Forum Discussion

Hermann2009's avatar
Hermann2009
Level 4
16 years ago
Solved

Index Server: Percentage completed: 295%

Dear all,
because of emory troubles on my Vault Server, i restart the Enterprise Vault Services.
Following the Index Process is rebuilding the Index of a User Vault.
Now i got the information 295% are completed ...
How could i finish the indexing process?

Server: Windows 2008, Vault 8 SP3, SQL 2005 SP3

the Log:
++++++++++++++++++++++++++++++++++++++++++++++
Log Name:      Symantec Enterprise Vault
Source:        Enterprise Vault
Date:          14.12.2009 14:38:06
Event ID:      7314
Task Category: Index Server
Level:         Information
Keywords:      Classic
User:          N/A
Computer:      Servername
Description:
Index volume update in progress

Index Volume: 12BB1CF84C973A343A465F21A59FD64721110000vie-eu-archiv/Volume:30 (Username)

Percentage completed: 295%
Estimated time of completion: 14.12.2009 15:20

Index Volume Path: E:\Index\index2\12BB1CF84C973A343A465F21A59FD64721110000vie-eu-archiv
Job Id: Vol_30
Job Author: archivuser
Job Description: Rebuild index volume

Job elapsed time: 05:37:48 (hours:minutes:seconds)
Processing time: 05:13:29 (hours:minutes:seconds)
Number of items added: 98320
Number of items additions abandoned: 0
Number of items deleted: 0
Number of items updated: 0

Basics
Enterprise Vault
Information Governance
  • Liam_Finn1's avatar
    Liam_Finn1
    16 years ago
     Your DTrace shows no sign of failure or error. Maybe a Dtrace to include the previous Indexserver and also the storage service but also add Storagewalker and storagebroker services to see if you can capture the error

    Try the rebuild first. If there is a failure then DTrace as above with the two additional and then try again to see if we can capture the error

    When you do Dtrace places set the trace to Verbose mode to capture as much detail as possable also set the dtrace buffer very high. To do this type dtrace 50000 to start Dtrace this will make the buffer larger to ensure nothing is missed

10 Replies

  • Liam_Finn1's avatar
    Liam_Finn1
    Level 6
    16 years ago
    Do a dtrace in the Indexbroker service and the Indexserver and also the storage service

    This will tell if there are any issues writing to the indexes and also if there are any issues when talking to storage to find the files to rebuild the indexes

    Post the results

    The results will get very big so keep an eye on the log files
  • Hermann2009's avatar
    Hermann2009
    Level 4
    16 years ago
    the Vault is really defective...


    Log Name:      Symantec Enterprise Vault
    Source:        Enterprise Vault
    Date:          15.12.2009 10:47:43
    Event ID:      7235
    Task Category: Index Server
    Level:         Error
    Keywords:      Classic
    User:          N/A
    Computer:      servername
    Description:
    A low level indexing operation has failed
    Error: a severe error occurred; cannot continue [AV:63]
    Index Id: 1AE898FE00470684CB2EF5EFBCC7DB7851110000vie-eu-archiv/Volume:36 (username)


  • MichelZ's avatar
    MichelZ
    Level 6
    16 years ago
    You may want to just delete the index and rebuild it from scratch... cheers
  • Hermann2009's avatar
    Hermann2009
    Level 4
    16 years ago
    hi
    any experience how long it should take to rebuild an index of 31540 object with 6334MB ?
  • MichelZ's avatar
    MichelZ
    Level 6
    16 years ago
    It should be identical to normal archiving troughtput, I believe....

    /Michel
  • Hermann2009's avatar
    Hermann2009
    Level 4
    16 years ago
    Dear all,
    in the same time as the Vault creates many errors in the eventlog i start to log with dtrace into a log file (see above)
    any ideas whats wrong in my Vault environment because the indexing services could not finish?
  • Liam_Finn1's avatar
    Liam_Finn1
    Level 6
    16 years ago
     It does not seem like it is a vault issue. The system is failing on indexing a volume.

    Does it always fail on the same Volume:36? 

    As Michel suggested do a complete rebuild on the index again and see how it works out. If it fails on the same volume again let us know. There is a way to rebuild an index in a way that it ignores errors and keeps going to completion but see if a complete rebuild of the index works first and see if it fails on the same volume each time

    If it continues to fail on the same place then we can give you the instructions on how to override the failure count to see if it will complete then you can address the errors on the volume if it is a volume issue
  • Liam_Finn1's avatar
    Liam_Finn1
    Level 6
    16 years ago
     Your DTrace shows no sign of failure or error. Maybe a Dtrace to include the previous Indexserver and also the storage service but also add Storagewalker and storagebroker services to see if you can capture the error

    Try the rebuild first. If there is a failure then DTrace as above with the two additional and then try again to see if we can capture the error

    When you do Dtrace places set the trace to Verbose mode to capture as much detail as possable also set the dtrace buffer very high. To do this type dtrace 50000 to start Dtrace this will make the buffer larger to ensure nothing is missed
  • Hermann2009's avatar
    Hermann2009
    Level 4
    16 years ago
    we start an index rebuild for this user, after 10 hours & 1000+1 error's in the eventlog ( event id 7313 + 7235 )
    the index works and now the Virtual Vault Cache sync works...
  • Liam_Finn1's avatar
    Liam_Finn1
    Level 6
    16 years ago
    Event ID 7313 could be a sign that you have insufficent disk space to house your indexes.

    can you post the complete Event info for the two above event ID's