Network Supervisor's Toolkit

home *** CD-ROM | disk | FTP | other *** search

/ Network Supervisor's Toolkit / Network Supervisor's Toolkit.iso / novell / fyi193 / 10feb93a.net

Wrap

Text File | 1996-07-10 | 271KB | 7,992 lines

FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: File Not Found Accessing CD-ROM through Mapping DOCUMENT ID#: FYI.P.8780 DATE: 10SEP92 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM "File not found" ISSUE/PROBLEM The user had an application that accesses data from a CD-ROM. It would report the following error when the user mapped to the CD-ROM as opposed to having it local. ┌──────────────┐ │File not found│ └──────────────┘ CD players are not like other DOS drives in all respects. For one thing, they have a device name. Normal DOS drives do not have a device name. The application seems to be searching for the device name as part of the path to the file. When the CD is remote, the local DOS will not have the device name in its internal tables. SOLUTION In this case, the problem was resolved by giving the network directory the same name as the CD-ROM device. Look at the command that is used to load MSCDEX.EXE to determine the device name. It will be the string after /D:. If the command (usually in the AUTOEXEC.BAT file) is "MSCDEX /D:MS001 /M:10 /L:F" then the device name is "MS001". The network directory was given the name MS001 with actual path F:\. The actual path is determined from the /L: parameter. See the NetWare Lite fax document 219 for more information on setting up CD-ROM drives. If that does not work, then to add the /L: switch. Even after trying these things some users still were not able to use their CD-ROM drive. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Upload of PUTIL3.ZIP DOCUMENT ID#: FYI.P.10969 DATE: 29DEC92 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM Concerning the file PUTIL3.ZIP available on available on NetWire or NetWare Express. PUTIL3.ZIP contains current printing utilities. The following files are contained in PUTIL3.ZIP: CAPTURE EXE 161103 10-28-92 9:17a NPRINT EXE 85425 10-28-92 8:26a PCONSOLE EXE 234069 10-28-92 8:21a PRINTCON EXE 161071 10-28-92 10:37a PRINTDEF EXE 183445 10-28-92 9:56a CAPTURE MSG 6309 10-28-92 9:17a ENDCAP EXE 13553 07-19-89 10:55a PSC EXE 22129 01-31-91 3:34p PSTAT EXE 28921 02-02-91 10:55a PCONSOLE HLP 40820 02-07-91 4:54p PRINTDEF HLP 34426 02-11-91 3:44p PRINTCON HLP 14223 01-10-91 9:59a PUTIL3 DOC ( This File ) CAPTURE.EXE NetWare v2.15c and above 01-28-91 49505 1) This EXE shipped with NetWare v2.2 and v3.11. 10-28-92 161103 1) Contains NCP Packet Signing. NPRINT.EXE NetWare v2.15c and above 02-05-91 85157 1) This EXE shipped with NetWare v2.2 and v3.11. 2) It fixes "file not found" when the first subdirectory in the path is 2 characters in length. 10-28-92 85425 1) Contains NCP Packet Signing. ENDCAP.EXE NetWare v2.15c and above 07-19-89 13553 1) This EXE shipped with NetWare v2.2 and v3.11. PCONSOLE.EXE NetWare v2.15c and above 02-11-91 233687 1) This EXE shipped with NetWare v2.2 and v3.11 2) Printers default to their common interrupt (instead of all to 7); the specified interrupt is shown (instead of displaying the default); you no longer have to assign the PSERVER to each queue; the "use interrupts" field is disabled when the printer is remote. 10-25-91 233873 1) Version 1.51 did not delete print server subdirectories when the print server was deleted. Version 1.52 now deletes these subdirectories properly. 10-28-92 234069 1) Contains NCP Packet Signing. PRINTCON.EXE NetWare v2.15c and above 01-26-91 160823 1) This EXE shipped with NetWare v2.2 and v3.11. 2) Defaults to byte stream instead of text. 10-28-92 161071 1) Contains NCP Packet Signing. PRINTDEF.EXE NetWare v2.15c and above 02-11-91 192283 1) This EXE shipped with NetWare v2.2 and v3.11. 10-28-92 183445 1) Contains NCP Packet Signing. PSC.EXE NetWare v2.15c and above 01-31-91 22129 1) This EXE shipped with NetWare v2.2 and v3.11. PSTAT.EXE NetWare v2.15c and above 02-02-91 28921 1) This EXE shipped with NetWare v2.2 and v3.11. PCONSOLE.HLP NetWare v2.15c and above 02-07-91 40820 1) This HLP shipped with NetWare v2.2 and v3.11. PRINTDEF.HLP NetWare v2.15c and above 02-11-91 34426 1) This HLP shipped with NetWare v2.2 and v3.11. PRINTCON.HLP NetWare v2.15c and above 01-10-91 14223 1) This HLP shipped with NetWare v2.2 and v3.11. CAPTURE.MSG NetWare v2.15c and above 10-28-92 6309 1) Used with CAPTURE.EXE v3.62 (NCP Packet Signing). SOLUTION NA FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Error 4 Opening $UPGRADE.BND" DOCUMENT ID#: FYI.P.10968 DATE: 29DEC92 PRODUCT: NetWare PRODUCT VERSION: v2.15, v3.11 SUPERSEDES: NA SYMPTOM While using the Upgrade backup method and trying to restore binderies, the system displayed the following error: ┌────────────────────────────┐ │Error 4 opening $upgrade.bnd│ └────────────────────────────┘ ISSUE/PROBLEM The user found that someone had deleted the CONFIG.SYS file. SOLUTION After the user created a CONFIG.SYS with files=20 and buffers=30, everything worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Cannot Print to HP IIIsi in Queue Server Mode DOCUMENT ID#: FYI.P.10954 DATE: 30DEC92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user was able to set up a HP IIIsi printer in QUEUE-SERVER mode initially and print fine; however, after rebooting the NetWare v3.11 file server, printing no longer functioned. The user replaced the Token-Ring network board in the PS/2 Model 95 server and the HP IIIsi printers on the network; however, during the final steps of setting up in PCONFIG (Hewlett Packard's printer configuration utility), the system displayed the following messages: ┌─────────────┐ │not logged in│ └─────────────┘ ┌────────────────────────────────┐ │cannot find queue server account│ └────────────────────────────────┘ ISSUE/PROBLEM Configuration: ■The NetWare v3.11 server is part of a large multi-server internet. When setting up the new server, the user used an internal IPX number for the new NetWare v3.11 server that was already being used somewhere on the internet as a network address. SOLUTION After the user changed the address to something unique, the symptom disappeared. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Invalid Sound Blaster Environment Settings DOCUMENT ID#: FYI.P.10947 DATE: 31DEC92 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM When booting the system, it returned the following error message (or something similar): ┌─────────────────────────────────────────────────────┐ │Error # -- invalid Sound Blaster environment settings│ └─────────────────────────────────────────────────────┘ The user Sound Blaster card would only work with some applications. ISSUE/PROBLEM MEMMAX -U must run in the AUTOEXEC before the two set statements for the sound blaster. SOLUTION After the user put the required MEMMAX -U lines in the AUTOEXEC.BAT, the sound blaster card is working fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Selected Hardware Not Available ..." - IBM Valuepoint DOCUMENT ID#: FYI.P.10946 DATE: 02JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user was trying to use ISADISK from ISA311.ZIP and could not get it to work. The system displayed the following error: ┌───────────────────────────────────────────────┐ │Selected Hardware not available for ISA drivers│ └───────────────────────────────────────────────┘ ISSUE/PROBLEM The user had purchased the new IBM PS Valuepoint 6384-C40. There is no BIOS on the disk controller, so there is no information available for ISADISK. Loading ISADISK on a SCSI controller would cause the same error. SOLUTION The user downloaded IDE386.ZIP (available on NetWire or NetWare Express) to solve the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Error 004: An Attempt to Add a Trustee..." - MIGRATE.EXE DOCUMENT ID#: FYI.P.10984 DATE: 04JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When running Migrate v1.08, same server method, the following error would occurred: ┌────────────────────────────────────────────────────── ────────┐ │Error 004: An attempt to add a trustee to a directory failed.│ │Object:GUEST(user) Directory path: Servername\sys:\MAIL\xxxxx│ └────────────────────────────────────────────────────── ────────┘ The xxxx indicates the user object id from the NetWare v2.2 server. ISSUE/PROBLEM This error is caused by not doing a data restore to the NetWare v3.11 server before running the second half of the migrate. SOLUTION According to Chapter 3 of Same Server Migrate manual, step 25, restore the data files from your backup device to the NetWare v3.11 server. Step 26, start the migration utility again. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: NMI Parity Errors Generated by I/O Check DOCUMENT ID#: FYI.P.10983 DATE: 04JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The system was getting NMI parity errors generated by I/O check. This had been working fine. The system displayed the errors when copying files from one volume to another. The errors seemed to occur only when copying to one particular volume. The user tried copying files to the workstation and then the other volume but the same problems occurred. They had some other errors and abends before the NMI errors. ISSUE/PROBLEM Configuration: ■File server - Everex 386/33 with 16 MG of RAM ■The server originally had 8MG of RAM. The user had added an additional 8MG. After removing the 8MG, the user could not reproduce the problem. This machine uses an on-board cache that takes 64KB off of the tail end of the first 384KB of extended RAM. The user tried disabling the cache; however, that did not solve the problem because disabling the cache does not release the RAM that the cache was using. This means that the memory board starting address was leaving a 64KB hole between the board and the remaining 320KB. SOLUTION After the user addressed the memory board to make up for this hole in memory, everything has been working fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Unexpected Error 16 (8801)" - Requester v2.0 DOCUMENT ID#: FYI.P.10982 DATE: 04JAN93 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM If an IBM XGA-2 display adapter (set to interrupt line 3, ROM Address CC000) is added to the workstation, the following error was produced when logging in under an OS/2 session: ┌──────────────────────────┐ │Unexpected error 16 (8801)│ └──────────────────────────┘ LOGIN was successful from a private VDM (DOS box) with NETX loaded or when a different video card was used. ISSUE/PROBLEM Configuration: ■Using NetWare Requester for OS/2 v2.0 ■Running OS/2 v2.0 on a Model 70 with an IBM Token-Ring network board set to interrupt 2. The Shared RAM size was set to 8KB on the Token-Ring card. SOLUTION After changing the Shared RAM size to 16KB (the recommended setting), the user was able to log in under OS/2 with the IBM XGA-2 adapter installed. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Load File Defines a Public Variable..." - NetWare v3.11 DOCUMENT ID#: FYI.P.10979 DATE: 04JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM "Load file defines a public variable that has already been defined symbol already defined: ASPI_ENTRY" ISSUE/PROBLEM The user was loading U24_31X.DSK for a UltraStore 24f and AESXNV31.DSK along with AXSDNV31.DSK for a DTC 3290 controller. No matter which driver he loaded first, loading the second would give the following error: ┌────────────────────────────────────────────────────── ───┐ │Load file defines a public variable that has already been│ │defined symbol already defined: ASPI_ENTRY" │ └────────────────────────────────────────────────────── ───┘ SOLUTION The user decided to use two of the same controllers, which made the error disappear. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Error Messages if DR DOS System Files Are Missing DOCUMENT ID#: FYI.P.10973 DATE: 04JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM A question was raised as to what error messages would be seen under DR DOS if various system files were missing. SOLUTION There are three system files for DR DOS 6.0. These are IBMBIO.COM (hidden), IBMDOS.COM (hidden) and COMMAND.COM (not hidden). When IBMBIO.COM is removed, the error message generated is the following: ┌───────────────────────────────────────┐ │Cannot load file - press a key to retry│ └───────────────────────────────────────┘ When IBMBIO.COM is restored and IBMDOS.COM is removed the following error message is displayed: ┌───────────────────┐ │Can't load DOS file│ └───────────────────┘ When IBMDOS.COM is restored and COMMAND.COM is removed, the following error message is generated ┌──────────────────────────────────┐ │Bad or missing command interpreter│ └──────────────────────────────────┘ FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Unable to Unlock File Server Console - MONITOR.NLM DOCUMENT ID#: FYI.P.10329 DATE: 29OCT92 PRODUCT: NetWare PRODUCT VERSION: v3.x SUPERSEDES: NA SYMPTOM Someone with supervisor equivalence had locked the file server console (from within MONITOR) and forgot the password they had set. The supervisor's password should unlock the console but did not. ISSUE/PROBLEM Under normal conditions, the password assigned in MONITOR.NLM or the supervisor (user) password will unlock the console. The bindery was probably corrupt, or the user mistyped the supervisor password. SOLUTION In this case, the only way to unlock the console was for a supervisor equivalent to change the supervisor's password and use the newly assigned supervisor's password to unlock the console. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: SBACKUP Documentation Error DOCUMENT ID#: FYI.P.10281 DATE: 17AUG92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user was having problems getting the following files from SBACK.ZIP (SBACKUP.NLM dated 3-5-92) to work with the HP35480A DAT backup system: TAPEDAI.DSK AHA1540.DSK ASPITRAN.NLM DIBIDAI.NLM Whenever the DIBIDAI.NLM option was selected from the DIBI2$DV.DAT file, the following error messages would appear: ┌───────────────────────────────────────────────────┐ │No device found, DIBI device drives did not respond│ └───────────────────────────────────────────────────┘ ISSUE/PROBLEM The user tried the following to solve the problem: ■Recopying the sbackup files into the system subdirectory, and making sure that a DIBI and TSA directory was present. ■Recreated the DIBI2$DV.DAT file with just the DIBIDAI.NLM option. ■Removed the .NLM extension in the DIBI2$DV.DAT file. Nothing worked even though the documentation says it is supported this way. The symptom was caused because the HP35480A is not hard coded into the TAPEDAI.DSK file (31576 bytes dated 5-11-92). Therefore, the documentation is wrong, and this device is not supported with these drivers. SOLUTION HP35480A will be available in the next release of the TAPEDAI.DSK. A workaround is to use the following drivers: AHA1540.DSK ASPITRAN.NLM NOVADIBI.NLM FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Invalid Drive Specification Error DOCUMENT ID#: FYI.P.9989 DATE: 26JUN91 PRODUCT: NetWare 286 PRODUCT VERSION: v2.15 SUPERSEDES: NA SYMPTOM Cannot map a drive to the server The server came up fine, and COMCHECK checked out fine. However, after loading IPX and NET3 and trying to change the F: drive, the system displayed an invalid drive specification error. There was no drive mapping to the server. The user tried to map a drive using the MAP command on the floppy. The MAP command executed fine; however, when doing a DIR on the directory that had been mapped to, it showed 0 files. ISSUE/PROBLEM Configuration: ■Running NetWare on a COMPAQ 386N ■Using a 650MB SCSI drive (configured as channel 1) ■Using the 3C503TP card The symptom was caused because the user was booting the server from the hard disk that was set at channel 1. SOLUTION When the user booted the server from floppy, everything worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Abend: GPI Running Process MUXPRC" - Null Modem DOCUMENT ID#: FYI.P.11025 DATE: 04NOV92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When doing a direct LOGIN over a null modem cable into a file server, the router on the other side of the null modem cable received the following error: ┌──────────────────────────────────┐ │ABEND: GPI Running Process MUXPRC│ └──────────────────────────────────┘ ISSUE/PROBLEM A user set up the following configuration with two file servers (SRV), two workstations (WS) and two routers (R), connected through a null modem cable with WNIM+ boards: ┌────────┐ ┌────────┐ ┌────┤ SVR1 │ │ SVR2 ├────┐ │ └────────┘ └────────┘ │ A│ │C A│ ┌────────┐ ┌────────┐ │C A├────┤ WS1 │ │ WS2 ├────┤C │ └────────┘ └────────┘ │ │ │ │ ┌────────┐ 0Modem ┌────────┐ │ └────┤ R1 ├────o─────────── o────┤ R2 ├────┘ NE2000 └────────┘WNIM+ BBB WNIM+└────────┘ NE2000 A direct login from WS1 to SRV2 will produce the ABEND on R2. A direct login from WS2 to SRV1 will produce the ABEND on R1. Everything is fine, when a log in happens on the same segment, which means from WS1 to SVR1, and then doing an ATTACH or MAP to SVR2. The same vice versa. Everything is also fine, when the following parameter is set: "Reply to get Nearest Server = OFF" For a login from WS1 to SRV2 set the parameter in SRV1. For a login from WS2 to SRV1 set the parameter in SRV2. The routers are configured as follows: R1- Router Type: Dedicated Protected Mode Communications buffers: 150 HDW config: NE2000 (v3.06ec), IRQ=3, Port=300h, Netaddress=AAA WNIM+, NW Async Remote Router WNIM v1.2 Port=280h, Ram Buffer at D000:0, Netaddress=BBB R2- Router Type: Dedicated Protected Mode Communications buffers: 150 HDW config: NE2000 (v3.06ec), IRQ=3, Port=300h, Netaddress=CCC WNIM+, NW Async Remote Router WNIM v1.2 Port=280h, Ram Buffer at D000:0, Netaddress=BBB SW config: Line number = 1 Line enabled = YES Flow control enabled = YES Data Rate = 19200 Modem Type = Special Configuration Modem Initialization String Dial Prefix = ATV0DT Phone number Auto Connect = YES Auto ID String = ID-CCC Life Class = Continuous The file servers were pure NetWare v3.11 server. The workstation were ordinary workstations that had the latest shells. The user tried several suggestions, such as the following: ■ Using the PTF299 (which did not even established a direct link between the routers, and also seems to be related to a problem that the patch has with ACK packets) ■ Not running IPX or NETx on the routers ■ Using CASTOFF and checking that all addresses are different on the servers and routers tested Then the user tried the following and had the same results: ■ Used all possible BAUD rates (2400) for the null modem ■ Checked if SLIST or DISPLAY SERVERS showed all information and did not find any hint (SLIST and DISPLAY SERVERS worked fine when not logged in through the router). SOLUTION Finally, the user found out that the routers should not run in protected mode. They have to run in real mode. The other important thing was that it is necessary to have two different addresses for the null modem line. For example, R1 should have BBB while R2 should have DDD for the WNIMs. Also the setup should be the following: Auto Connect = YES on the second started Router Auto Connect = NO on the first started Router Flow Control = NONE Baud rate = 2400 Line Enabled = YES Modem Type = NONE Another point to look for is intermittent problems with NE2000 and WNIM+ boards. So, for a test, change the NE2000 to a NE1000. Instead of using PTF299, use the patches in the ARO.ZIP file on available on NetWire or NetWare Express. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Problems with DOS Set Command in Login Script DOCUMENT ID#: FYI.P.11017 DATE: 21OCT92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM While following some suggestions for login scripts in a manual for setting a DOS environment variable, the user had some trouble executing the application. The manual suggested a line in the login script as follows: DOS SET NETNAVC="F:\APPS\NETNAV\CODE" The user copied this string using lowercase letters instead of uppercase. The application requiring this variable being set was not able to interpret the command. The user checked the SET settings under DOS and found the following command: NETNAVC=f:\appsetnav\code ISSUE/PROBLEM A note in the NetWare Version 3.11 Installation manual states that the backslash in the DOS SET command is used as a special programming character and that instead of having one backslash (such as, in a path), two backslashes should be used. The only combination of a backslash and a character following with a special interpretation was "\n." Anything else (especially "\N") works; two backslashes work in all cases. SOLUTION Do not use "\n" parameter in that command. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: SBACKUP Not Finding a Wangtek Tape Drive DOCUMENT ID#: FYI.P.11016 DATE: 19OCT92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When SBACKUP was loaded, the user was presented with the device select screen that displayed the following message: ┌────────────────────────────────────────────────────── ──────┐ │The ASPI control layer did not detect any sequential devices│ └────────────────────────────────────────────────────── ──────┘ ISSUE/PROBLEM Configuration: ■Server : No name 486/33, 386sx 16 ■Disk controller: AHA1542B ■Tape : Wangtek 5150 ES ■Drivers : Using newest version The symptom still occurred after the user check things such as cables and SCSI IDs. The error is caused because the driver or SBACKUP does not properly recognize the tape if no tape is inserted. SOLUTION After the user inserted a tape, SBACKUP worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: SFT III Procedure Europe DOCUMENT ID#: FYI.P.11015 DATE: 16OCT92 PRODUCT: NetWare SFT Level III PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM Q1: Is SFT III a public available product? A1: No, SFT III is currently in a Controlled release. Only certain dealers may sell SFT III. These dealers have been trained for servicing this product. Q2: What is a Controlled release? A2: During the BETA phase, many companies had to chance to look at this product under a nondisclosure agreement. The response from this market was strong enough that Novell decided to offer those companies early access to this product by a Controlled release. This procedure will ensure that the final product will meet the market requirements. Users who wish to participate in this program must meet certain site requirements. These requirements are checked and verified by Novell. During this verification, the ordering and shipment procedure occur. The user will receive an upgrade to the General release free of charge, and support will be done by Novell through the reseller. Q3: What are the limitations of SFT III? A3: SFT III currently supports only DOS, Windows, and OS/2 clients. NetWare for SAA, TCP/IP, NFS, and NetWare for Macintosh are currently not supported. Also, IPX checksums for Ethernet_802.3 are not supported. Q4: What if users need one of the products mentioned above? A4: You may install these products on another server on the same network. Q5: Will these products ever be supported? A5: TCP/IP, NFS, and NetWare for Macintosh will be supported in a general release of SFT III. NetWare for SAA will probably not be supported soon because of the complexity of this product. Q6: Will the general release be based on NetWare v4.0? A6: Although no final decision has been made at the time of this wirting, the General release will probably be based on NetWare v3.x. SOLUTION NA FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Returning Abended Server Back to DOS DOCUMENT ID#: FYI.P.11014 DATE: 16OCT92 PRODUCT: NetWare PRODUCT VERSION: v3.x SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM When NetWare 386 servers receive abends, in most cases it is not possible to return to the DOS prompt, quickly. Especially, in cases of the new PS/2 90/95 machine, the self-check after restart is quite time consuming. If the server does not have a reset button, it has to be turned off and back on again. SOLUTION A solution to this scenario is to enter the debugger by typing 386DEBUG, then type the command Q for quit and the Y for yes. Unless DOS has been unloaded, the debugger will return to DOS. If DOS is no longer present in that machine, the debugger will cause the machine to reboot thus saving a power OFF/ON sequence. When switching back to DOS, the server will reenter the real mode world; thus, all drivers are left somewhere in the flat memory area. Most drivers will reinitialize the hardware during load time. Do not try to resume the server with the debugger after receiving a serious abend. Although, sometimes this looks like working internal structures are damaged (after a Page Fault or a GPPE), the server software will continue but is not in a defined state. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Truncated or Segmented Print Files with DOS Print DOCUMENT ID#: FYI.P.11011 DATE: 08OCT92 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM The user has problems printing files with DOS commands (PRINT or COPY/TYPE with redirection). Files in the queue are truncated or split into several queue files. ISSUE/PROBLEM The user used CAPTURE and the DOS PRINT command under Windows to send files to a network printer. In many cases, the queue contained only parts of the print file; in other cases, the whole file was printed while in different queue jobs. The shell version NETX v3.26 and v3.27 would not fix the problem. SOLUTION The user had a combination of several problems: 1. The DOS PRINT does not work very efficiently. It reads the file and sends it to the print queue in 64 bytes pieces in resident mode (as a background TSR). The time interval between these mini blocks may be quite large, especially in a Windows DOS box. Therefore, the printout of 10KB may even take 3 minutes or more. "Copy file LPT1" or "Type file>LPT1" will be much faster. NPRINT will be fastest with large files (larger than 150KB). Timeout--If PRINT is used, the CAPTURE timeout has to be set to a relatively high value (or to 0 = no timeout). If the printing DOS window is made a background task in MS Windows, PRINT may even completely pause printing until the window is active again. If your timeout is too small, the result is fragmented jobs. AutoEndcap (AE)--Autoendcap means: an EndCap signal (EndOfJob) is sent by the shell as soon as the application returns to DOS. If you have your CAPTURE flags set to AE (this is the default), your shell will sent an EndCap to the server whenever any application on your PC returns to DOS. So every command entered, even typing DIR at the prompt will send the EndCap signal. If you still have a background PRINT command sending data to a queue, this EndCap signal will cause the server to close the current queue job and create a new one for the next data sent by PRINT. The result is fragmented jobs. Therefore, when using DOS PRINT, use the following capture statement: Capture TI=0 NoAuto Approximately, the same procedure applies to "Copy file LPT1" or "Type file>LPT1" when you switch to other Windows tasks while printing large jobs. 2. DOS PRINT, COPY, and TYPE should only be used for pure text files. If binary data are in your file, print jobs may be truncated. The job will go into the queue until the first 0x1A byte (EndOfText) is found in the source file. So, do not use PRINT or TYPE with preformatted print data (such as print files from spread sheets or word processors) that may contain control codes. "COPY /B file LPT1" may be used; however, the best solution is to use NPRINT or PCONSOLE. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Abend: Invalid Cache Buffer Handle Used" DOCUMENT ID#: FYI.P.11010 DATE: 08OCT92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user saw the following message on the file server: ┌────────────────────────────────────────┐ │Abend: Invalid cache buffer handle used.│ │Interrupted process: Polling Process. │ └────────────────────────────────────────┘ ISSUE/PROBLEM This message can be found in SERVER.EXE. The user had five servers with identical hardware, however, only one of them received the abends. SOLUTION This abend can often be avoided by turning off the internal cache for the machine. Also, the user needs to run a thorough diagnostics on the memory being used to ensure all chips are running at the same speed. The user lost confidence in his PC and replaced the complete system board. This fixed the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: DIR Shows Garbage after SSTOR Compression DOCUMENT ID#: FYI.P.11005 DATE: 05JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM DIR of floppy drive shows garbage ISSUE/PROBLEM After running the SSTOR compression on the hard disk, when the user does a directory listing of the floppy drive, all he sees is garbage and he cannot view or execute any of the files on that floppy. A directory listing on the C: drive worked just fine. CHKDSK showed no errors. SOLUTION The user disabled the SuperPCK caching of the floppy drive and added the /-a* switch to the SuperPCK line in AUTOEXEC.BAT. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: History Buffer Not Available in DESQview DOCUMENT ID#: FYI.P.11001 DATE: 05JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM History not working inside of DESQview ISSUE/PROBLEM When running inside a DOS session from Quarterdeck's DESQview, the history buffer is not available, and pressing the up arrow key to recall the last line entered does not do anything. SOLUTION This is a known issue inside DESQview, Windows, or probably any other program that uses DOS sessions. When a DOS session is opened, a new copy of COMMAND.COM is loaded and the DOS "master" environment is passed to that DOS session. However, the controlling application (such as DESQview or Windows) has to trap for its own "hot" keys. Because of this, the history buffer, which lets you call up previously entered commands by pressing the up arrow or <Ctrl>+<E> keys, is not handled by the application in a DOS session the same way; it is just at a normal DOS prompt. This is not a bug on either side, just an inherent issue with this type of application. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Copying PRINTCON.DAT from Supervisors Mail Directory DOCUMENT ID#: FYI.P.10991 DATE: 05JAN93 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM When logged in as Supervisor, the administrator created a new Print Job Configuration in PRINTCON and then tried to copy the file to another user within PRINTCON (an option only available for Supervisor). The user could not see the configuration file in the MAIL directory. ISSUE/PROBLEM The PRINTCON.DAT file is not written or modified until exiting PRINTCON. In this case, there were no configurations to copy because the user had just created them and had not saved the PRINTCON.DAT. SOLUTION After the user exited PRINTCON then ran and copied it again, the copy feature worked. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Insufficient Memory Available..." - EISA Configuration DOCUMENT ID#: FYI.P.11054 DATE: 07JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When the user tried to mount a volume, the system displayed the following message at the console and the volume would not mount: ┌─────────────────────────────────────────────┐ │Insufficient memory available to mount volume│ └─────────────────────────────────────────────┘ The user had 64MB RAM and 2GB disk space. MEMORY at the console showed 63.xxMB available. The user had to use the REGISTER MEMORY command to register the memory even though it was an EISA machine. ISSUE/PROBLEM The user had not configured the memory in the EISA CONFIGURATION. SOLUTION After configuring the memory in the EISA CONFIGURATION, the user no longer had to register memory manually and the volumes mounted fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: GPPE with CPQSCSIT.NLM DOCUMENT ID#: FYI.P.11048 DATE: 07JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When choosing the append option in SBACKUP, the server hung or received a GPPE without displaying a message. ISSUE/PROBLEM Configuration: ■Using the latest SBACKUP and the CPQSCSIT.NLM dated 02-21-92 Compaq said that this driver will not support the APPEND option. They have an older driver, with other problems, that does support append but returns and error on new tapes. SOLUTION You cannot use the append option with the CPQSCSIT.NLM driver. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: DPATCH.NLM Not Reporting Disk I/O Statistics DOCUMENT ID#: FYI.P.11055 DATE: 07JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user was running STAT.NLM that collects LAN I/O statistics on a server. The user recently implemented DPATCH.NLM along with its respective PATCHMAN (this patch allows STAT.NLM to collect disk I/O statistics). After installing the patch, the disk statistics still reported 0s. The user also tried using the PATCHMAN from 311PTB.ZIP. Neither action seemed to work. ISSUE/PROBLEM The user had loaded most all the patches from 311PTB.ZIP. SOLUTION Novell recommends that users add only patches that affect the problems experienced. After the user unloaded the patches from 311PTB.ZIP, the disk statistics began being collected. Novell recommended that the user add only one patch at a time back to the operating system to pursue finding which NLMs conflict. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Not Reading Device LPT1..." - NetWare v3.11 DOCUMENT ID#: FYI.P.11045 DATE: 07JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM "Not reading device LPT1...fail on INT 24 LPT1" ISSUE/PROBLEM After loading NETX (not logging in) and when trying to print locally through DOS (no CAPTURE involved) on a workstation, the system displayed the following error message: ┌─────────────────────────────────────────────┐ │Not reading device LPT1...fail on INT 24 LPT1│ └─────────────────────────────────────────────┘ Also after logging in, WordPerfect could be set to print directly to the LPT port. It printed with no problem. After the user removed the AUTOEXEC.BAT and CONFIG.SYS, the system still displayed the error. The user was using NETX.COM v3.26 dated 02-11-92 and was 50KB while the NETX.COM in DOSUP5.ZIP was dated 02-11-92 and reported being 30KB. Using the NETX.COM from DOSUP5.ZIP, did not eliminate the symptom. SOLUTION Put "Local printers=0" in the SHELL.CFG. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Profile Login Script Not Executing - NNS DOCUMENT ID#: FYI.P.11043 DATE: 07JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.10, v3.11 SUPERSEDES: NA SYMPTOM The users logging in through an access server using ONLAN would have their Domain and User login scripts executed but not their Profile login script. The log in problem happened all of a sudden with no administrative changes on the network. ISSUE/PROBLEM Configuration: ■Using NetWare Name Service (NNS) ■Three servers in the domain The remote users were executing LOGIN.EXE v3.10 and not the NNS version. The LOGIN must have been interpreting the Domain script as the System and the User script as the User. SOLUTION After the remote users executed the correct LOGIN, everything worked as expected. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Out of Dynamic Memory Workspace During GetDirectoryEntry" DOCUMENT ID#: FYI.P.11037 DATE: 06JAN93 PRODUCT: NetWare PRODUCT VERSION: v2.2 SUPERSEDES: NA SYMPTOM Intermittently, all workstations attached to a specific server simultaneously received the following error: ┌────────────────────────────────────────────────────── ──┐ │Out of dynamic memory workspace during GetDirectoryEntry│ └────────────────────────────────────────────────────── ──┘ ISSUE/PROBLEM After the user replaced the server and regenerated the operating system, the error still occurred. When the error occurred and the user tried to shut down the server, it hung. If the user did not immediately reboot the server, other servers on the network started to get the same symptoms. The user ran a machine with FCONSOLE loaded and when the server crashed, no parameters were even close to the maximum limits. The error also occurred with many or just a few users attached. SOLUTION The user found, through talking to some other instructors at the school where this was happening, that one workstation was constantly having communication problems. After the user replaced that board, the error has not returned. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Unable to Identify Hard Disk" - DR DOS, Norton DOCUMENT ID#: FYI.P.11030 DATE: 06JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM The user was trying to run Norton's hard-disk speed benchmark test after installing the DR DOS, April update, on the machine. When trying to run it, the system displayed the following message: ┌────────────────────────────┐ │Unable to identify hard disk│ └────────────────────────────┘ ISSUE/PROBLEM When loading Windows after the DR DOS update, the user let Windows load SmartDrive in the AUTOEXEC.BAT. SOLUTION After the user removed SmartDrive from the AUTOEXEC.BAT, everything worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Error Insufficient Memory for Directory Tables" DOCUMENT ID#: FYI.P.11028 DATE: 06JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user tried to run VREPAIR, it would not complete and it displayed the following message: ┌──────────────────────────────────────────────┐ │Error Insufficient Memory for Directory Tables│ └──────────────────────────────────────────────┘ The user could not mount the volume due to FAT corruption caused by a power outage and wanted to start over clean. The user deleted the volume, deleted the NetWare partition, recreated the partition, then tried to recreate the volume. It would give the above error and would not create the volume. All memory pools in MONITOR looked normal, and the server had sufficient memory. The user was trying to recreate everything as it was previously configured. ISSUE/PROBLEM The above error message is returned when the operating system makes a call to allocate memory for a NLM or process and sufficient memory cannot be found. In this case, memory corruption was caused by the power problems and manifested itself by showing less memory than what was actually available when the operating system tried to load the VREPAIR and INSTALL utilities. SOLUTION The user did the following to solve the problem: 1. Deleted the partition again. 2. Shut down the server. 3. Booted to DOS. 4. Ran SERVER.EXE. 5. Loaded install. 6. Recreated the partition and volume with no errors. Then the volume mounted and everything ran fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Running NetWare on a Machine with No Keyboard DOCUMENT ID#: FYI.P.11076 DATE: 11JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM The user had a machine that had a faulty keyboard handler and wanted to know what would happen if he wanted to run the machine without a keyboard for a few weeks. Does NetWare require a keyboard to be present? SOLUTION NetWare does not require a keyboard. However, when you reconnect it, the keyboard may not work. Also, you must shut down the server through RCONSOLE, so ensure all the necessary NLMs are loaded before you disconnect the keyboard. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "LC Unable to Load Commexec..." - NetWare for SAA DOCUMENT ID#: FYI.P.11100 DATE: 12JAN93 PRODUCT: NetWare for SAA PRODUCT VERSION: v1.3 SUPERSEDES: NA SYMPTOM The user would get the following error when loading COMMEXEC.NLM: ┌────────────────────────────────────────────────────── ───────────┐ │LC unable to load commexec module agent type 11AH error code 90BH│ └────────────────────────────────────────────────────── ───────────┘ ISSUE/PROBLEM A power failure had corrupted the SAA files and the user had recopied them from the original disks. Trying to load it again would produce a message that it was already loaded. Trying to unload it would produce a message that it was not loaded. SOLUTION The user must do a complete reinstall of NetWare for SAA instead of just copying all the files. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: RPRINTER Is Slow on Austin 386 Machine DOCUMENT ID#: FYI.P.11096 DATE: 12JAN93 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM When they send any job to the printers using RPRINTER, it would print one line each 30 seconds. ISSUE/PROBLEM Configuration: ■Two HP LaserJet IID printers running RPRINTER on Austin 386-sx20 machines ■LAN board set for interrupt line 5 ■Printers attached to LPT1. ■The LPT port is on the system board The machine's LPT1 port was set for interrupt line 5 that conflicted with the LAN board interrupt line setting. SOLUTION After the user changed the interrupt line assignment in the CMOS, the printer printed fine. Keywork: IRQ, motherboard, card FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Abend: Bounds Check Processor Exception" - NetWare v3.11 DOCUMENT ID#: FYI.P.11093 DATE: 12JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user was running NetWare Runtime v3.11 on a Model 90 with 60MB SCSI and did the following: ■ Low-level formatted the drive with the Reference Disk. ■ Created a 5MB DOS partition with DOS v5.0. ■ Ran SERVER.EXE. ■ Loaded INSTALL.NLM. ■ Then when the user selected YES to create a NetWare partition on the rest of the drive in the INSTALL utility, the system displayed the following error: ┌───────────────────────────────────────┐ │ABEND: bounds check processor exception│ └───────────────────────────────────────┘ ISSUE/PROBLEM The user was using a corrupted working copy of the SYSTEM-2 diskette. SOLUTION After the user created a new copy of the SYSTEM-2 diskette, the install worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "General Protection Error in WPWIN.EXE at 0005:4487" DOCUMENT ID#: FYI.P.11090 DATE: 12JAN93 PRODUCT: DRDOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM After clicking on the icon, Windows returns the error message "General protection error in WPWIN.EXE at 0005:4487." ISSUE/PROBLEM After installing DR DOS and resolving all other Windows related issues, WPWIN would not run. After clicking on the icon, Windows returns the error message "General protection error in WPWIN.EXE at 0005:4487." ┌──────────────────────────────────────────────────┐ │General protection error in WPWIN.EXE at 0005:4487│ └──────────────────────────────────────────────────┘ SOLUTION WordPerfect suggested regenerating the WP{WP}WP.SET file that contains all the program settings. This is done by renaming the WP{WP}WP.SET to WP{WP}WP.OLD. The next time the user starts WPWIN, the program will regenerate a new WP{WP}WP.SET file for the current environment. Implementing this suggestion solved the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "0XFFFFFFBE" Entering Wrong Password for SBACKUP DOCUMENT ID#: FYI.P.11084 DATE: 12JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user was running SBACKUP.NLM and when entering the supervisor password, the system displayed the following error message then prompt the user for another log in attempt. ┌──────────┐ │0xFFFFFFBE│ └──────────┘ ISSUE/PROBLEM The user thought he was logging in correctly; however, when he logged into another workstation as supervisor, he remembered that he had changed the password. SOLUTION After the user entered the correct password, the system allowed him to log in and continue normally. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Drive Deactivation Troubleshooting Tips DOCUMENT ID#: FYI.P.11081 DATE: 12JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM Drive Deactivation Tips and Techniques for NetWare v3.11. The purpose of this document is to establish what drive deactivation is and why it occurs. Drive deactivation is when one or more devices (drives) in the server become unable to communicate with the operating system (OS). Drive deactivation is caused by an inability of NetWare and the disk driver to communicate with the device. When the driver cannot communicate with the drive, it will normally retry several times. If, after a predetermined number of retries, the driver still cannot establish a communication link with the drive, the driver tells the OS that the drive is not responding and the OS then deactivates the device. Drive deactivation can occur across any of the disk interfaces, such as IDE, ESDI, or SCSI. Drive deactivation is also not confined to just one manufacturer. The following list is a "checklist" that Novell suggests you go through in troubleshooting a drive deactivation problem. In many cases, you must be very specific with hardware, firmware, and software revision levels as you work with your third-party Host Bus Adapters (HBAs) and drive manufacturers in solving your problems. The priority of the checklist is based on solutions that will cost the least and still receive the highest priority. For checklist purposes, this discussion will be restricted to the SCSI interface. All suggestions should be easily translated for any interface. CHECKLIST 1. Verify that termination along the SCSI bus is correct. The first basic rule to remember is that both ends of the bus must be terminated. You must be careful that if you have drives in an internal or external cabinet that you remove the termination from the individual drives and then just terminate the cabinet externally. This avoids future headaches; because, as you add more drives, you do not have to worry about which drive was terminated in which cabinet. Another problem arises when you use both the internal and external ports on an HBA. When you do this, you must remove the termination from the HBA and terminate both ends of the SCSI bus, which will usually be at the drives themselves unless an external cabinet is used. 2. Verify that the SCSI IDs are set properly. In all situations, the HBA will have a SCSI ID of 7. This may be set either through a hardware setting (jumpers) or through software utilities (such as EISA CONFIGURATION or REFERENCE). In PS/2 situations and possibly with HPs, the HBA takes a SCSI ID of 7 and the attached devices start at 6 and work down to 0. In all other situations, the HBA takes an ID of 7, and the attached devices start at 0 and work up to 6. In the PS/2 environment, there are specific things that need to be addressed. ■ Under reference, the "fairness" needs to be turned ON for the HBA. ■ Using the OPTION diskette v1.0 from IBM, the settings for Fairness are reversed. Normally, fairness is turned ON. However, if you are using the OPTIONS diskette v1.0, you must set "Fairness" to OFF. On all other versions of the OPTIONS diskette, simply set fairness to the normal setting of ON. 3. Verify the disk driver. Verify with your third-party HBA manufacturer that you are running the latest Novell certified driver and that it is compatible with the firmware revision level on your HBA. 4. Verify the HBA firmware. Verify with your third-party HBA manufacturer that the version of firmware on your HBA is the correct version applicable to the version of the disk driver that you are using. 5. Verify the firmware revision level on the drive. Verify with the third-party drive manufacturer that you have the proper or latest version of firmware on your drive. 6. Check and if necessary replace the cabling on the SCSI bus. Reseat all connectors, then reseat the HBA into the bus. Look for pinched or broken leads on all connectors and cables. 7. Verify proper wattage of the power supply. Novell had seen many drive deactivation problems solved by upgrading or replacing a weak or flaky power supply. Drive deactivation problems seem to be an ever increasing problem as drives grow larger in capacity yet shrink in physical size. This smaller size allows more drives to be placed into the server box itself thus placing a larger demand on the server's power supply. 8. Examine other devices attached to the SCSI bus. If you have a tape unit, CD-ROM, or other devices attached to the SCSI bus, ensure you enable the "disconnect" feature on those devices. This allows the HBA to issue a command or series of commands to that device and then disconnect and return to servicing disk requests. If the "disconnect" feature is not enabled, the SCSI bus must wait for the device to return with a completion code and by so doing may not be able to service the drivers requests to "talk" to the drive. 9. Examine other bus mastering boards. If your server has bus mastering boards, you may need to examine these closely. If a bus mastering board holds the bus too long, this may prevent the disk driver from establishing a communication link with the drive. Sometimes it may be an issue of getting a newer driver for the bus mastering device or taking drastic steps of replacing or upgrading the suspected bus mastering device. Experience has shown that this can happen, but it is rare. 10. Replace the HBA. If the above steps have failed to solve the drive deactivation problem, then it is time to take drastic action. Replace the HBA with preferably the same make and model number. If a different brand or model number is chosen, a reformatting and repartitioning of the drive may be necessary. 11. If you are still unsuccessful in solving the drive deactivation, it is time to seriously look at replacing the drive. Experience has shown that replacing the drive with the same make and model number may solve the problem; however, in some situations a switch to a different manufacturer may be necessary. Drive deactivation can sometimes be very finicky and may require drastic steps to stabilize the server and eliminate the drive deactivations. SOLUTION NA FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: External Router Shipped with NetWare v3.11 Is Not SNMP Compliant DOCUMENT ID#: FYI.P.11080 DATE: 12JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM Is the external router software that comes with NetWare v3.11 SNMP compliant? SOLUTION No, because it will not pass TCP/IP packets. Users wishing to run SNMP will have to get the multi-protocol router software. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Memory Allocation Errors Running MAKEUSER DOCUMENT ID#: FYI.P.11069 DATE: 11JAN93 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM When running MAKEUSER, the workstation received memory allocation errors. ISSUE/PROBLEM The administrator wanted to add 8000 users (university students and professors) with MAKEUSER. SOLUTION MAKEUSER allocates a structure for every user. Therefore, when the user tried to create 8000 users all at one time, the system required more memory than the workstation had available. Instead, the user could run MAKEUSER multiple times for fewer users. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "File Pre-Read Error..." - NetWare v3.11 DOCUMENT ID#: FYI.P.5671 DATE: 26FEB92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When mounting the SYS: volume, the system immediately fills the screen with the following error: ┌────────────────────────────────────────────────────── ───┐ │File Write Pre-Read error <server>/SYS:SYSTEM\SYS$LOG.ERR│ └────────────────────────────────────────────────────── ───┘ ISSUE/PROBLEM The system error log (SYS$LOG.ERR) was corrupted. The SYS$LOG.ERR file could not be deleted because NetWare was keeping it open. The console kept displaying the following message repeatedly: ┌────────────────────────────────────────────────────── ───┐ │File Write Pre-Read error <server>/SYS:SYSTEM\SYS$LOG.ERR│ └────────────────────────────────────────────────────── ───┘ VREPAIR did not fix the problem. SOLUTION The users could still log in the system, so after the system manager did a system backup, reinitialized SYS: volume, and reinstalled NetWare, the problem was solved. Note: When a "write" is performed, the area to which the data will be written is first "read." If for some reason, there is a problem with the read, such as controller or drive failure, then the above message is returned. Novell has seen this happen particularly when a device is deactivated. This could also happen if the block to which the data was to be written needed to be hot-fixed, but hot-fix was full. In the latter case, the situation is more serious. Resizing the hot-fix area will cause a complete reinitialization of the partition tables and require a complete restore from a backup. In the first case of a device deactivation, you could try to reactivate or remount the volume. This may or may not work; however, the drive deactivation needs to be addressed, regardless. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Filename Not Found" - Running JUMPERS Utility DOCUMENT ID#: FYI.P.11119 DATE: 13JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When the user ran the JUMPERS utility from the NetWare drive on an IPX located on the local drive, the system displayed the following error: ┌──────────────────┐ │Filename not found│ └──────────────────┘ The system displayed the same error after the user copied the JUMPERS utility to the local hard drive in the same directory where IPX was. ISSUE/PROBLEM When checking the DOS file attribute of IPX.COM, the user found it had been set to "Read Only." SOLUTION After the user changed IPX.COM DOS file attribute to Read/Write, the system worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Upload of OS2UTI.ZIP DOCUMENT ID#: FYI.P.11118 DATE: 13JAN93 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM Concerning OS2UTI.ZIP available on NetWire or NetWare Express. The following files are included: OS2UTI DOC USERLIST EXE 19523 07-16-92 4:37p WHOAMI EXE 28047 06-30-92 11:54a LOGOUT EXE 16343 06-30-92 11:02a These OS/2 utilities are for all versions of the Requester and fix a problem that occurs when a user performs a log into a network between 12:00 pm and 1:00 pm. The old utilities reported the log in time as AM rather than PM. The new utilities report the correct PM log in time. SOLUTION NA FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Abend: GPPE Running MONITOR Background Process" DOCUMENT ID#: FYI.P.5045 DATE: 20DEC91 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Error on file server: "abend: GPPE running monitor background process" ISSUE/PROBLEM Two users with completely different setups that were running NetWare v3.11 software had the following identical problem: ■ The server would abend every month (almost to the day) with the following message: ┌──────────────────────────────────────────────┐ │abend: GPPE running monitor background process│ └──────────────────────────────────────────────┘ The only thing these users had in common was that they were running NetWare v3.11 software and using the MONITOR NLM v1.67 dated 02-07-91. SOLUTION Apply the patch in MON175.ZIP available on NetWire or NetWare Express. The following is a description of the patch and what it fixes: MON175.ZIP is a NetWare v3.11 MONITOR.NLM and will replace MON311.ZIP. This file (MONITOR.NLM v1.75) includes a fix to keep the console locked in MONITOR under all conditions. In previous versions of MONITOR if an application such as the PSERVER was unloaded remotely through PCONSOLE, it would cause the console to unlock. This file (MONITOR.NLM v1.75) corrects a problem seen in previous versions of MONITOR.NLM. These previous versions could cause a "General Protection Processor Exception (GPPE) running process Monitor Background process" error on the server. This error would generally occur when MONITOR was loaded for extended periods of time (30 days or so) and was caused by resource contention between the foreground and background processes of MONITOR. This version supersedes previous versions and includes all such fixes. This new MONITOR.NLM will also be included in NE386b.ZIP also available on available on NetWire or NetWare Express. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Upload of 311PTC.ZIP DOCUMENT ID#: FYI.P.11117 DATE: 13JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM The document concerning 311PTC.ZIP located on NetWire and NetWare Express. This file replaces 311PTB.ZIP. This file contains the latest release of NetWare v3.11 operating system patches. The following files are included in this release: ┌───────────────────────┐ │User Versions Supported│ ├───────┬───────┬───────┤ │ 1 │ 5-250 │ 1000 │ 311PTC DOC (This file ╞═══════╪═══════╪═══════╡ AFPFIX1 NLM 2663 1-06-92 7:22p │ │ X │ │ ASNCRDFX NLM 1521 9-23-92 5:03p │ X │ X │ X │ BIGRFIX NLM 1425 4-22-92 3:36p │ X │ X │ X │ CLPHFIX NLM 1238 6-16-92 7:59a │ X │ X │ X │ CROPNFX NLM 1040 12-18-91 3:18p │ │ X │ │ DIAGRFSX NLM 2593 3-18-92 8:21a │ │ X │ │ DIRLPFIX NLM 1508 12-12-91 1:07p │ │ X │ │ DOSUNFIX NLM 1910 4-22-92 10:35a │ X │ X │ X │ DTRSTFIX NLM 1384 9-17-92 1:48p │ X │ X │ X │ DUPFIX NLM 1513 9-16-92 10:22a │ X │ X │ X │ EAERRFIX NLM 1241 9-24-92 10:39a │ X │ X │ X │ EAINFIX NLM 1788 5-15-92 9:33a │ X │ X │ X │ EVENTFIX NLM 2498 6-29-92 3:55p │ X │ X │ X │ FATFIX NLM 3357 4-23-92 12:53p │ X │ X │ X │ GETQFIX NLM 1211 3-23-92 2:57p │ │ X │ │ GETRFIX NLM 1311 4-22-92 3:38p │ X │ X │ X │ GETSEFIX NLM 1463 4-22-92 10:41a │ X │ X │ X │ GNSFIX NLM 2509 5-15-92 9:35a │ │ X │ X │ HTFXFX NLM 1036 5-15-92 9:36a │ X │ X │ X │ INDSKFIX EXE 10686 9-24-92 8:34a │ X │ X │ X │ MAXCDCFX NLM 962 5-15-92 9:36a │ X │ X │ X │ MFIXOPEN NLM 1459 10-03-91 5:00p │ │ X │ │ MKDIRFIX NLM 2295 5-15-92 10:02a │ │ X │ X │ MPRPSFIX EXE 10674 9-23-92 9:33a │ X │ X │ X │ NCOMFIX NLM 2109 12-14-92 8:00a │ X │ X │ X │ NCPSPFIX NLM 1760 5-15-92 10:13a │ │ X │ X │ OBPRGFIX NLM 1640 11-06-92 10:49a │ X │ X │ X │ OS2OPNFX NLM 1389 11-22-91 2:45p │ │ X │ │ PAPFIX1 NLM 1145 12-02-91 3:18p │ │ X │ │ PAPFIX3 NLM 1310 1-08-92 10:32a │ │ X │ │ PATCHMAN NLM 9051 4-16-92 11:21a │ X │ X │ X │ PCHMN223 DOC 6110 5-07-92 4:33p │ - │ - │ - │ REMDEVFX NLM 1299 10-03-91 5:04p │ │ X │ │ REQFIX NLM 1251 5-15-92 9:38a │ │ X │ X │ SPXFIX1 NLM 1533 10-21-92 2:31p │ X │ X │ X │ SPXFSFIX NLM 1742 7-20-92 3:02p │ X │ X │ X │ SYNCTIME NLM 2211 11-19-92 2:37p │ X │ X │ X │ TRSTFIX NLM 1688 10-02-92 11:20a │ X │ X │ X │ TTSFIX NLM 1227 5-20-92 8:41a │ X │ X │ X │ UNKFIX NLM 1058 5-15-92 9:41a │ X │ X │ X │ UNLCKFIX NLM 1247 5-15-92 8:01a │ X │ X │ X │ VREPAIR NLM 88941 9-10-92 5:02p │ X │ X │ X │ WORMROFX NLM 1081 10-03-91 4:49p │ │ X │ │ WSDFIX NLM 1753 6-29-92 4:05p │ │ X │ X │ XMDFIX NLM 1496 9-15-92 3:29p │ X │ X │ X │ └───────┴───────┴───────┘ This file supersedes previous releases of the NetWare v3.11 operating system patches distributed on NetWire under the following names: 311PT1.ZIP 311PT4.ZIP 311PT7.ZIP 311PTA.ZIP 311PT2.ZIP 311PT5.ZIP 311PT8.ZIP 311PTB.ZIP 311PT3.ZIP 311PT6.ZIP 311PT9.ZIP The following changes were made to create 311PTC.ZIP: DOTFIX NLM 1343 12-12-91 1:29p (Removed) NCOMFIX NLM 2109 12-14-92 8:00a (New) RENFIX NLM 1363 11-12-92 8:58a (Removed) SYNCTIME NLM 2211 11-19-92 2:37p (Replacement file for SYNCTIME dated 12-17-91 size 2131) DOTFIX.NLM has been removed as a precaution; even though, no actual problems have been encountered. RENFIX.NLM has been removed because this functionality will no longer be supported. SYNCTIME.NLM is now 1000 user compatible and solves the abend problem. ***IMPORTANT NOTES:*** 1. There are some specific notes regarding the use of these patches with any NetWare v3.11 operating system patches dated before October of 1991. See NOVELL Technical Bulletin TB.P.298. 2. A second documentation file by the name of PCHMN223.DOC is included, which accompanies PATCHMAN.NLM. PATCHMAN.NLM is required when loading the dynamic patches (NLMs). 3. Each patch included in this release is listed below. 4. Also included below is a brief discussion of the patch concept and installation instructions. 5. Please note that specific instructions are given for the use of FATFIX.NLM below the description of that patch. 6. The 1-user version or Runtime version patches have been combined into this ZIP file. (1U311 will no longer be available) 7. This ZIP file contains static (EXE) and dynamic (NLM) extensions. The following is a description of each patch: DESCRIPTION OF THE PATCHES AFPFIX1.NLM 2663 01-06-92 7:22p This patch is only for NetWare for Macintosh. AFPFIX1.NLM corrects a problem with copying folders (containing files) into drop boxes on a NetWare v3.11 server. This patch fixes all known drop box problems in System 6 and 7 for all released versions of AFP.NLM. This patch supersedes the original AFPFIX1.NLM version and the correction in AFP.NLM v3.01. ASNCRDFX.NLM 1521 09-23-92 5:03p This patch fixes a problem that may occur when multiple threads do asynchronous reads. It is possible for a thread to read data that was written by another thread. This problem has been known to occur with the BTRIEVE.NLM application if a Btrieve data file page size is not 512, 1024, 2048, or 4096 bytes. If the page size is not one of these sizes, the data file may become corrupt. This version will work with BNETX. BIGRFIX.NLM 1425 04-22-92 3:36p This patch allows the operating system to prevent "read" requests larger than 512KB from overrunning the stack. This problem has been known to occur when loading large NLMs. CLPHFIX.NLM 1238 06-16-92 7:59a This patch fixes the "ClearPhantom called with not primary directory number" abend. CROPNFX.NLM 1040 12-18-91 3:18p This patch fixes a problem found when a client makes a request to create or open a file. When a user makes a CreateOpenFile request, the server creates the file, but the server's reply to the user gets lost. The client then makes a second request to create the file; however, because it has already been created, the user gets the following error message: "file already created" Additionally, when a client request is made to open a file, the server opens the file but its response to the client is lost. The client makes a second request to open the file, and the server opens the file a second time. This scenario can result in the server keeping the file open while the user thinks it has been closed. This problem becomes apparent when the user attempts to delete the file and receives the following error message: "File in use" DIAGRFSX.NLM 2593 03-18-92 8:21a This patch fixes two network management errors. The first error caused the server not to respond to requests issued by the network management console (only rebooting would fix the problem). The second error did not allow the file server to respond to network management requests directed to other networks connected to the file server other than the internal network. DIRLPFIX.NLM 1508 12-12-91 1:07p This patch fixes a problem in NetWare 386 v3.11 that results in an infinite loop when the DOS DIR command is used in a subdirectory where more than 65,535 files exist. This loop will also occur with the FindFirst and FindNext DOS function calls. DOSUNFIX.NLM 1910 04-22-92 10:35a This patch fixes a problem with NetWare/NFS v1.2. If a file is renamed from 99999999.v0 to 99999999.V0, it causes the server to ABEND with the following message: "DOSGenerateUniqueName tried over 100000000 names and couldn't get a unique one." DTRSTFIX.NLM 1384 09-17-92 1:48p DTRSTFIX.NLM fixes a problem in which various GPPEs and abends occur because of an invalid Trustee Node pointer in a Directory Entry. DUPFIX.NLM 1513 09-16-92 10:22a This patch fixes a problem in which a certain sequence of events would cause duplexed drives to report all mirrored drives were synchronized after a hardware failure when they were not. DUPFIX.NLM must be loaded in memory at the time a hardware error occurs with the disk drive. EAERRFIX.NLM 1241 09-24-92 10:39a EAERRFIX.NLM fixes an Invalid Semaphore abend that is initially caused by an ERR_EA_NOT_FOUND being returned in the Error Code on a Read Extended Attribute NCP reply. EAINFIX.NLM 1788 05-15-92 9:33a This patch fixes a problem with Extended Attributes when two or more work stations execute XCOPY multiple times. The result is that after about ten minutes the server will stop and send the following ABEND message: "Free called with a memory block that has an invalid resource tag." EVENTFIX.NLM 2498 06-29-92 3:55p This patch is required by any NLM that needs object deletion notification. The symptoms of this bug will vary depending on the NLM being run. It addresses a problem that occurs when a "delete bindery object" call is made to the operating system. The operating system reports the event after the object is already deleted, thus preventing the application from accessing information on the object when the event is reported. Under this patch, the bindery object is not deleted until after the event is reported. FATFIX.NLM 3357 04-23-92 12:53p This patch fixes a problem with TurboFAT and FAT corruption, in general, when multiple workstations randomly access large database files. (The minimum size of a TurboFat database file is, using a default 4KB block size and 64 blocks as the TurboFat, 64 * 4kb = 256kb. If you have set up a different block size, substitute that number.) ********* IMPORTANT SPECIFIC INSTRUCTIONS FOR FATFIX.NLM ********* It is important to note that the latest version of VREPAIR.NLM should be run before implementing this patch. (Use the VREPAIR.NLM that is included with this ZIP file.) The FATFIX.NLM will prevent corruption of the FAT and TURBOFAT; however, VREPAIR is required to correct previous corruption. The steps below should guide you through using FATFIX.NLM. 1. To use this patch, add the following two lines to your AUTOEXEC.NCF file after the assignment of the "IPX INTERNAL NET" number (See page 150 of the NetWare Version 3.11 Installation manual for an example of this statement): LOAD PATCHMAN LOAD FATFIX It is important that these lines are added immediately after the assignment of the internal network number. After editing, be sure to save the file. 2. After editing the AUTOEXEC.NCF file, do the following things: 2a. LOAD VREPAIR at the console. 2b. Dismount the volumes. 2c. Run VREPAIR on all volumes until there are no further errors reported (this may take one or more attempts). 2d. Shut down the server. 2e. Bring the server back up with the modified AUTOEXEC.NCF file. GETQFIX.NLM 1211 03-23-92 2:57p This patch fixes a problem that caused a large number of deleted .SRV files in the queue subdirectories in SYS:SYSTEM. GETRFIX.NLM 1311 04-22-92 3:38p This patch fixes an abend caused by a driver requesting a valid linked request that has been changed by another process. A zero is now returned to the calling process instead of stopping the server and sending an abend message. GETSEFIX.NLM 1463 04-22-92 10:41a This patch fixes a problem found when more than 31 security equivalences for a user or group ID are set up. This can result in a system hang which may be accompanied by a message indicating a "multiple abend." GNSFIX.NLM 2509 05-15-92 9:35a This patch fixes a problem found when rebooting and attempting to get a connection to the file server. If all the connections are in use (such as 20 for the 20-user version or 50 for the 50-user version) and one of the users reboots, the user will not be able to log back into the server. This problem will only occur if there is only one network. If the user is on an internet, a reconnection will be possible. The problem is that after a user reboots, the connection is not terminated at the server until a time-out limit is reached (usually 15 minutes). This patch corrects the problem by checking the source node address of the incoming packet to the node address of all workstations that the server recognizes it is attached to. If a match is found, the server responds to the workstation thus recreating a connection in the same slot the workstation had before rebooting. HTFXFX.NLM 1036 05-15-92 9:36a This patch corrects a problem when using disk block sizes larger than 4KB and cache buffer sizes larger than 4KB. If a block was read from the redirection area while copying a file in the above circumstance, it could cause the destination block to become corrupted. INDSKFIX.EXE (static) 10686 09-24-92 8:34a InDskFix.EXE changes the SERVER.EXE file on disk to fix the following abend: "The sleeping process was prematurely awakened during Delay." MAXCDCFX.NLM 962 05-15-92 9:36a This patch changes the maximum limit for the Set parameter "Maximum Concurrent Disk Cache Writes" from the old range of 10 to 100 to a new range of 10 to 1000. This change will be transparent to most users. The performance of some high throughput disk controller drivers is limited by the slower limit. You may want to load this NLM if your system uses high throughput EISA drives. MFIXOPEN.NLM 1459 10-03-91 5:00p This patch is intended for all number-of-user versions of NetWare 386 v3.11 that are running the Macintosh gateway VAPs instead of the AFP NLMs. This patch fixes the problem that Macintosh users have when trying to open files that are flagged Read-Only. The user is unable to open the file and receives the following error message: File not found MKDIRFIX.NLM 2295 05-15-92 10:02a This patch corrects invalid error codes returned by the "create directory" call or when copying a file to an invalid path. It replaces and supersedes the previously released patch, DIRHFIX1.NLM. Because DIRHFIX1.NLM and MKDIRFIX.NLM patch some of the same areas, they cannot coexist. To run MKDIRFIX.NLM, the DIRHFIX1.NLM must be removed from the system. Additionally, this fix changes error codes returned from "ERR_BAD_DIR_HANDLE-program aborted" to "ERR_INVALID_PATH" in the case where the server is scanning a directory path and a file name exists in the middle of that directory path. (For example: MAKEDIR: DIR1/DIR2/DIR3 where DIR2 is a filename.) MPRPSFIX.EXE (static) 10674 09-23-92 9:33a MPRPSFix.EXE changes the SERVER.EXE file on disk to allow the Maximum Physical Receive Packet Size to be 24,682 bytes. Previously the allowable limit was set at 4,202 bytes. NCOMFIX.NLM 2109 12-14-92 8:00a NComFix.NLM fixes the abends "Resource tag link error detected in NCompleteRequest" and "Invalid disk request passed to ReturnDiskRequest." NCPSPFIX.NLM 1760 05-15-92 10:13a This patch prevents a zero-length WriteToSpool NCP from being processed. This problem has been known to cause an ABEND:GENERAL PROTECTION PROCESSOR EXCEPTION and corrupt an entire volume. OBPRGFIX.NLM 1640 11-06-92 10:49a ObPrgFix.NLM fixes the abend "A string with too long a length was passed to the output formatter." OS2OPNFX.NLM 1389 11-22-91 2:45p This patch allows a user to use the TYPE command to view a file even though SCAN file rights have not been granted. Without this patch, if the user is granted all rights but the SUPERVISOR and SCAN file rights (including Read), the file still cannot be viewed using the TYPE command. This patch fixes this problem. PAPFIX1.NLM 1145 12-02-91 3:18p This patch is only for v3.0 NetWare for Macintosh. PAPFIX1.NLM is a patch to APPLETLK.NLM v3.0. This patch resolves a problem where server PAP sessions of greater than 255 transactions would be interrupted and fail. The AppleTalk stack would then generate an error message stating that the Tickle Timer popped on the affected connection. This problem would cause ATPS.NLM to consistently fail to print files over one megabyte in size or to return font lists greater than 255 entries to Macintosh clients. In these circumstances, ATPS would return the following message: "PAP write: no such device or address" This patch is for use with NetWare for Macintosh v3.0; the problem is resolved with NetWare for Macintosh v3.01. PAPFIX3.NLM 1310 01-08-92 10:32a This patch is only for NetWare for Macintosh v3.0. PAPFIX3.NLM modifies APPLETLK.NLM v3.0 to resolve a problem that ATPS.NLM has connecting to certain non-Apple printers that require an incrementing value in the WaitTime field of PAP OpenConn packets. The symptom of this problem visible to users would be that ATPS.NLM generates messages such as "Printer busy, status- idle" and may not advertise the affected queue on AppleTalk. This problem is resolved with NetWare for Macintosh v3.01. PATCHMAN.NLM 9051 04-16-92 11:21a PATCHMAN.NLM is the Patch Manager that tracks and manages all dynamic patches for NetWare 386 v3.10 or above. All NetWare v3.11 Operating System patches included with this documentation will require PATCHMAN.NLM v2.20 or above. The version of PATCHMAN included in this file is v2.23. Please see PCHMN223.DOC that is included with this file for further explanation on PATCHMAN.NLM. REQFIX.NLM 1251 05-15-92 9:38a This patch reinitializes the number of requests made from a station each time a new user logs in. This problem was visible when looking at data returned by PAUDIT. If an individual workstation was used to log in for several different users, the "requests" value displayed in PAUDIT was not reinitialized for each user. REMDEVFX.NLM 1299 10-03-91 5:04p This patch fixes a problem formatting removable devices. If INSTALL.NLM is loaded and FORMAT is run, the NetWare v3.11 operating system causes an abend with the following message: "ABEND: Kernel detected a process switch during interrupt time." SPXFIX1.NLM 1533 10-21-92 2:31p This patch corrects problems for certain applications that may hang due to lost Event Control Blocks (ECBs). SPXFSFIX.NLM 1742 07-20-92 3:02p This patch fixes a problem that results in a file server GPI in the SPXFindSession. This ABEND is known to occur when using Madge "Smart IPX" Token-Ring drivers. The problem is caused by a connect request SPX packet that arrives at the server without the system bit being set. This causes the server to look up the session with ID = FFFF (session ID for system packets); however because the system bit is not set, the server looks in the session table for a client with a corresponding ID. This causes SPXFindSession to look beyond the end of memory, resulting in a GPI ABEND. The patch adds a test to SPXFindSession to check for sessions beyond the end of the session table. With this patch, if a connect-request SPX packet is sent without the system bit set, the packet will be discarded as invalid and the workstation that sent the invalid packet will hang. SYNCTIME.NLM 2211 12-19-92 2:37p This patch forces the server's date/time to be synchronized to the hardware's clock at specified intervals. These NLMs eliminate the gradual loss of time experienced by the server due to "missed" clock ticks while the server is in real mode. **Note: This file should not be confused with TIM386.ZIP located in forum NOVLIB, Library 4, that is for NetWare 386 v3.10a and specifically refers to v3.10a in the documentation that comes with that file. TRSTFIX.NLM 1688 10-02-92 11:20a This patch fixes a problem found when nine or more trustee directory assignments for individual users or groups are made at the root level of the same volume (usually volumes other than the SYS: volume). Although the assignments are made as they should be, the ninth and above cannot be seen from SYSCON. (They do appear in FILER or TLIST.) This patch adds a check for the parent of the trustee node that contains the ninth trustee assignment. If the parent node is the root of the volume, it checks for trustee assignments. This patch also fixes a problem found when granting trustee rights to an existing group. The rights appear to be correct in SYSCON when viewed from user SUPERVISOR, but these new rights do not appear when viewed by the individual user. These rights are not in effect. When viewed as a group member, the system displays the following error: "GetTrusteePaths returned error 242" TTSFIX.NLM 1227 05-20-92 8:41a TTSFIX.NLM fixes a bug met when certain applications execute explicit Transition Tracking System (TTS) and file locking is also used. This results in an application hang at the workstation and a GPPE ABEND at the file server. UNKFIX.NLM 1058 05-15-92 9:41a This patch fixes a problem that was causing an abend on a NetWare v3.11 file server. This is caused by entering a complete file name of a valid NetWare file at the file server console. This file may be located on any of the server's search paths. For example: Typing "MONITOR.NLM" at the console instead of "LOAD MONITOR.NLM" will cause the above condition. Typing "MONITOR" without the .NLM extension will not cause the abend. If the named file is not an .NCF file, the server will repeatedly display: "Unknown command" and "Invalid line in command file-line length exceeds 511 characters" until it eventually the system would abend with the message: "ABEND: General Protection Processor Exception running console command process." The operating system is incorrectly attempting to open and read the NLM as if it was an .NCF file. UNLCKFIX.NLM 1247 05-15-92 8:01a This patch corrects a problem that causes an abend with the following message: "UnlockDirectoryEntry called with entry that was already unlocked" This problem occurs when the system tries to read the primary directory table and fails. It then attempts to read the mirrored table and even though it appears to read the table successfully, the information is corrupt and the system stops and send the abend message. VREPAIR.NLM 88941 09-10-92 5:02p This version fixes problems where VREPAIR has not been able to correct errors in the FAT or directory tables. VREPAIR has been enhanced to check for errors and in the algorithm used in comparing values for the correct entry. WORMROFX.NLM 1081 10-03-91 4:49p Currently, if a removable device is initially mounted read/write and then the cartridge is removed and a new device is inserted that has the write disable tab set, the operating system does not recognize that the device is now Read-Only. This patch fixes this problem. WSDFIX.NLM 1753 06-29-92 4:05p This patch fixes a problem with APIs that allow users to access files on volumes where they have no rights. This patch generates a new handle so that rights are reenumerated with each volume change. XMDFIX.NLM 1496 09-15-92 3:29p This patch corrects a problem with utilities that use the SPX protocol to maintain communication with the server. Typically the client will lose the connection to the server and the following message will be displayed: "No more response buffers available" The most common occurrence is with a remote printer (RPRINTER.EXE) losing its connection to the print server (PSERVER.NLM); however it may happen with other utilities. This is happens when a call to SPXSendSequencedPacket() resets a session retransmit count that causes it not to abort the connection if the connection is broken in one direction and maintained in another. SOLUTION NA FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Names Stored in Bindery Are in Upper Case Only DOCUMENT ID#: FYI.P.11109 DATE: 13JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Application returns full name all in capital letters. ISSUE/PROBLEM Using the MAKEUSER utility, the user specified a full name in lowercase because he had an application that looked for this name. When the name was called, it was returned in all caps. The user wanted to know why. SOLUTION All objects stored in the bindery are in uppercase only. Whenever there is a call made to get the name, it will be returned in uppercase. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "There Are No Accessible Disk Drives with NetWare Partitions" DOCUMENT ID#: FYI.P.11101 DATE: 13JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM After registering memory and then loading the disk driver, the user could not mount the volumes. The server displayed the following errors: ┌────────────────────────────────────────────────────── ─────┐ │There are no accessible disk drives with NetWare partitions│ └────────────────────────────────────────────────────── ─────┘ ┌───────────────────────────────┐ │Abend: ...AES no sleep process.│ └───────────────────────────────┘ If the user loaded the disk driver before registering memory, he could mount the volumes. ISSUE/PROBLEM Configuration: ■Dell 486-50 EISA ■Adaptec 1740 HBAs The user was setting auto register memory=off in STARTUP.NCF and manually registering memory in the AUTOEXEC.NCF. However, the user was improperly registering memory using the following command: REGISTER MEMORY 1000000 4F00000 SOLUTION After the user changed the command to the following, everything worked fine: REGISTER MEMORY 1000000 5000000 On page 216 of the NetWare Version 3.11 System Administration manual, it states the following: "...(the length) must be on an even paragraph boundary (a number that is divisible by 10h)" Because it was within the 96MB boundary, the user thought that he could use 4f00000. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: DOS SET Command Not Working with LOGIN_NAME DOCUMENT ID#: FYI.P.11142 DATE: 14JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM DOS set command not working with Login_Name variable ISSUE/PROBLEM Configuration: ■LOGIN.EXE v3.60 and v3.65 Connection 7 on the server could not execute the following command: "Dos set menuworks=/swap:f:\menuworks\%login_name.$$$" SOLUTION Novell was unable to duplicate this symptom. Using P_Station instead of Login_Name worked fine for all connections. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Slow Floppy Diskette Access - DR DOS, Norton Desktop DOCUMENT ID#: FYI.P.11137 DATE: 14JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM Slow floppy diskette access ISSUE/PROBLEM Configuration: ■Using Norton's Desktop for Windows on a 386 The user's floppy diskette access is very slow from Windows or from the DOS prompt. This is a known problem with the first time the floppy diskette is accessed; however, in this case the floppy diskette access is always slow. SOLUTION Initially, when the user removed SUPERPCK, the floppy diskette access returned to normal. After the user added to the following to the SuperPCK line, the floppy diskette access is operating at normal speeds with the cache enabled: /-a* /-b* /-e FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Changing an NNS File Server Name DOCUMENT ID#: FYI.P.11135 DATE: 14JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Changing an NNS file server name causing random password assignments ISSUE/PROBLEM Why would using the following procedure to change the name of a NetWare Name Service (NNS) server in a domain cause the user's 600-user accounts to have random passwords assigned? 1. Edit the AUTOEXEC.NCF file to the new name. 2. Shut down the server. 3. Bring the server back up in the same domain. SOLUTION The NNS binderies throughout the domain still had the old server name in the bindery. Renaming the server without first removing it from the domain corrupted the bindery. Complete the following procedure to rename a NNS server: 1. DELETE the server from the domain. 2. Rename the server in the AUTOEXEC.NCF file. 3. ADD the server back into the domain. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Problems Resynchronizing Duplexed Drives DOCUMENT ID#: FYI.P.11130 DATE: 14JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM While running on a NetFrame, a primary drive of a duplexed pair went bad and deactivated itself. The secondary drive took over, and for a day and a half updates were made to the data on the secondary drive. The user ran VREPAIR on the primary drive and reinstalled it into the system. After the user shut down the server and brought it back up, the drives came up duplexed; however the updated information was not kept (there was no visible sign that the drives were doing any resynchronizing). The secondary drive resynchronized itself to the primary rather than the other way around; therefore, all the data that had been updated over the last 36 hours was lost. ISSUE/PROBLEM Apparently, in some rare cases, the synchronization bit on the secondary drive is not reset or updated to show that it has more recent information. Because of this, when the primary drive is brought back online, the operating system checks the synchronization bits on both drives and sees no changes or updated time stamps, so it mirrors them back as before. All the updated information is lost. SOLUTION Novell was able to recreate this in some rare cases and built a fix called DUPFIX.NLM (in 311PTB.ZIP and later) available on NetWire or NetWare Express. Novell has a list of some of the specific scenarios of this problem. It is not easy to recreate but does happen. Before this case, Novell has not seen the problem on a NetFrame. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: How NetWare v3.11 Handles Volumes Spanning Drives DOCUMENT ID#: FYI.P.11150 DATE: 15JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM How does the NetWare 3.11 server handles volumes spanning drives? SOLUTION When NetWare spans two disk drives with one volume, the operating system will stripe the data across the two drives (the data is written across both drives). This striping will take place until one of the drives runs out of space. For example, if the SYS: volume is on one drive and it is half full of data and another drive is added and the segment is added to the SYS: volume, then striping will start until the first drive is filled. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: In-place Migration Utility Aware of Duplexed Drives DOCUMENT ID#: FYI.P.11149 DATE: 15JAN93 PRODUCT: NetWare PRODUCT VERSION: v2.x, v3.x SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM Can the in-place MIGRATION utility for NetWare v2.2 to v3.11 handle mirrored or duplexed situations? Can it handle the situations where one of the drives or channels has failed, especially if it is the primary drive or channel? SOLUTION Yes. The in-place migration utility is recognized and searches for mirrored or duplexed situations. The in-place migration utility is designed to handle situations when one of the drives or channels has failed. Novell technicians ran the following test to ensure the in-place MIGRATION utility does recognize and handle the previously mentioned situations. The Novell technicians set up a NetWare v2.15 server with mirroring. They failed the primary drive, then shut down the server. They brought the server back up by booting to a floppy diskette to verify that the NetWare v2.15 server would come up and recognize a problem with the primary drive. Then they followed the necessary step to run the MIGRATION utility. During the step where the migration utility searches for drives, it recognized drive 0 as not having a NetWare v2.15 or v2.2 partition on it and recognized drive 1 as having the SYS: volume. This result is what they expected. Then the in-place MIGRATION utility upgraded the secondary drive and allowed a DOS partition to be put on it as well. Then they shut down the NetWare v3.11 server and brought it back up after having corrected the disk problem. The old NetWare v2.15 primary drive appeared in INSTALL as a pre-386 NetWare partition. Then they erased that partition and created a 386 partition and reestablished mirroring. Everything worked as expected. Keyword: Inplace FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "No Free Connection Slots" - SMCARC.LAN, ARCnet DOCUMENT ID#: FYI.P.11147 DATE: 15JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user's system was receiving both of the following errors at different times: ┌────────────────────────────────┐ │A file server could not be found│ └────────────────────────────────┘ ┌────────────────────────┐ │No free connection slots│ └────────────────────────┘ ISSUE/PROBLEM The user only knew that he had an ARCnet board in the server and that he was using a TRXNET driver that loaded fine. Track On showed that Get and Give Nearest Server were being completed. However, the DOS C: partition showed that the user had a TRXNET.LAN and SMCARC.LAN board. SOLUTION After the user loaded the SMCARC driver, everything worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Abend: Subdirectory Was Deleted..." During Backup DOCUMENT ID#: FYI.P.11146 DATE: 15JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM "ABEND: Subdirectory was deleted during FindDirectoryEntry" ISSUE/PROBLEM The user was doing a server tape backup. During the backup, the user was logged into the network to clean some directories. While cleaning the directories, the server stopped and sent the following abend message: ┌────────────────────────────────────────────────────── ───┐ │ABEND: Subdirectory was deleted during FindDirectoryEntry│ └────────────────────────────────────────────────────── ───┘ SOLUTION When deleting directories on the file server while the tape backup was running, the user deleted files and a directory that the tape backup was currently backing up, thus causing the error. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Sector Not Found Reading Drive A:" - DR DOS DOCUMENT ID#: FYI.P.11170 DATE: 18JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM "Sector not found reading drive A:" ISSUE/PROBLEM The user recently did an FDISK on the drive and installed DR DOS 6.0. When trying to reinstall other software, many of the disks that he tried to use in the A: drive gave the following error: ┌─────────────────────────────────┐ │Sector not found reading drive A:│ └─────────────────────────────────┘ Normally, this is just associated with faulty floppy diskettes or misaligned or dirty drive heads. SOLUTION After some research, the user said that he also experienced a problem with doing a DIR on a floppy diskette, changing the diskette, and then getting the same DIR list as the first diskette. This symptom is generally caused from a SUPERPCK problem. This "Sector not found" problem was also being caused by SUPERPCK. After the user added /-a* parameter to the SUPERPCK line in the AUTOEXEC.BAT (/-a* means ignore the A: drive), both symptoms disappeared. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: RCONSOLE Copies SYSTEM and PUBLIC Files from Network Drive DOCUMENT ID#: FYI.P.11168 DATE: 18JAN93 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM The INSTALL.NLM with -j option allows you to copy SYSTEM and PUBLIC files from the A: or B: drive but not from a network drive. How can I copy these files from a network drive? SOLUTION To copy SYSTEM and PUBLIC files from a network drive, use the RCONSOLE from a workstation. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Using In-place UPGRADE with Existing DOS Partition DOCUMENT ID#: FYI.P.11166 DATE: 18JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM The user was doing a NetWare in-place Upgrade. When performing the NetWare v2.x to v3.x in-place upgrade, you have the option of creating a DOS partition at the beginning of the drive before the creating the NetWare v3.11 partition. What happens if you already have a DOS partition at the end of the drive, past the NetWare partition? SOLUTION If you select the option to create a DOS partition, it will do so and leave the other DOS partition intact as well, giving you two DOS partitions on each end of the drive. The reason that this happens is that the in-place UPGRADE only leaves free space on the drive for a partition to be created. To complete the creation of the partition, you still need to run DOS's FDISK and format the partition. Keyword: Inplace FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Abend: Invalid Tnode Found by AddTrusteeRights" DOCUMENT ID#: FYI.P.11164 DATE: 18JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM "Abend: Invalid Tnode found by AddTrusteeRights" ISSUE/PROBLEM The user's system received the following error two or three times a day: ┌──────────────────────────────────────────────┐ │Abend: Invalid Tnode found by AddTrusteeRights│ └──────────────────────────────────────────────┘ SOLUTION Applying the DTRSTFIX.NLM found in 311PTC.ZIP, available on NetWire or NetWare Express, solved the problem. Note: The README for DTRSTFIX.NLM states the this file "fixes problems in which various GPPEs and abends occur because of an Invalid Trustee Node pointer in a Directory Entry." FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Local Mappings Override Network Search Drives - RPL DOCUMENT ID#: FYI.P.11163 DATE: 18JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Search drives getting written over ISSUE/PROBLEM When the user logged in after doing a Remote Program Load (RPL), the local mapping overrode the network search drivers. SOLUTION Removing "last drive = z" from the CONFIG.SYS file solved the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Number of Partitions Compressed with DR DOS DOCUMENT ID#: FYI.P.11160 DATE: 18JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM Problems compressing the ninth partition ISSUE/PROBLEM Is there a limit to the number of partitions you can compress with the DR DOS v6.0 of SuperStor? SOLUTION Yes. Eight is the limit. If you SSTOR nine, you will experience unpleasant behavior, such as being unable to access one of the compressed drives or the data it contains. A possible way to recover the compressed drive that is not accessible in the above scenario it the following: Backup one SSTOR partition (one that is still accessible), then remove it and reboot. At that point, there are again only eight partitions and you may get the "lost" one back. Note: This is only a theory, and it may not work. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Copying Binary Files to LPT1 with DR DOS DOCUMENT ID#: FYI.P.11159 DATE: 18JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM "Error writing to device, 0 files copied." ISSUE/PROBLEM While using DR DOS v6.0 and the "COPY FILENAME.EXT LPT1: /B" syntax to send a binary file to LPT1, the drive came on as if it was executing the command then displayed the following error: ┌────────────────────────────────────────┐ │Error writing to device, 0 files copied.│ └────────────────────────────────────────┘ When the user booted with MS-DOS, executing the copy command worked. SOLUTION The functionality of DR DOS allows you to place the /B switch (or any other switch) on the source file or the destination device for the file to be copied. The above syntax was only allowing the /B switch to be applied to the destination device for the COPY command. By changing the syntax to the following, the /B switch was applied to the source file only, which was successfully "copied" to LPT1: port: COPY FILENAME.EXT /B LPT1 FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: AUTOEXEC.NCF on C: or A: Drive Not Found DOCUMENT ID#: FYI.P.11195 DATE: 18JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM AUTOEXEC.NCF on C: or A: not found. ISSUE/PROBLEM After the SYS: volume mounts, the server no longer looks to the C: or A: drive for the AUTOEXEC.NCF. The user wanted to use a floppy drive to boot the server. He wanted to have SERVER.EXE, STARTUP.NCF and AUTOEXEC.NCF on the A: drive. Some testing was done with the following results: STARTUP.NCF (on A: drive) set auto register memory above 16 megabytes=OFF load a:aha1540 port=330 int=b AUTOEXEC.NCF (on A: drive) file server name NTS_BC ipx internal net 1083600 The user tried the above configuration on the C: partition as well. In each case, the SYS: volume mounted immediately after the AHA1540 driver loaded. After the SYS: volume mounted, the "FILE SERVER NAME:" prompt appeared on the console indicating that AUTOEXEC.NCF had not been found. Clearly, SERVER.EXE does not search on local drives for AUTOEXEC.NCF after the SYS: volume has mounted. SOLUTION The following setup worked: STARTUP.NCF (on A: drive) set auto register memory above 16 megabytes=OFF AUTOEXEC.NCF (on A: drive) file server name NTS_BC ipx internal net 1083600 load aha1540 port=330 int=b mount all The same order of commands was tested from the C: drive using the ISADISK and AHA1540 with the same results. Besides the previous solution, if you omit the STARTUP.NCF, the AUTOEXEC.NCF will be recognized and executed properly. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Errors Mounting Volume and Registering Memory Above 16 Meg DOCUMENT ID#: FYI.P.11192 DATE: 19JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Everything had been running fine until the user copied more data to the VOL1: volume (700MB of data). After copying more data to the VOL1: volume and rebooting, it would not mount and the system returned the following errors when attempting to mount it: ┌────────────────────────────────────────┐ │Insufficient memory for directory tables│ └────────────────────────────────────────┘ ┌─────────────────────────────────────────────────────┐ │1.1.39 Cache memory allocator out of available memory│ └─────────────────────────────────────────────────────┘ ┌───────────────────────────────────────┐ │Not enough memory to allocate directory│ └───────────────────────────────────────┘ The SYS: volume mounted fine. ISSUE/PROBLEM Configuration: ■80MB RAM in an EISA machine ■SYS: volume was 1.9GB ■VOL1: volume was 2.3GB When the user added the data to his VOL1, apparently more directory entries were allocated causing the system to require more memory than was available before the memory above 16MB was registered. The SYS: volume was automatically mounting when the disk drivers were loaded in the STARTUP.NCF file and before the memory above 16MB was registered with the REGISTER MEMORY command in the AUTOEXEC.NCF. This clone EISA machine would not automatically-register the memory above 16MB. SOLUTION Novell instructed the user to do the following to solve the problem: 1. Move the AUTOEXEC.NCF to the C: partition and load the disk drivers from the AUTOEXEC.NCF instead of the STARTUP.NCF. This kept SYS: volume from mounting automatically. 2. Put the REGISTER MEMORY command before the MOUNT command so that the volumes could use more than 16MB when mounting. This allowed VOL1 to mount. With drivers other than Adaptec's, it is not always necessary to load the driver below 16MB. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Bad Resource Tag Passed..." - Under Heavy Loads DOCUMENT ID#: FYI.P.11187 DATE: 19JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Under very heavy load, the file server would stop and send a "Bad Resource Tag passed to LSLReturnRcvECB" abend message. ISSUE/PROBLEM Under a very heavy load, the user's system received the following error: ┌──────────────────────────────────────────┐ │Bad Resource Tag passed to LSLReturnRcvECB│ └──────────────────────────────────────────┘ An analysis of the Core dump indicated a problem with the GEEBV3.LAN driver v1.22B (920810) for Gateway's 32-bit EISA Ethernet LAN board. However, after the user switched to a SMC 32-bit EISA board, the system still periodically hung under heavy usage. SOLUTION Replacing the server LAN board with a 16-bit board has resolved the issue. The user was happy with this solution and was unwilling to debug any more with the 32-bit LAN board. Keywords: card, 32bit, 16bit FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: OS/2 Clients Unable to Process SAPs from Cisco Router DOCUMENT ID#: FYI.P.11186 DATE: 19JAN93 PRODUCT: NetWare Requester for OS/2 PRODUCT VERSION: v2.0 SUPERSEDES: NA SYMPTOM Occasionally, OS/2 Clients could not connect to Microsoft SQL (v4.2) Named Pipe Servers on the other side of a Cisco router. DOS Clients would connect to SQL every time. ISSUE/PROBLEM The OS/2 Client could not process all the Service Advertising Protocol (SAP) traffic coming from the Cisco router. Because the OS/2 named Pipes table is kept up to date from the SAPs, the SQL connection was very intermittent. SOLUTION Increasing the number of buffers in the Link Support section of the NET.CFG gave enough buffering so the client could process all the SAPs. This specific user was on Token-ring and changed the buffers as shown: From: Link Support buffers 14 4202 To: Link Support buffers 28 2154 The user's internetwork had about 200 SAP devices. The Cisco router would order these SAPs according to SAP type, starting with the smallest to the largest type (File Server=Type 4h, Named Pipes Server=type 9Ah). With 200 SAP devices on the network, the Cisco Router would send out about 30 SAP packets (seven SAP devices in each packet) every minute. The OS/2 Client was set to have 14 buffers. The Cisco would burst out these 30 packets. The packets with the Named pipe server type=9Ah started at about packet 15 in the 30 packet burst. Therefore, by increasing buffers to 28 2154, the system was able to pickup all the SAP information needed. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Error Attempting to access Illegal..." - NetWare v3.11 DOCUMENT ID#: FYI.P.11209 DATE: 20JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The system kept receiving the following error when the user dialed into a NetWare v3.11 network through a Vectra modem using Carbon Copy and tried to log in as Supervisor: ┌────────────────────────────────────────────────────── ───┐ │Error attempting to access illegal or downed file server.│ └────────────────────────────────────────────────────── ───┘ All other users could log in fine. ISSUE/PROBLEM The user Supervisor was a member of eight different groups. SOLUTION Taking the user Supervisor out of these groups solved the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Removing a Password on a Directory - DR DOS DOCUMENT ID#: FYI.P.11202 DATE: 20JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM Unable to remove a password on a directory using the documented syntax on the Help screen ISSUE/PROBLEM The user successfully put a password on a directory but cannot remove it. The documentation on Help screen (PASSWORD /H) says that using /NP removes password protection from the directory. When the user tried using the syntax given in the Help screen, a variety of errors is returned. SOLUTION First, enter the password-protected directory by entering the password. For example, if the directory is called TESTDIR\PROTECTD and the password is "sesame," enter it by typing the following command: C:\> CD TESTDIR\PROTECTD;sesame After entering the directory, remove the password by typing the following: C:\> PASSWORD C:\TESTDIR\PROTECTD /NP Note: It is not necessary to give the password again! If you give the password again after the /NP, you will the following error message: "invalid attempt to both set path password (with /P option) and remove it with /N option" FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Abend: AllocSemiPermMemory..." - SFT NetWare III DOCUMENT ID#: FYI.P.11199 DATE: 20JAN93 PRODUCT: NetWare SFT Level III PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM "Abend: AllocSemiPermMemory got an invalid resource tag." ISSUE/PROBLEM When the user tried to activate the SERVER in both IO_ENGINES at the same time, the server stopped and sent the following abend message: ┌───────────────────────────────────────────────────────┐ │Abend: AllocSemiPermMemory got an invalid resource tag.│ └───────────────────────────────────────────────────────┘ SOLUTION Activating SERVER in both IO_ENGINES is not allowed and will generate the previously mentioned error message. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: How to Bypass Default Login Script without User Script DOCUMENT ID#: FYI.P.11197 DATE: 20JAN93 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM How can a user bypass the default login script without having a user login script? SOLUTION The later versions of LOGIN.EXE (such as v3.7 in LOG370.ZIP available on NetWire or NetWare Express) have added a parameter called NO_DEFAULT that allows the user to bypass the default login script without having an individual login script. As another solution, do the following: Put EXIT at the end of the system login script if you want to bypass the default login script without creating an individual login script. This procedure is documented in Example 1 on page 287 of the NetWare Version 3.11 Installation manual. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Unable to Migrate Binderies from NetWare v3.11 to v3.11 DOCUMENT ID#: FYI.P.11183 DATE: 19JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When the user tried to migrate from a NetWare v3.11 to v3.11 binderies only using the option for 3.X to 3.11 migration, the system displayed the following error: ┌───────────┐ │Error 1.0.1│ └───────────┘ ISSUE/PROBLEM Configuration: ■Using MIGRATE from MIGRAT.ZIP available on NetWire or NetWare Express. This functionality is not built into the MIGRATE utility yet. It should be in future releases. SOLUTION As a workaround, Novell instructed the user to run BINDFIX, copy .OLD files to the other server, and run BINDREST. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Token-Ring DMA Bus Master 4 Microcode Not Loaded" DOCUMENT ID#: FYI.P.11182 DATE: 19JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When the user loading TOKENDMA for the bus master Token-Ring board from IBM (16/4 adapter/A), the system displayed the following error: ┌────────────────────────────────────────────────────── ──┐ │Error: token ring DMA bus master 4 MicroCode not loaded.│ └────────────────────────────────────────────────────── ──┘ ISSUE/PROBLEM FIRMLOAD.COM searches for two files, MONT400.BIN and WRTRAM.BIN, that contains the microcode for this board. FIRMLOAD (formerly DOWNLOAD.COM) places that microcode on the board. FIRMLOAD.COM must be run each time the machine is turned on (automate in the AUTOEXEC.BAT) SOLUTION When the user executed FIRMLOAD.COM before running SERVER, the board initialized and the error disappeared. Note: FIRMLOAD.COM comes on the SYSTEM-2 diskette or TOKENB.ZIP. TOKENB.ZIP is available on NetWire or NetWare Express. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Files and Directories Created Running a Print Server DOCUMENT ID#: FYI.P.11179 DATE: 19JAN93 PRODUCT: NetWare PRODUCT VERSION: v2.2, v3.11 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM What files/directories are created when using a print server? SOLUTION SYS:SYSTEM <DIR> C100002D <DIR> - printserver ID PRINT .000 - printer definition for printer 0 QUEUE .000 - queues servicing printer 0 for each NOTIFY .000 - notify list for printer 0 print server PRINT .nnn - printer definition for printer n QUEUE .nnn - queues servicing printer n NOTIFY .nnn - notify list for printer n 8600003D.QDR <DIR> - print queue ID 0086001 .Q - print job 1 for each 0086002 .Q - print job 2 queue 008600n .Q - print job n Print server and queue IDs are found in PCONSOLE. 8600003D.QDR <DIR> Q_0086 .SRV Q_0086 .SYS These two hidden files are also created. The file (FILESERV) is created in the print server directory when the print server is configured to service servers other than the one on which it is created. In the queue directory, the .SYS file is technically the queue. Each print job that is submitted has a corresponding record created in this file that allows the management of the print job by Queue Management System (QMS) and the print server. The .SRV file keeps track of print servers currently attached to the queue. Both of these files are opened and kept open when the queue is first accessed after the file server is brought up, so someone cannot view them after that. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "C0000005" Error - NetWare Requester DOCUMENT ID#: FYI.P.11178 DATE: 20JAN93 PRODUCT: NetWare Requester for OS/2 PRODUCT VERSION: v2.0 SUPERSEDES: NA SYMPTOM "C0000005" is displayed on the screen and the machine is inoperable. ISSUE/PROBLEM The above error can happen with video boards other than Western Digital (see FYI.P.8239). SOLUTION FYI.P.8239 states the following: "The file BVHSVGA.DLL contains a fix for this problem: OEM machines with Western Digital super VGA video that has 1MB of video memory crash after first boot then the C0000005 message occurs." Another option is to use a different video board than the one currently installed. Several video boards besides the Western Digital produce this error. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "C0000005" Error - NetWare Requester DOCUMENT ID#: FYI.P.11231 DATE: 21JAN93 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM When anyone logged in and printed, the name on the banner page was always the same (System Administrator's username). ISSUE/PROBLEM The users did all their printing through Windows and set up their captures through Windows. Everything that was printed came out with the same name. The user's were not using any print job configurations. SOLUTION Novell instructed the user to leave Windows, issue an Endcap, then reissue a basic capture. Now, the banner name was always printed as %LOGIN_NAME. They also found some DOS set variables in the user's system login script that had the incorrect syntax. When the user removed the incorrect variables, the problems disappeared. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Insufficient Memory Available" - DR DOS, SuperStor DOCUMENT ID#: FYI.P.11241 DATE: 22JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM While booting, the following error message was reported: ┌─────────────────────────────┐ │Insufficient Memory Available│ └─────────────────────────────┘ ISSUE/PROBLEM The error was displayed because of a problem with the SUPERPCK line in the AUTOEXEC.BAT. The user had added the following commands: SUPERPCK /em /s:100 /mn /&u+ The /&u+ was the cause of the problem. When you specify /&u+ and the system cannot find enough upper memory available, it will return an error message instead of trying to go into Conventional memory. SOLUTION After the user changed the command to the following, the problem disappeared: SUPERPCK /em /s:1024 FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Unsynchronized Mirrored Drive Causing Slow Performance DOCUMENT ID#: FYI.P.11237 DATE: 22JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user's file server's performance slowed down considerably within the last week. The volume was also taking longer to mount than usual. MONITOR showed "Current disk request = 2" even when no one was logged on to the server. Utilization climbed to 40 through 100 percent when accessing Disk Information in INSTALL. ISSUE/PROBLEM While looking at Mirroring in INSTALL, the user discovered the mirrored partition was unsynchronized. Apparently, the operating system was continually trying to resynchronize the drive, which caused the slowing and current disk request to show as 2. SOLUTION After the user deleted the unsynchronized drive, all the symptoms disappeared. The Current Disk Request fell to 0 and utilization fell to normal. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "DOS Partition Relative Sector..." - Adaptec 1542 DOCUMENT ID#: FYI.P.11236 DATE: 22JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Whenever the user tried to access the partition table in INSTALL, the system displayed the following error: ┌───────────────────────────────────────────────┐ │DOS partition relative sector xx and size xxxx │ │do not match the ending cylinder x, sector xx │ └───────────────────────────────────────────────┘ ISSUE/PROBLEM Configuration: ■Adaptec 1542b HBA ■Micropolis 1924 2.1GB drive The 1542 does a DOS translation that would conflict with what NetWare was seeing, thus the error message. SOLUTION When the user removed Jumper J6 (pin 2) that disables this translation (not needed for the small DOS partition on the drive), the error disappeared. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Deleting a Volume in NetWare v2.2 DOCUMENT ID#: FYI.P.11228 DATE: 21JAN93 PRODUCT: NetWare PRODUCT VERSION: v2.2 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM Configuration: ■Four volumes, two volumes on each of the two hard drives The user cannot mount the second volume on their second drive, and VREPAIR does not repair the volume so that they can mount it. They user must remove the volume. SOLUTION In the File Server Definition screen, you can change volume sizes and rename volumes. However, if you change the volume size of any volume, all the data on that drive is lost but does not delete the volume. Hopefully, you have a backup of all volumes on that drive; because to create the same size volume on that drive, you must remove the volumes on the drive (with something such as ZTEST or low-level format) and recreate the all the volumes on that drive in INSTALL again. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Core Dump Is Not a NetWare v2.2 Feature DOCUMENT ID#: FYI.P.11219 DATE: 21JAN93 PRODUCT: NetWare PRODUCT VERSION: v2.2 SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM Is there a way to do a core dump in the NetWare v2.2 operating system? SOLUTION No. This feature was not built into the operating system until NetWare v3.0. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Adding More Segments on a Drive - NetWare v3.11 DOCUMENT ID#: FYI.P.11217 DATE: 21JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM All the segments on disk drive were used. ISSUE/PROBLEM In the NetWare v3.11 operating system, the disk drive can only have up to eight segments on the drive, and the limitation for the volume is 32, spanning multiple drives. So if you have eight segments on four different drives, you have 32 total segments on one spanned volume under NetWare. The question that the user had was the following: "I have one drive that I set up with three volumes and left space unused at the end of the drive. I have used that area at the end of the drive to add to the three volumes as they needed more space. My problem now is that I am out of segments (used eight segments on the drive). If I delete the VOL2: volume, which had three segments, then recreate it as one segment of the same size, does that free up the other two segments so I can add more space to the SYS: volume?" SOLUTION Yes. The user would have six segments used on the drive and would be able to add the other two segments to any volume. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Servers Not Mirrored Correctly Causing Error DOCUMENT ID#: FYI.P.11252 DATE: 25JAN93 PRODUCT: NetWare SFT Level III PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM "Drive deactivated due to drive failure" ISSUE/PROBLEM The user had installed SFT III. When the user shut down the primary server, the secondary server displayed the following message: ┌────────────────────────────────────────────────────── ────────┐ │Drive deactivated due to drive failure. Sys volume not mounted│ └────────────────────────────────────────────────────── ────────┘ SOLUTION By looking at the mirrored server status in the I/O engine, the user found that during INSTALL the servers had not mirrored correctly or completely. After getting the servers mirrored correctly, the error disappeared. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Surface Test Failing on Adapted 1740 Drive DOCUMENT ID#: FYI.P.11251 DATE: 26JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Surface Test failing using an Adaptec 1740 in Enhanced mode. ISSUE/PROBLEM Configuration: ■ALR Pro VEISA 486/66 ■an Adaptec 1740A ■Fujitsu 1.7GB drive The user was running the optional Surface Test from the NetWare v3.11 INSTALL.NLM as a stress test on the drive channel. It would go through about 2000 blocks, then fail. He tried three other 1740 controllers with the same results; however, other controllers (such as the Bus Logic 32-bit EISA controller) worked fine. The user also tried the 1740 controllers in other file servers and had the same problem. Novell was not able to recreate the problem. Although, the user was not running this utility to low-level format the drive, he was concerned that a failure of this test indicated that the utility would fail when other applications made similar calls. SOLUTION According to Novell, the Surface Test is not a good indication of how the NetWare v3.11 applications will run. In NetWare v3.0, Novell required all disk driver writers to include the Application Program Interface (API) calls that would support the surface test. However, as the SCSI technology evolved, this requirement was dropped and the manufacturers had to write their own utilities for formatting the drive. Therefore, the Surface Test may be making API calls that are not included in the disk driver. However, an application that was written with NetWare v3.11 API calls will not be making these calls and consequently will not fail as the Surface test. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "RPLODI Already Loaded" - TOKENBM.COM DOCUMENT ID#: FYI.P.11258 DATE: 25JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM "RPLODI already loaded" and "Error reading boot image file" after TOKENBM executed. ISSUE/PROBLEM When trying to remote boot with a Token-Ring 16/4 adapter 2 board and DOS 5.0, the system displayed the following messages: ┌─────────────────────┐ │RPLODI already loaded│ └─────────────────────┘ ┌─────────────────────────────┐ │Error reading boot image file│ └─────────────────────────────┘ Remote boot worked fine when the user did not use DOS 5.0. The user tried using RPLFIX with DOS 5; however, that did not fix the problem. The user was also using the latest TOKENBM.COM from the IBM BBS. SOLUTION Using RPLODI dated 03-21-91 (in DOSUP6.ZIP available on NetWire or NetWare Express) instead of 02-1-91 took care of the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: SSTOR Reserving Drive F: - User Unable to Log into Network DOCUMENT ID#: FYI.P.11271 DATE: 26JAN93 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM The user had two machines on the network that were running DR DOS and were compressed using SSTOR. If the user ran the SSTOR driver in CONFIG.SYS, he could not get the F: drive so he could log in to the network. Disabling the SSTOR driver allowed him to log in normally. ISSUE/PROBLEM The user had C:, D:, and E: partitions on the hard drive with the E: partition completely compressed. LASTDRIVE was set to E. NetWare takes the next drive letter after LASTDRIVE and sets it up as the LOGIN directory. SSTOR was reserving drive F: as a SSTOR reserved drive and was apparently conflicting with NetWare by trying to set up F: for the user to log into the network. SOLUTION After the user changed the LASTDRIVE to the F: partition and rebooted, he could log into NetWare on the G: drive. Making the appropriate changes in login scripts should get the system running fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: DR DOS Reads DCONFIG.SYS Before CONFIG.SYS File DOCUMENT ID#: FYI.P.11269 DATE: 26JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM The user had installed a CD-ROM drive; however, the driver MSCDEX.EXE was not loading. The system displayed an error saying that the drive was not initialized. The user verified that MTMCDE.SYS (the CONFIG.SYS driver for the CD-ROM) was in the CONFIG.SYS file and the syntax was correct. No confirmation message appeared as it was loading. The user tried adding a ? before the driver line and rebooted; however, no prompt to load the driver appeared. ISSUE/PROBLEM At one time when the user had been using SSTOR, he edited the DCONFIG.SYS file to include all the regular CONFIG.SYS commands and removed the chain statement to C:\CONFIG.SYS. Since then, the user had removed the SSTOR drive; however the DCONFIG.SYS file still remained in the root directory without being deleted. DR DOS will always read the DCONFIG.SYS file before the CONFIG.SYS file, so the DCONFIG.SYS was being loaded, and the CONFIG.SYS file was ignored. SOLUTION When the user deleted the DCONFIG.SYS, everything worked as expected. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "NWD0115 Error..." - Requester, Token Ring, ROUTE.SYS DOCUMENT ID#: FYI.P.11263 DATE: 26JAN93 PRODUCT: NetWare Requester for OS/2 PRODUCT VERSION: v2.0 SUPERSEDES: NA SYMPTOM "NWD0115 error getting connection ID (0x880f)" ISSUE/PROBLEM When the drivers for the NetWare requester for OS/2 v2.0 were being loaded, the following error message was displayed on the screen: ┌────────────────────────────────────────────┐ │NWD0115 error getting connection ID (0x880f)│ └────────────────────────────────────────────┘ The user was trying to connect to a NetWare v2.15a file server on the Token-Ring. At the NetWare v2.15a file-server console, GET NEAREST SERVER was not displayed when the NetWare drivers were being loaded at the work station. The preferred server was specified in the NET.CFG, and the interrupt line was changed from 2 to 3 at the workstation. SOLUTION There was only one Token-Ring, so removing the device driver ROUTE.SYS from the CONFIG.SYS solved the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Access Denied" Copying Files to Floppy Drive - DR DOS DOCUMENT ID#: FYI.P.11261 DATE: 26JAN93 PRODUCT: DR DOS PRODUCT VERSION: v6.0 SUPERSEDES: NA SYMPTOM When copying files from the hard drive to the floppy drive, the system displayed the following message: ┌─────────────┐ │Access Denied│ └─────────────┘ ISSUE/PROBLEM In the DR DOS(R) User Guide under "Error Messages" suggests that a password or read-only files attributes could cause this error. The user did not find anything when he checked for the passwords or read-only attributes on the files. After the user tried copying to both floppy drives from the hard drive (d:\npp\*.* directory), thinking that it could be hardware, both drives showed the same error. The user took it another step further and copied files to the floppy drives from the DR DOS directory, and this time no error reported. SOLUTION When copying files to a floppy-diskette root directory, the maximum number of files cannot exceed 128 in the root. The subdirectory the user was trying to copy had above this number. The user must make a subdirectory and copy the files to that subdirectory. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: SYSCON Not Showing Trustee Rights for Users or Groups DOCUMENT ID#: FYI.P.9369 DATE: 04JUN91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM SYSCON v3.01 not showing trustee rights for groups or users ISSUE/PROBLEM A Supervisor was experimenting with the volumes on his network. He did things such as renamed the volume, deleted the volume, and then looked to see what happened to the user's trustee rights. Suddenly, the rights started disappearing. He could add rights for a directory to a user in SYSCON, go to another user; then when he went back to the first user, the rights would be gone in SYSCON. After investigation, the Supervisor found that the user did have the rights that were granted to him for that directory; however, he could not see them in SYSCON. SOLUTION The user changed to using SYSCON v3.63 from NetWare v3.11 software. Using this version of SYSCON solved the problem of seeing the rights for both users and groups that stopped showing up for some unexplained reason. Note: Remember when checking effective rights that the only the RIGHTS utility shows effective rights while SYSCON and TLIST only show specific instances where a user or group was granted some rights and not the sum of all rights the user may have from group membership. Therefore, even though SYSCON was used to remove trustee rights from a particular user, that user may inherit rights to that directory from a group and still access that directory. SYSCON.EXE v3.66 is available on NetWire and should be used. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Insufficient Memory" During Install of NetWare v3.1 DOCUMENT ID#: FYI.P.9370 DATE: 04JUN91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM When executing INSTALL, the system displayed the following message: ┌───────────────────┐ │Insufficient Memory│ └───────────────────┘ ISSUE/PROBLEM Configuration: ■15MB of RAM installed on a Gateway 2000 386 machine The machine came set up with a HIMEM.SYS driver being called in the CONFIG.SYS file. SOLUTION When the user removed the HIMEM.SYS driver from the CONFIG.SYS file, the system worked fine. NetWare needs to be able to control all the memory. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Error "Fatal - Access to DIR Denied" - NetWare v2.15c DOCUMENT ID#: FYI.P.9371 DATE: 03JUN91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM "Fatal - Access to DIR denied." ISSUE/PROBLEM When the user tried to save a file in Lotus or Accpac, the system displayed the following message: ┌─────────────────────────────┐ │Fatal - Access to DIR denied.│ └─────────────────────────────┘ SOLUTION Increasing the number of directory entries fixed the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Running Nondedicated Router DOCUMENT ID#: FYI.P.9372 DATE: 03JUN91 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM When the user ran the router dedicated, it worked fine. When the user ran the router nondedicated and logged in to the network from it, the other workstation's packets could not cross the router. ISSUE/PROBLEM Configuration: ■Using Western Digital Ethernet Plus boards in an external router When running the router nondedicated, the user was loading IPX and NET3 after loading the router. Although the system did not display any errors, this was causing the problem because the router program already loads the IPX that is needed. SOLUTION When running routers nondedicated, the user only needs to load NET3 and not IPX. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Plugging Workstation into UPS Causes Beaconing Errors DOCUMENT ID#: FYI.P.9367 DATE: 31MAY91 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM Adding a new workstation to the fouth ring caused beaconing errors. When the user attached the workstation to one of rings other than the fourth ring, the errors disappeared. ISSUE/PROBLEM Configuration: ■The user had a large multi-ring Token-Ring network using IBM Token-Ring 4MB network boards. The cause of the error was a faulty UPS. SOLUTION The workstation was attached to a UPS when it was on fourth ring. Plugging the workstation directly into a wall outlet resolved the issue. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Errors 251, 255, and 257 in NetWare v3.1 DOCUMENT ID#: FYI.P.9368 DATE: 31MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM When in SYSCON, the system displayed ERROR 251 or ERROR 255. When in PCONSOLE, the system displayed ERROR 257. ISSUE/PROBLEM These errors typically indicate bindery corruption. Novell instructed the user to run BINDFIX. When he tried this, the system displayed the following message: ┌───────────────────────────────────────┐ │Unable to open the bindery NET$BIND.SYS│ └───────────────────────────────────────┘ Novell said this error indicated severe bindery corruption. When the user tried BINDREST, the system displayed the following message: ┌─────────────────────────────────────────┐ │Unable to set file NET$BIND.SYS to normal│ └─────────────────────────────────────────┘ SOLUTION Novell then instructed the user to do a backup, recreate the SYS: volume (SYS was the above volume), and reload. Doing this procedure solved the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Disk Command at Server Console DOCUMENT ID#: FYI.P.9364 DATE: 24MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15x SUPERSEDES: NA SYMPTOM NA ISSUE/PROBLEM When running the console command DISK in a NetWare v2.15 operating system, there is a series of rows and columns of zeros under the "PHYSICAL DISK STATUS AND STATISTICS" title. There are about 5 columns and 11 rows of zeros. The user would like to know what these numbers represent. SOLUTION When Novell tried to reproduce this in NetWare v2.2, these numbers did not come up; however, they did show up in the NetWare v2.15 software. These numbers are a series of debug statements that Novell engineering did not remove. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: MAS90 Accounting Application Problems and WD EtherCard Plus DOCUMENT ID#: FYI.P.9365 DATE: 24MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM MAS90 Accounting Application Problems and WD EtherCard Plus ISSUE/PROBLEM Configuration: Software: ■NetWare SFT v2.15c - dedicated Hardware: ■Dell 486/25 ■Hard Drives: Two 640MB Micropolis 1568-15 mirrored ■UltraStor 12f controller ■8MB file server memory ■Western Digital EtherCard Plus - 16-bit in file server, 8-bit in workstation When printing from an accounting application called MAS90 from State of the Arts, the system displayed the following message: ┌───────────────────────────────────┐ │Abend: NMI (Non Maskable Interrupt)│ └───────────────────────────────────┘ If the user ran a report from within this application, the workstation would hang and they would receive the NMI error message. Also, sometimes when using the same application, the workstation received the following message: ┌─────────────────────────────────────────┐ │Abend: GPI (General Protection Interrupt)│ └─────────────────────────────────────────┘ SOLUTION When the user ran a Dell diagnostic, it showed no errors with memory or the motherboard. When the user tried a new controller from UltraStor, the system had the same errors. When the user replaced the board from Western Digital with a new board, the abend message disappeared. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: VOLINFO, CHKVOL, and NDIR Show Running Out of Disk Space DOCUMENT ID#: FYI.P.9366 DATE: 22MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10a SUPERSEDES: NA SYMPTOM The backed up data size did not match the data size on the volume. ISSUE/PROBLEM Configuration: ■NetWare v3.10a running on a file server When the user used VOLINFO, CHKVOL, or NDIR, the system showed that it was running out of space on the hard disk. The tape backup show at least 100MB more space was on the hard disk than VOLINFO indicated. SOLUTION The user runs an application that creates many small (100 byte) files. The user also indicated that there was about 13,000 of these small files on the server. NetWare 386 defaults to a 4KB block for the hard disks, which means that each one of the small files was actually taking up 4KB of disk space. Therefore, VOLINFO and the other Novell utilities are reporting the correct information. Some third-party backup systems report bytes allocated that agree with Novell utilities. Other backup systems report only the bytes backed up (actual file size), and they do not agree with the Novell utilities. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Setting Up Bernoulli Box As NetWare Drive DOCUMENT ID#: FYI.P.9362 DATE: 20MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM Problems running NETGEN ISSUE/PROBLEM The user tried to set up Bernoulli box as a NetWare drive on an Advanced NetWare 2.15c operating system. After linking the operating system in NETGEN, each time the user pressed <Enter> on the "NetWare Installation" option to check the hardware, the system went out to the A: drive. SOLUTION The user had to change the jumper setting on the Bernoulli box to from DMA to PIO even though the Bernoulli manual states that jumper must be set to DMA for the box to work on a server. The user also had to put a cartridge in the Bernoulli box even though the manual states not to do this when you run NETGEN. After making these two changes, the user was able to successfully complete the operating system generation. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "PRINTDEF Database Is Currently Being Modified" - NetWare v2.15c DOCUMENT ID#: FYI.P.9363 DATE: 20MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM The user downloaded PRINTDEF from NetWire and copied the .EXE into the PUBLIC directory, replacing the old one, then deleted NET$PRN.DAT. After the user tried to go into PRINTDEF to create a new definition, the system displayed the following error: ┌────────────────────────────────────────────────────── ──────┐ │PRINTDEF database is currently being modified, this PRINTDEF│ │session will be limited to previous definition of PRINTDEF │ │information. Rerun PRINTDEF. │ └────────────────────────────────────────────────────── ──────┘ ISSUE/PROBLEM None of the 10 workstations that the user checked to see if he had loaded PRINTDEF on had the file open. SOLUTION After taking down the server and rebooting it, the user was able to access PRINTDEF. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Problems Printing Graphics Files with HP LaserJet III DOCUMENT ID#: FYI.P.9355 DATE: 18MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15 SUPERSEDES: NA SYMPTOM HP LaserJet III printer worked just fine with small files. When the user tried printing a graphics file, its status light just sat there blinking and finally went back to normal status. ISSUE/PROBLEM The printer only had 1MB of memory, which is not enough memory to print graphic files. SOLUTION After the user put in 2MB more of memory, printing graphic files worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Menu Program Will Not Execute DOCUMENT ID#: FYI.P.9356 DATE: 17MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15a SUPERSEDES: NA SYMPTOM The user had some menus that had been working fine on all stations. One station began having trouble with MENU not executing. When the user made a selection from the menu, it did not execute the commands and the menu reappeared. ISSUE/PROBLEM Configuration: ■Using MENU v1.22 The workstation with the problem had MS-DOS 3.3 on it, and a technician on site had changed it to IBM 3.3 to match other stations. The symptom began after the technician reformatted the hard disk and changed the COMSPEC appropriately. The user discovered a "Shell = c:\command.com" command in the CONFIG.SYS file. SOLUTION After the user deleted the shell command, everything worked as expected. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: LOGIN Reports "Network Error Abort, Retry" Error DOCUMENT ID#: FYI.P.9360 DATE: 17MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM The user had a local ring and a remote ring tied together by IBM bridges over a 56KB line. They could run IPX and NET3 and get attached; however, when the user ran LOGIN, the system displayed the following message: ┌──────────────────────────┐ │network error abort, retry│ └──────────────────────────┘ ISSUE/PROBLEM Configuration: ■TOKENB.ZIP drivers v2.62 ■IPX v3.02a ■NET3 v3.01e ■Running source routing IBM Token-Ring states that you can only run 1KB packets across a 56KB line. SOLUTION After the user did an IPX o,tbz = 1024 command, the system came up. The system was slow, but it came up. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Insufficient Memory Errors Sunday Virus Causing Memory Errors in Novell MENU DOCUMENT ID#: FYI.P.9359 DATE: 17MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.1x SUPERSEDES: NA SYMPTOM The system had been operating fine for some time. Suddenly, when executing Novell's MENU, the system displayed memory errors. The system also displayed the errors when executing other .EXE and .COM files. ISSUE/PROBLEM The user found the Sunday virus. The characteristics of this virus are similar to the ones from the Jerusalem B virus. SOLUTION After removing the virus from the system, everything worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Routing Buffers Max at 65,000 - ELS NetWare v2.15c DOCUMENT ID#: FYI.P.9357 DATE: 17MAY91 PRODUCT: ELS NetWare Level II PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM Routing buffers increase until they exeeded the maximum limit at 65,000. At that point, the server would locked and workstations lost connection. ISSUE/PROBLEM Configuration: ■AT 386/25 clone ■4MB of RAM, ■Maxtor 200MB drive ■CNET 16-bit network board This sysmptom has generally been related to network boards or drivers, but the CNET is not certified. SOLUTION When the user tried a Tiara network board, everything worked just fine. The user also tried some other boards that also worked. On a hunch, the user put the CNET board back in; and instead of using their TURBO driver v1.2, when he used the RXNET driver, it worked great. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Error Receiving on Network" - NetWare v3.1 DOCUMENT ID#: FYI.P.9358 DATE: 17MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM Soon after the user upgraded from NetWare v2.15 to a NetWare v3.1 operating system, the system displayed the following error: ┌──────────────────────────┐ │Error receiving on network│ └──────────────────────────┘ ISSUE/PROBLEM Configuration: ■IBM PS/2 Model 80 file server, ■One MAU (Multiple Access Unit) ■One ring ■Three workstations The user was using the new 16/4 boards with old drivers. SOLUTION After the user upgraded the Token-Ring driver to the latest driver from TOKENB.ZIP, the error disappeared. TOKENB.ZIP is available on NetWire or NetWare Express. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Server Hangs with No Errors - LANSpool VAP DOCUMENT ID#: FYI.P.9354 DATE: 16MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM Server hangs or sends GPIs ISSUE/PROBLEM When the user ran the LANSpool VAP, the server either hang with no error or with a GPI error. The server had this problem about twice a day. SOLUTION Upgrade to LANSpool v2.3 dated 04-23-91. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Workstation Losing Server Connection - Thomas Conrad 4045 Bus Master Board DOCUMENT ID#: FYI.P.9353 DATE: 16MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM After logging into the the NetWare v3.10 file server from an IBM PS/2 55SX workstation and being logged into the network for a short time (10 minutes), the workstation lost its connect to the server. At other times, the same workstation would not be able to log in to the file server. ISSUE/PROBLEM Thomas Conrad said that the board somehow is losing its universally assigned address after the board has been in use for a short time. SOLUTION The user shipped his board back to the manufacturer for a replacement. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Zenith Z125AA Line Printer in Advanced NetWare v2.15c DOCUMENT ID#: FYI.P.9347 DATE: 15MAY91 PRODUCT: NetWare PRODUCT VERSION: 2.15c SUPERSEDES: NA SYMPTOM Printing corruption ISSUE/PROBLEM Configuration: ■Set Zenith Z125AA line printer up for 9600 baud, ■7 bit ■Even parity ■One stop bit ■XON/XOFF set to OFF When printing through the network using core printing, the printer printed garbage. It worked fine using the printer from DOS with those same settings. SOLUTION When the user changed the printer settings to 8-bit, one stop bit and no parity, it worked just fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Printing Graphics with PageMaker in NetWare v3.10 DOCUMENT ID#: FYI.P.9348 DATE: 15MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM Distorted print jobs ISSUE/PROBLEM When printing graphics documents from PageMaker, the printing came out distorted when printing on the network printer. SOLUTION After setting up a print job in PRINTCON for the HP laserjet in use, everything worked fine through the network printer. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Volume Full Errors Saving Files - NetWare v3.1 DOCUMENT ID#: FYI.P.9349 DATE: 15MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM After installing a new driver for the IDA controller (v2.0) when the user tried to save files, the system kept displaying volume full errors. VOLINFO showed the disk space was low. ISSUE/PROBLEM Configuration: ■Compaq SystemPro with Intelligent Disk Array Controller ■Six drives with total of 1.2GB, 12MB of RAM The new driver (v2.0) must be used with the v0.96 IDA ROMs. The user had an older version of IDA ROMs. SOLUTION After the user installed the v0.96 ROMs, the full disk space came back. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Error Offending Segment" - NetWare v2.15c DOCUMENT ID#: FYI.P.9346 DATE: 15MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM While trying to access diskettes in NETGEN, the system displayed the following error: ┌───────────────────────┐ │Error Offending Segment│ └───────────────────────┘ ISSUE/PROBLEM The user had an AST 386/33 with a built in floppy controller. The user had also added a second ESDI Ompti controller which also had a floppy disk controller enabled on it. SOLUTION After the user disabled the cache and floppy address on the Ompti controller, everything ran fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Mixing ESDI and SCSI Controllers - NetWare v2.15c DOCUMENT ID#: FYI.P.9342 DATE: 14MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: SYMPTOM The user was able to get the SCSI host adapter or ESDI controller and drives to work separately but not together. The system displayed the following error when running COMPSURF: ┌───────────────────────────────────────────────┐ │Abend: Error reading disk parameter information│ └───────────────────────────────────────────────┘ ISSUE/PROBLEM The SCSI configuration was working fine; however the user wanted to add the ESDI on channel 0. SOLUTION The user was able to get the system to work by going into the REFERENCE diskette and doing the following: 1. Turned off the BIOS for the SCSI and by changing the SYS: volume name on the SCSI to something else (wanted SYS on the ESDI). 2. Changed the SCSI to a nonbootable partition. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Windows 3.0 Only Runs in Real Mode DOCUMENT ID#: FYI.P.9341 DATE: 14MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Every time Windows v3.0 was run from the network, it would come up in real mode. WIN /s and WIN /3 would start Windows then pop back to the DOS prompt. When the user installed Windows v3.0 on the local drive, it worked in the real, standard, and enhanced modes. ISSUE/PROBLEM The user discovered that if he did not install the mouse, then the standard and enhanced modes worked fine on the network. Microsoft said the MOUSE.SYS dated 04-01-90, size 4896, was the problem and that the user needed a MOUSE.SYS dated after 05-01-90. SOLUTION After the user installed MOUSE.SYS dated after 05-01-90, the system worked as expected. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "1.1.120 User... at Station... Terminated..." - NetWare v3.11 DOCUMENT ID#: FYI.P.11286 DATE: 27JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM "1.1.120 User USERNAME at station xxxx connection terminated User did not log out within 5 minutes after security watchdog notice" ISSUE/PROBLEM Even though the user had used set parameter to turn off watchdog notices to the console, the system displayed the following message at the console: ┌────────────────────────────────────────────────────── ──────────┐ │1.1.120 User USERNAME at station xxxx connection terminated User│ │did not log out within 5 minutes after security watchdog notice │ └────────────────────────────────────────────────────── ──────────┘ SOLUTION This is not watchdog but security that is terminating the connection. Therefore, the set parameter has no effect. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Missing or Invalid Value for CHECKSUM..." - NET.CFT Parameter DOCUMENT ID#: FYI.P.11281 DATE: 27JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM The user's system displayed the following message: ┌────────────────────────────────────────────────────── ─────────┐ │SHELL-331-58: Missing or invalid value for CHECKSUM parameter │ │was specified on line XX of configuration file. Entry ignored.│ └────────────────────────────────────────────────────── ─────────┘ ISSUE/PROBLEM The NETX v3.31 of the shell no longer supports the following parameter in the NET.CFG: checksum yes The following is an example of how it worked with all versions on NETX before v3.31: LINK DRIVER IPTUNNEL checksum yes SOLUTION The UDP checksum is now enabled by default and the above parameter is no longer necessary. Leave it out of the NET.CFG. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Unable to Span Partitions to Make NetWare See One Partition DOCUMENT ID#: FYI.P.11173 DATE: 18JAN93 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM Unable to duplex drives. ISSUE/PROBLEM A user had been told by technical support that he could take one 1GB drive with SYS: and VOL1: volumes and duplex it with two 650MB drives off another controller; however, he was unable to duplex the drives in this manner. SOLUTION There is some confusion about what can be done with two drives. If you are dealing with volumes, then two drives can be spanned by adding the segment on the second drive to the volume on the first. However, with mirroring and duplexing, you are dealing with physical partitions on the drives. Therefore, there is no way to span physical partitions to make NetWare recognize it as one partition so that the two drives will appear as one for mirroring purposes. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: ARCserve Not Finding IBM$RUN.OVL DOCUMENT ID#: FYI.P.10887 DATE: 17DEC92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: NA SYMPTOM When running the EMSNETX shell with Dell's DOS 5.0 and trying to run ARCSERV.NLM from this machine, ARCSERV.NLM returned with an error that it could not find IBM$RUN.OVL. ISSUE/PROBLEM The user checked the short machine type in the NET.CFG file. It had not been set; therefore, it should be returning IBM. To verify this, the user did a write statement in the login script. Sure enough, the %SMachine returned a blank. SOLUTION As a test, the user assigned the %SMachine type in the NET.CFG file; and everything worked. ARCSERV.NLM also worked by removing EMSNETX and running just NETx. Novell was unable to recreate this problem using MS-DOS. It may be a problem with Dell's DOS. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Environment Variables Outside Login Script Does Not Work with MAP DOCUMENT ID#: FYI.P.10855 DATE: 15DEC92 PRODUCT: NetWare PRODUCT VERSION: v2.2, v3.11 SUPERSEDES: NA SYMPTOM Environment variables not recognized outside the login script when using MAP command. ISSUE/PROBLEM One day a user set an environment variable to the following: set TERMTYPE=Whatever Then when the user used the MAP command with this following syntax, it mapped the drive correctly to the WHATEVER subdirectory: map s16:=sys:\%TERMTYPE% The next day when he issued the MAP command, the system displayed the following message: ┌─────────────────────────────────┐ │Directory ****** is not locatable│ └─────────────────────────────────┘ SOLUTION You can not map the drive from the command line. However, you can map the drive in a batch file or login script as shown. Batch File MAKEMAP.BAT MAP S16:=SYS:%TERMTYPE% Login Script MAP S16:=SYS:%<TERMTYPE> FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Workstation Loses Connection - SMC PC-270E Boards DOCUMENT ID#: FYI.P.09058 DATE: 20MAR92 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: FYI.P.5870 SYMPTOM Workstation losing connection ISSUE/PROBLEM Configuration: SMC PC-270e boards connected in linear bus topology The fourth workstation that tried to attach to file server was often unable to do so. If it was successful at logging in, it lost connection. When a fourth workstation tried to log into the file server, sometimes it would log in and sometimes it would not. If the workstation was able to log in, it lost connection. SOLUTION When the user added an active hub, all stations were able to log into the server and stay connected. Note: SMC PC270e boards should be able to have up to 10 stations connected in linear bus. Check the basics. Make sure certified cabling is being used, and verify terminators are 100ohms. Make sure node addresses are unique for all workstations and that there are sufficient network connections on the server. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Running RCONSOLE with No Password Required DOCUMENT ID#: FYI.P.09060 DATE: 18SEP92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: FYI.P.08854 SYMPTOM NA ISSUE/PROBLEM The supervisor does not want a password to be required when users run RCONSOLE at the workstations. Normally, the password for RCONSOLE is set in the syntax to load REMOTE.NLM. If no password is defined in the syntax, the NLM will not automatically load and the server console halts and prompts the supervisor to enter a password. What is the syntax to allow REMOTE.NLM to automatically load and accept a null password? SOLUTION Solution 1: The following configurations will allow the REMOTE LOAD without a password: LOAD REMOTE < C:\password.fil Create a "password.fil" with text that has only a carriage return in it and place it in the c:\ directory. Solution 2: Put the following in the syntax in the AUTOEXEC.NCF file: load remote ^ FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Garbage Printing to RPRINTER Worstation DOCUMENT ID#: FYI.P.09061 DATE: 06APR92 PRODUCT: NetWare PRODUCT VERSION: v2.2, v3.11 SUPERSEDES: FYI.P.07058 SYMPTOM Printing garbage ISSUE/PROBLEM Configuration: ■PS/2 model 50 ■HP III printer Garbage was printed to an RPRINTER workstation. SOLUTION The user implemented RPRINTER 1.22r in PSERV.ZIP that fixes this problem by using the Tx (x is 1-9) parameter. As a reminder, the new RPRINTER in PSERV.ZIP has a new parameter whose syntax is RPRINTER <PSERVER_NAME> <PRINTER_NUMBER> T<x> (x is 1-9). This parameter adjusts the strobe time on the port. Setting this parameter to 1, 2, 3, or 4 will generally fix any timing problems between the printer and the computer. Remember, the higher you go (5, 6, 7,...), the slower you will print. Basic Information on Strobe It is the strobe signal that informs the printer that there is valid data present on the data lines and to "pull" it in, or accept the data into its data buffer. A parallel printer port has three basic registers (data, status and control). Strobing is a mechanism for flow control between the printer and the parallel port. When a byte of data is sent to the port, the control register is strobed on and then off. The strobe bit must be set for a minimum of 0.5 microseconds to pulse the data into the printer (time may vary on ports, thus the setable parameter). Data must be present in the printer data register for a minimum of 0.5 microseconds before and after this stobe is set. Because of the differences in ports and machines, the strobe time in the interrupt service routine can be adjusted to speed up or slow down the time between strobe on and stobe off so the data byte can successfully be received by the printer. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "This Program Cannot Be Run in DOS Mode" DOCUMENT ID#: FYI.P.09063 DATE: 19MAR92 PRODUCT: NetWare PRODUCT VERSION: v2.15c, v2.2 SUPERSEDES: FYI.P.05838 SYMPTOM The user added a NetWare v2.2 server to an existing NetWare v2.15c network. After copying the v2.2 NetWare utilities to the NetWare v2.15c server's SYSTEM and PUBLIC directories, the workstations displayed the following message each time SYSCON, FILER, or another NetWare utility command was issued: ┌──────────────────────────────────────┐ │This program cannot be run in DOS mode│ └──────────────────────────────────────┘ ISSUE/PROBLEM Because of the file sizes, Novell determined that the user had accidentally loaded the OS2 UTILITIES diskettes from a NetWare v3.11 set. SOLUTION After loading the v2.2 DOS utilities, everything worked. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Abend: Invalid Serial Number" - NetWare v3.11 Premium Package DOCUMENT ID#: FYI.P.09064 DATE: 15MAY92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: FYI.P.07543 SYMPTOM The user upgraded from NetWare v3.11 100-user version to a NetWare v3.11 250-user PREMIUM package and replaced the old SERVER.EXE file with the new. When rebooting the system, it failed with the following message: ┌────────────────────────────┐ │ABEND: Invalid serial number│ └────────────────────────────┘ ISSUE/PROBLEM The user had old patches loading through the AUTOEXEC.NCF file. The NetWare v3.11 PREMIUM package has a different serialization scheme than the earlier NetWare v3.11 package. The patches have been updated to support the new serial numbers, but the user still had the earlier patches. SOLUTION To prevent the ABEND and change the patches, load the server with an -NA parameter. If necessary, the AUTOEXEC.NCF file can be modified and the updated patches (311PTC.ZIP found on NetWire or NetWare Express) installed. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: RPRINTER Hangs in Windows with Stacker Loaded DOCUMENT ID#: FYI.P.09066 DATE: 23SEP92 PRODUCT: NetWare PRODUCT VERSION: v2.2, v3.11 SUPERSEDES: FYI.P.08903 SYMPTOM RPRINTER does not print or hangs workstation in Windows 3.1 ISSUE/PROBLEM Configuration: ■RPRINTER ■Windows 3.1 The user could print fine outside of Windows; however, whenever the user tried to print from an application inside Windows, nothing printed or the workstation hung. SOLUTION The user was loading Stacker. When Stacker was not loaded, the printing worked fine. Because using Stacker was desired, the user used the following workaround steps: 1. Load Stacker. 2. Go into Windows. 3. Go to a DOS prompt. 4. Run RPRINTER. 5. Go to Windows application. Using this procedure kept Stacker from overwriting RPRINTER in memory. If RPRINTER was loaded in the virtual DOS session and that session was left running as a background process, the user could print from other Windows applications. However, depending on the background priority, everything crawled or ran slowly because of time slicing. The hardware version of Stacker (with the coprocessor board) also requires an EMMEXCLUDE statement in the [386 Enh] section of the SYSTEM.INI. If a third-party memory manager is included, check the documentation to see if the EMMEXCLUDE is also required in the CONFIG.SYS. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Insufficient File Handles at Workstation" - NCOPY.EXE DOCUMENT ID#: FYI.P.09068 DATE: 23AUG91 PRODUCT: NetWare PRODUCT VERSION: v2.12, v3.11 SUPERSEDES: FYI.P.03883 SYMPTOM "Insufficient File Handles at Workstation" ISSUE/PROBLEM Configuration: ■NCOPY.EXE v3.56 ■MS-DOS v5.0 ■3.02a IPX ■3.22 NETX NCOPY from a NetWare v2.12 to v3.11 server would result in the following error: ┌────────────────────────────────────────┐ │Insufficient File Handles at Workstation│ └────────────────────────────────────────┘ This error occurred after several minutes of successful copying. The error was seen at the workstation and then continued with the following error for for each remaining file: ┌───────────────┐ │File not copied│ └───────────────┘ Increasing maximum file handles in SHELL.CFG did not help. SOLUTION Novell instructed the user to try earlier SHELL/IPX/DOS. Not all combinations were tried; however, the one that worked was the following: DOS 4.0, NETX v3.22, IPX 3.02a Two other workarounds for this problem are the following: ■ Flag all files shareable. ■ Make sure that all nonshareable files are closed before using NCOPY to copy them. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Unable to Delete User Home or Subdirectories - MAKEUSER v3.61 DOCUMENT ID#: FYI.P.09069 DATE: 09APR92 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: FYI.P.07095 SYMPTOM MAKEUSER will not delete the users home directory or subdirectories ISSUE/PROBLEM MAKEUSER will not delete the users home directory or subdirectories using the following commands: ■ #home_directory sys:users ■ #purge_user_directory ■ #delete user1; user2^ In some cases but not all, MAKEUSER (v3.61) will delete the home directory. In some cases but not all, MAKEUSER will delete subdirectories. In all cases, the user is deleted. SOLUTION Execute the MAKEUSER version in MKUSER.ZIP available on NetWire or NetWare Express. This version of MAKEUSER fixes this problem and deletes all appropriate directories. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Trustee Assignments Missing after Restore with NBACKUP DOCUMENT ID#: FYI.P.09071 DATE: 22JUL92 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: FYI.P.08270 SYMPTOM NA ISSUE/PROBLEM Configuration: ■Two existing NetWare v3.11 servers (FS1 and FS2) The user wanted to combine the two servers into one (FS3). FS3 was a different box than FS1 and FS2, so the user did the following to combine the servers: ■ The user did a NBACKUP of FS1, selecting "Backup File Server" in the "BACKUP" menu (backed up the data, trustee assignments, binderies - everything). ■ When the user restored this session to FS3, everything restored as expected. ■ Next, the user did a NBACKUP of FS2, excluding the bindery but including the trustee assignments and all data. When the user RESTORED the session to FS3, the data came across; however, the users no longer had their trustee assignments. SOLUTION Page 3-17 of the NetWare System Interface Technical Overview manual states that the trustee assignments are linked to the OBJECT ID (which is randomly generated on each server) and not the actual username/groupname. Because the users on FS1 and FS2 did not have consistent OBJECT IDs on both servers, the user was forced to recreate the trustee assignments for those directories and files restored from the second server (FS2). FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Possible Problems Mirroring with IDE Drives DOCUMENT ID#: FYI.P.09072 DATE: 13OCT92 PRODUCT: NetWare PRODUCT VERSION: v2.2, v3.11 SUPERSEDES: FYI.P.10088 SYMPTOM NA ISSUE/PROBLEM Use of IDE Drives in Mirrored Environments Mirroring (as opposed to duplexing) IDE drives is a fairly pointless exercise. This is because the secondary drive is highly dependent on the first drive. Normally, where two drives are operating in mirrored fashion from a single (external) disk controller, each drive is an equal and independent peer; and provided the controller board itself does not go down, the failure of either drive should not lead to the failure of the other. However, with IDE drives, the controller is integrated into the drive itself; hence, the name Integrated Drive Electronics (IDE). When you run IDE drives as a pair, the second drive's controller is disabled and it runs as a "Slave" from the first drive's built-in controller. This effectively becomes a "fault-intolerant" configuration, because a failure in the primary IDE drive has a very high possibility of bringing its internal disk controller down. This would effectively make the second drive inoperative, even though the data would still be intact in most cases (if the controller did not write some garbage on the secondary drive if it died). For fault tolerant applications, IDE drives should only be operated in duplexed configurations (Example, from entirely separate IDE interfaces on the PC bus). This would require using a secondary IDE network board, operating at a second port address, or using PCs that have dual IDE interface ports built into the system board. Because of increasing support problems resulting from design limitations of some IDE drives and Host Bus Adapters (HBA), Novell has issued the following configuration recommendation. Novell recommends that when using IDE drives in SFT II mirroring configurations, duplex the drives by placing the mirrored pair on separate channels (duplex or channel mirroring) rather then mirroring (simplex mirroring) the drives on a single channel. On some implementations of the IDE interface, when the Master drive fails, the Slave drive may also fail when both drives are attached to a single HBA. This is a limitation of some manufacturer's implementation of the IDE interface and not a limitation of NetWare itself. Users wishing to mirror a pair of drives on a single IDE HBA should confirm with the drive manufacturer that the configuration has been certified by Novell for use in mirrored environments. After extensive testing, Novell has determined that when using IDE drives, the advantages of SFT II are fully realized when duplexing the mirrored pair by using separate Host Bus Adapters for each side of the mirrored pair. Caution should also be exercised when two drives are added to a single HBA and duplex mirrored against two other drives on a separate channel. In that configuration, if one drive in a given HBA fails, the second drive on that HBA may also fail. The following table summarizes this recommendation: ┌────────────────────────────────────────────────────────────── ────┐ │Recommended │ │ │ │ Duplex Mirroring (a mirrored pair on separate Host Bus Adapters) │ │ │ │ HBA Drive A │ │ │ │ HBA Drive A' │ ╞══════════════════════════════════════════════════════════════ ════╡ │Recommended (Use Caution) │ │ │ │ Duplex Mirroring of dual drives (two mirrored pairs on │ │ separate Host Bus Adapters) │ │ │ │ HBA Drive A Drive B │ │ │ │ HBA Drive A' Drive B' │ ╞══════════════════════════════════════════════════════════════ ════╡ │Not Recommended │ │ │ │ Simplex Mirroring (a mirrored pair on a single Host Bus Adapters)│ │ │ │ HBA Drive A Drive A' │ ╞══════════════════════════════════════════════════════════════ ════╡ │Not Recommended │ │ │ │ Simplex Mirroring on Dual Host Bus Adapters (two mirrored │ │ pairs on each of two HBAs) │ │ │ │ HBA Drive A Drive A' │ │ │ │ HBA Drive B Drive B' │ │ │ └────────────────────────────────────────────────────────────── ────┘ SOLUTION NA FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Problems Compiling Programs - NetWare v2.2 DOCUMENT ID#: FYI.P.09073 DATE: 30MAR92 PRODUCT: NetWare PRODUCT VERSION: v2.2 SUPERSEDES: FYI.P.05961 SYMPTOM Error compiling with Microsoft Corp's Quick BASIC Professional Development System (PDS) 7.1 ISSUE/PROBLEM Using NetWare 2.2, there were problems compiling programs with Microsofts Corp's Quick BASIC. Some of the code in the programs they were compiling contained "Open COM1 or COM2" statements. A workstation, running NETX shell v3.22, would hang or reboot when the compiled program tried to open the COM ports. Microsofts Corp's Quick BASIC 4.5 exhibited no problems opening the COM ports and having the NETX shell loaded. SOLUTION When disabling COM3 and COM4 using shared interrupts at the work station, the problem goes away. By removing the IBM SERIAL/PARALLEL board that was in the workstation, the user was able to compile the programs with no problems. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: LOGIN /NO /SCRIPT Returns Error 18 (8801) DOCUMENT ID#: FYI.P.09076 DATE: 29JUL92 PRODUCT: NetWare PRODUCT VERSION: v2.2, v3.11 SUPERSEDES: FYI.P.08378 SYMPTOM Errors: "Warning: Unexpected error 18 (8801) Can't get connection information." ISSUE/PROBLEM When the user issued a LOGIN command with the "no attach (/no)" and "script (/script)" options using LOGIN.EXE v3.58, the system displayed the following error message: ┌────────────────────────────────────────────────────── ──┐ │Warning: Unexpected error 18 (8801) Can't get connection│ │information │ └────────────────────────────────────────────────────── ──┘ This error can be duplicated by doing the following: LOGIN /no /script c:\junk.log servername/username Where JUNK.LOG has the contents, write the following: Good Day %Login_Name! When LOGIN is run, it will immediately display the following error and write only "Good Day!" SOLUTION LOGIN.EXE v3.70 found in LOG370.ZIP works with the /no /script options and fixes the problem. LOG370.ZIP is available on NetWire or NetWare Express. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Unable to Lock Disk..." - NetWare v3.11 DOCUMENT ID#: FYI.P.09078 DATE: 11MAY92 PRODUCT: NetWare PRODUCT VERSION: v3.11 SUPERSEDES: FYI.P.07462 SYMPTOM "Unable to lock disk, may be in use by another process" ISSUE/PROBLEM Configuration: ■Compaq DeskPro 386/33 machine as the file server ■Compaq system manager (server manager) board installed Each time the user tried to delete the NetWare partition from a drive, the system displayed the following error message: ┌─────────────────────────────────────────────────────┐ │Unable to lock disk, may be in use by another process│ └─────────────────────────────────────────────────────┘ No one was logged into the network. The drive used to be a mirrored drive with the external subsystem. SOLUTION After unloading the NLM (IDAMON.NLM) used to run the system manager (server manager) board, the user was able to delete the partition. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Error Using Stacker with DR DOS 6.0 DOCUMENT ID#: FYI.P.10511 DATE: 18NOV92 PRODUCT: NetWare PRODUCT VERSION: All versions SUPERSEDES: NA SYMPTOM When the user loaded SUPERPCK, the system displayed an error message referring to heads and sectors not matching previous... ISSUE/PROBLEM Configuration: ■Using DR DOS v6.0 ■Using Stacker When SUPERPCK loads, it looks at the drive and sends this as a warning message because of the way Stacker handles the drive. SOLUTION The user must press <Enter> twice to get past this message. This is only a warning. SUPERPCK works fine with Stacker and the user can ignore the message. Use the following a workaround. If SUPERPCK is set up as Superpck /EM /S:1024, change it to read as follows: Superpck /EM /S:1024 /K- >NUL The /k- switch takes the pause out for issuing warning messages. The >nul switch makes it so the user does not see the message. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Machines Hang with No Error Messages Displayed DOCUMENT ID#: FYI.P.9345 DATE: 14MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c, v3.10 SUPERSEDES: NA SYMPTOM NE2000 DMA failures ISSUE/PROBLEM Configuration: ■Club American 386/25 and 386/30 servers. Each machine has two NE2000 network boards. All machines were hanging with no error displayed. The NetWare v3.10 server was getting NE2000 Direct Memory Access (DMA) failure errors. The test the user ran on the cable's ohms showed they were good. SOLUTION The user found several bad connections. One was so bad it pulled out of the T-connector when the user picked up the cable. Fixing the cables took care of the "hanging" problems. Another solution is the following: While working on the NetWare v3.10 server, the LPT/video board quit. Replacing it slowed down the number of NE2000 DMA failures. The user then removed the NE2000 boards and discovered that one had the W16 jumper removed and the other did not. The BIOS in the machine was from Siemens; so after the user put the jumper on, the DMA failures disappeared. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Error "Invalid Load File" - PCN2 Drivers and NetWare v3.1 DOCUMENT ID#: FYI.P.9343 DATE: 14MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.1 SUPERSEDES: NA SYMPTOM "Invalid Load File" ISSUE/PROBLEM The user was trying to use PCN2 drivers from PCN2.ZIP with a NetWare v3.1 operating system. During load from floppy or hard drive, the system displayed the following message: ┌─────────────────┐ │invalid load file│ └─────────────────┘ SOLUTION The PCN2 drivers have only been certified for use with NetWare v3.11 operating systems but not for NetWare v3.10 systems. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Serial Printer Not Printing from COM1 Port DOCUMENT ID#: FYI.P.9344 DATE: 14MAY91 PRODUCT: NetWare PRODUCT VERSION: 2.15c SUPERSEDES: NA SYMPTOM Xerox 4045 model 150 not printing from server port ISSUE/PROBLEM Configuration: ■NULL modem cable for printer Print jobs went in and out of the print queue as if printing; however, nothing printed on the printer. Printing worked fine under DOS. If the user used the same printer and cable hooked to a workstation and ran PSERVER.EXE, it printed fine. The user tried polled and interrupt mode but neither helped. SOLUTION Novell ran a test with the Xerox 4045 printer and came up with the following pin configuration that should work: PC-port Printer 1 --------- 1 2 --------- 3 3 --------- 2 5 --------- 20 7 --------- 7 When the user tried Novell's pin configuration with a new cable, it fixed the problem. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: GPIs Loading VAPs in NetWare v2.15c DOCUMENT ID#: FYI.P.9335 DATE: 13MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM The user had physically moved the server from one building to another. After the move, the system received GPI errors whenever the user answered YES to the prompt of loading VAPs. Changing the VAPs did not remove the symptom. The user had tried MAC.VAP and PSERVER.VAP, and each one returned the same GPI. ISSUE/PROBLEM The user was using Disk Manager - N(R) because of the IDE drive. It writes and uses some information on track 0. During a successful boot, the server screen will display a message that Ontrack's boot loading has been successful (or something to that effect). SOLUTION The user reinitialized the disk and regenerated the operating system so the Novell cold boot loader was being used instead of the Ontrack information. This removed the GPI problems. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Abend: General Protection Interrupt" - VAP #0 DOCUMENT ID#: FYI.P.9334 DATE: 13MAY91 PRODUCT: NetWare PRODUCT VERSION: 2.15c SUPERSEDES: NA SYMPTOM With the PSERVER VAP loaded, the server occasionally went down and display the following message: ┌───────────────────────────────────┐ │ABEND: General Protection Interrupt│ └───────────────────────────────────┘ The error occurred mostly with high server utilization. ISSUE/PROBLEM The server only had one File Service Process (FSP). They were able to bring that up to two; however, the server still went down. SOLUTION After the user ran PSERVER.EXE instead of PSERVER.VAP, the file server no longer had problems even with two FSPs. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Abend: Attempt To Configure Non-Existent Disk" DOCUMENT ID#: FYI.P.9336 DATE: 13MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM The user received the following message after installing an external drive to the Compaq 386/25 file server: ┌───────────────────────────────────────────────┐ │Abend: Attempt to configure a non-existent disk│ └───────────────────────────────────────────────┘ ISSUE/PROBLEM Configuration: ■The user was running a 320MB ESDI internal drive and had attached this external drive both running off ESDI controllers. ■The user had set the secondary drive to run at IRQ=15. After checking for things such interrupt conflicts, wrong jumper settings, and hardware conflicts, the user discovered the problem was caused from the STONED virus. After attempting to boot the file server several times the server responded that the computer was stoned. SOLUTION The first disk was the one getting the abend error because of the virus. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Slow Performance on Compaq DeskPro 386/25 DOCUMENT ID#: FYI.P.9337 DATE: 13MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10a SUPERSEDES: NA SYMPTOM The workstations logged into one of the servers was experiencing extreme slowness and would sometimes lose connections. The user tried a myriad of things involving the drivers and boards, but nothing worked. When the user tried to switch the machines around (except for the subsystem), the machine that was having the problems sped up and the other one slowed down. The problem seemed to be caused from the location of the server and not the server itself. The user ran several tests on that segment's cabling without finding the cause of the problem. The only thing he had not changed was the subsystem. ISSUE/PROBLEM Configuration: ■Two identical Compaq DeskPro 386/25 servers, each with a 650MB Anthem subsystem and Adic DCBs. Finally, while booting the suspect server, the user mentioned that the speed rating on the DeskPro was only 50. When checking the MODE SPEED for the machine on the COMPAQ USER PROGRAM diskette, the user discovered that it was set for AUTO, which should let it run at 25MHz. It was not. For some reason with that particular Anthem subsystem (the other one was identical to it), it was kicking the DeskPro 386/25 down to COMMON speed, which is 6MHz. SOLUTION After the user changed the MODE SPEED from AUTO to HIGH (25MHz), the speed jumped up to over 300 and there were no more speed problems with the workstation. From: Adayley,wvoss FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: MAP ROOT Command Problems in NetWare v3.10 - IBM PCs DOCUMENT ID#: FYI.P.9340 DATE: 13MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM The user was having various problems with the MAP ROOT command. Installing GEATWS21.ZIP from Gateway's BBS fixed the problems on the NCR 386 and IBM ATs but did not help the IBM PCs at all. ISSUE/PROBLEM Configuration: ■Gateway Ethernet board in NCR 386s, IBM ATs, and IBM PCs The problem could be with the Gateway driver. Other boards (for example an NE1000) work fine in an IBM PC when doing a MAP ROOT. It could also have something to do with speed differences between the slow PC and the NCR 386sx/16MHz file server. It is possible that the slow PCs cannot respond fast enough to the file server. SOLUTION Try putting a faster board in the PC or putting an 8-bit board in the 386 and connect all the PCs off it. Another option is to put in an external bridge between the PCs and the 386 file server to slow down the response from the file server. The user tried an SMC 3008 Ethernet board (option 0: IRQ 3,300). With this board, the PC worked fine; it also cleared up another problem of the PC locking with a DIR command. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Only Default Login Script Executes NetWare v2.15 DOCUMENT ID#: FYI.P.9339 DATE: 13MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM Users get only the default login script ISSUE/PROBLEM When users logged in, only the default login script would execute. Also, a DIR of SYS: volume's directories and subdirectories showed a limited number of files existing for the users. They could not even type "rights" because RIGHTS.EXE did not show up in the PUBLIC directory. This was not a problem for Supervisors; everything appeared normal for them. SOLUTION The group EVERYONE had become corrupted, and BINDFIX and VREPAIR would not fix it. 1. In SYSCON, rename the group EVERYONE. 2. Create the group EVERYONE again. 3. Then insert the users into the newly created group EVERYONE. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Servers Hangs or GPI in NetWare v2.15c with LANSPOOL VAP DOCUMENT ID#: FYI.P.9338 DATE: 13MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM File server hangs and gets GPIs ISSUE/PROBLEM A user has quite a few sites that are having problems with their NetWare v2.15c servers locking and sometimes sending GPIs. The users experience hangs at least twice a day. The LANSPOOL.VAP is used in these sites. SOLUTION The problem was resolved by upgrading to LANSpool v2.3. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Invalid Drive Specification" - NetWare v3.10 DOCUMENT ID#: FYI.P.9332 DATE: 10MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10 SUPERSEDES: NA SYMPTOM The user had a NetWare v3.10 server with three NE3200 boards. One of the NE3200 boards serviced a Macintosh backbone. One serviced a PC backbone, and the other NE3200 serviced a backbone that had Macintosh and PC machines on it. All the PCs started experiencing the following error when they tried to switch to the F: drive to log into the network. ┌───────────────────────────┐ │Invalid Drive Specification│ └───────────────────────────┘ If the could get to the F: drive, the system displayed the following message: ┌─────────────────────────────────────────┐ │Extended error 55: No network connection"│ └─────────────────────────────────────────┘ ISSUE/PROBLEM One of the Macintosh machines had taken the LOGIN directory (it appears as a folder on the Macintosh) and moved it into another folder, a different directory. SOLUTION When the directory was restored to its appropriate location, things started to work just fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: LOGIN.EXE v3.08 Truncates Path DOCUMENT ID#: FYI.P.9333 DATE: 10MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15a SUPERSEDES: NA SYMPTOM LOGIN truncates part of the path. ISSUE/PROBLEM LOGIN v3.08 would truncate any part of the path that pointed to a drive greater than the C: drive. SOLUTION After the user upgraded from a NetWare v2.15a to v2.15c operating system, everything worked fine. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "Phantom Files Removed = 1" DOCUMENT ID#: FYI.P.9331 DATE: 08MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.10a SUPERSEDES: NA SYMPTOM Every time the user shut down the server and brought it back up, the system displayed the following message when mounting the SYS: volume. ┌─────────────────────────┐ │Phantom files removed = 1│ └─────────────────────────┘ ISSUE/PROBLEM For some reason when the user shut down the server, he tried to write to a file that did not exist, thus creating a phantom file. SOLUTION After the user ran VREPAIR on the volume, the phantom file disappeared and the error no longer displayed. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Unable to Boot from Hard Drive on PS/2 Model 80 DOCUMENT ID#: FYI.P.9328 DATE: 06MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM After installing the operating system on a PS/2, model 80 with two ESDI drives duplexed and mirrored, the user was unable to boot from hard drive. A boot from a floppy diskette allowed the server to come up. ISSUE/PROBLEM Both the hard drives were terminated on channel 0. SOLUTION After the extra terminators were removed, the user was able to boot from hard drive. Only the last drive in the chain needs to be terminated. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: "0 Object Nodes and 0 Property Nodes Free" - Running BINDFIX DOCUMENT ID#: FYI.P.9330 DATE: 06MAY91 PRODUCT: NetWare PRODUCT VERSION: v2.15c SUPERSEDES: NA SYMPTOM The user thought the binderies were full because the system displayed the following message when running BINDFIX: ┌────────────────────────────────────────┐ │0 object nodes and 0 property nodes free│ └────────────────────────────────────────┘ ISSUE/PROBLEM The displayed error does not mean the binderies are full. SOLUTION NetWare allocates a block of disk space for the initial bindery table that holds the list of object and property nodes. They are not on a one-to-one basis. One object node can have several property nodes. When that table is full, the operating system will allocate additional space for more nodes until the set maximum is reached that was specified in NETGEN. The message means that the initial table is full. If users are deleted, then that message will change (depending on the number of deletions) to X object nodes and X property nodes free. This means that some entries in the initial table have been deallocated. NetWare will allocate more object nodes and property nodes when needed. FYI (Note: The origin of this information may be internal or external to Novell. Novell makes every effort within its means to verify this information. However, the information provided in this document is FOR YOUR INFORMATION only. Novell makes no explicit or implied claims to the validity of this information.) TITLE: Second Print Server Not Responding DOCUMENT ID#: FYI.P.9329 DATE: 06MAY91 PRODUCT: NetWare PRODUCT VERSION: v3.1 SUPERSEDES: NA SYMPTOM Second print server not responding ISSUE/PROBLEM All workstations were logging into the first file server (from the concentrator) and then logging into the second file server through the hub. When they went into PCONSOLE and looked at "status/control" for the first file server's print server everything was fine; however, when they looked at the second print server, the system received an error that the print server was not responding. SOLUTION The addresses for the boards going through the two concentrators were the same and needed to be different; the two addresses on the hub need to be the same. In most cases, you will only get router configuration errors if you have LAN boards sharing the same cable and using different addresses. They must be the same address when sharing the same cable. On a large internetwork, the router configuration errors may spread across the network to other servers or routers not even connected to the cable.