Server abends during purge of volume (toolbox purge, dos prompt purge, filer purge, or netware client purge)

(Last modified: 06Sep2005)

This document (10098599) is provided subject to the disclaimer at the end of this document.

fact

Novell NetWare 6.0 SP5

nw6nss5b.exe post sp5 patch applied

symptom

Server abends during purge of volume (toolbox purge, dos prompt purge, filer purge, or netware client purge)

Server abends every 5 days and 25 minutes

deleted.sav directory did not exist on root of volume

Server resets itself during NSS /PoolVerify=<poolname> of the volumes parent pool

cpu hog detected by timer abend every 5 days 25 minutes

note

Server OKCNV001 halted Saturday, August 13, 2005   9:56:06.125 am
Abend 2 on P00: Server-5.60.05-1937: CPU Hog Detected by Timer

Registers:
    CS = 0008 DS = 0010 ES = 0010 FS = 0010 GS = 0010 SS = 0010
    EAX = FCE17833 EBX = CD4F0E40 ECX = CB25EDA8 EDX = 00000000
    ESI = 00000246 EDI = 00000000 EBP = CA5A4B1C ESP = CA5A4B08
    EIP = 00000000 FLAGS = 00000046
   

Running process: Server 00:5 Process
Thread Owned by NLM: SERVER.NLM
Stack pointer: CA5A4F80
OS Stack limit: CA59D040
CPU 0 (Thread CA593260) is in a NO SLEEP state
Scheduling priority: 67371008
Wait state: 3030070  Yielded CPU
Stack: --FCE17833  ?
       CAFE51C9  (LIBNSS.NLM|nssSpinLock+9)

LIBNSS.NLM|nssSpinLock+9  -- showed up in virtually all abend.log files

 

change

Server hardware disk array drive failure occurred, during the drive replacement and subsequent server restart, the option for writing the cached information from the drive array controller to the disk drives was NOT selected.

cause

File system corruption had occurred during the physical drive failure, by not selecting to write the cached information from the controller to the disk array, information destined for the file system was lost.

Each time the server is restarted, NSS timed background tasks are started.  At 5 days, the NSS background check for the Salvage Name Tree is started.  Due to the corruption that occurred during the drive failure, the Salvage Name Tree and Purgable Entries did not match.

fix

Corrected pool inconsistencies:

  • Commented out third party applications
  • Created deleted.sav on volume root
  • Flagged deleted.sav on volume root as hidden
  • Reset server
  • NSS /PoolRebuild=<poolname> /Purge
  • NSS /PoolRebuild=<poolname>
  • NSS /PoolActivate=<poolname>
  • Mount <volumename>

Validated that salvage system worked after rebuild as follows:

  • Created temporary directory on rebuilt volume
  • Copied files to temporary directory
  • Deleted temporary directory (which also deleted the files)
  • Ran filer to salvage the directory and file

Returned server to previous state:

  • Uncommented Third Party applications
  • Reset server
  • Tested Third Party applications (including backup and restore tools for rebuilt pools volumes)

document

Document Title: Server abends during purge of volume (toolbox purge, dos prompt purge, filer purge, or netware client purge)
Document ID: 10098599
Solution ID: NOVL103090
Creation Date: 13Aug2005
Modified Date: 06Sep2005
Novell Product Class:Netware

disclaimer

The Origin of this information may be internal or external to Novell. Novell makes all reasonable efforts 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.
Any trademarks referenced in this document are the property of their respective owners. Consult your product manuals for complete trademark information.