[Deploying Sakai] Search tool and rebuilding index

Tue May 3 23:16:57 PDT 2011

Thanks Anand,

You've been of a great help.
I'm testing it now on one server and tomorrow I'll test it with shared "
sharedJournalBase at org.sakaiproject.search.api.JournalSettings=<common
directory accessible by all tomcats>/search"
on NFS share.

1) Should I expect any problems given the fact that 2 servers would access
sharedJournalBase at org.sakaiproject.search.api.JournalSettings location
simultaniously?

2) What would be the reason for the following messages (after pressing "Rebuild
Whole Index") and how to mitigate it:

2011-05-04 15:37:50,859  WARN Timer-1
org.sakaiproject.search.optimize.shared.impl.JournalOptimizationOperation -
Failed to compete optimize
org.sakaiproject.search.optimize.api.OptimizedFailedIndexTransactionException:
Failed to Optimize indexes
        at
org.sakaiproject.search.optimize.shared.impl.OptimizeSharedTransactionListenerImpl.prepare(OptimizeSharedTransactionListenerImpl.java:292)
        at
org.sakaiproject.search.transaction.impl.IndexTransactionImpl.firePrepare(IndexTransactionImpl.java:313)
        at
org.sakaiproject.search.transaction.impl.IndexTransactionImpl.prepare(IndexTransactionImpl.java:147)
        at
org.sakaiproject.search.optimize.shared.impl.JournalOptimizationOperation.runOnce(JournalOptimizationOperation.java:82)
        at
org.sakaiproject.search.journal.impl.IndexManagementTimerTask.run(IndexManagementTimerTask.java:139)
        at java.util.TimerThread.mainLoop(Timer.java:512)
        at java.util.TimerThread.run(Timer.java:462)
Caused by: java.io.FileNotFoundException: no segments* file found in
org.apache.lucene.store.FSDirectory@/srv/apache-tomcat-5.5.26/sakai/indexwork/journal-optimize-import/3:
files:
        at
org.apache.lucene.index.SegmentInfos$FindSegmentsFile.run(SegmentInfos.java:516)
        at org.apache.lucene.index.IndexReader.open(IndexReader.java:185)
        at org.apache.lucene.index.IndexReader.open(IndexReader.java:167)
        at
org.sakaiproject.search.optimize.shared.impl.OptimizeSharedTransactionListenerImpl.prepare(OptimizeSharedTransactionListenerImpl.java:194)
        ... 6 more
2011-05-04 15:39:20,930  INFO Timer-1
org.sakaiproject.search.optimize.shared.impl.DbJournalOptimizationManager -
Rolled Back Failed Shared Index operation a retry will happen on annother
node soon
2011-05-04 15:39:21,060  INFO Timer-1
org.sakaiproject.search.journal.impl.MergeUpdateOperation - Local Merge
Operation
        Merged Journal 59 from the redolog in 64 ms
        Merged Journal 60 from the redolog in 62 ms

2011-05-04 15:39:23,524  WARN Timer-1
org.sakaiproject.search.component.adapter.contenthosting.ContentHostingContentProducer
- Failed to digest /content/group/963f86af-9012-4626-bdcd-e8592c60c1db/MIHSR
Leadrership course module 2.pdf wi
th
org.sakaiproject.search.component.adapter.contenthosting.PDFContentDigester at 50700ae1cause:
Failed to get content for indexing: cause: Error: expected hex
character and not  :32
2011-05-04 15:39:23,524  INFO Timer-1
org.sakaiproject.search.component.adapter.contenthosting.ContentHostingContentProducer
- Digested /content/group/963f86af-9012-4626-bdcd-e8592c60c1db/MIHSR
Leadrership course module 2.pdf into 43 ch
aracters with Default Digester
2011-05-04 15:39:33,461  WARN Timer-1
org.sakaiproject.search.component.adapter.contenthosting.ContentHostingContentProducer
- Failed to digest /content/group/963f86af-9012-4626-bdcd-e8592c60c1db/MIHSR
Leadrership course module 1.pdf wi
th
org.sakaiproject.search.component.adapter.contenthosting.PDFContentDigester at 50700ae1cause:
Failed to get content for indexing: cause: Error: expected hex
character and not  :32
2011-05-04 15:39:33,461  INFO Timer-1
org.sakaiproject.search.component.adapter.contenthosting.ContentHostingContentProducer
- Digested /content/group/963f86af-9012-4626-bdcd-e8592c60c1db/MIHSR
Leadrership course module 1.pdf into 43 ch
aracters with Default Digester
2011-05-04 15:39:47,987  INFO Timer-1
org.sakaiproject.search.journal.impl.SharedFilesystemJournalStorage - ++++++
Saving /srv/tomcat/sakai/indexwork/indexer-work/indextx-1304483547557 to
shared
2011-05-04 15:39:48,629  INFO Timer-1
org.sakaiproject.search.indexer.impl.TransactionalIndexWorker - Indexed 198
documents in 27529 ms 7.192415271168586 documents/second into save point 61
2011-05-04 15:39:48,650  INFO Timer-1
org.sakaiproject.search.optimize.shared.impl.DbJournalOptimizationManager -
Locked 59 savePoints

Thanks,
Leon Kolchinsky

On Wed, May 4, 2011 at 00:10, Anand Mehta <anand.mehta at yahoo.com> wrote:

> Hello Leon,
>
> You can add the following settings to sakai.properties:
>
> localIndexBase at org.sakaiproject.search.api.JournalSettings
> =${sakai.home}/indexwork
> sharedJournalBase at org.sakaiproject.search.api.JournalSettings=<common
> directory accessible by all tomcats>/search
>
> After you have restarted all Tomcats, go to the search tool in the admin
> workspace (add it if not there), click on Admin link and then click "Rebuild
> Whole Index". I hope this helps.
>
>
> Thanks,
> Anand
> ------------------------------
> *From:* Leon Kolchinsky <lkolchin at gmail.com>
> *To:* production <production at collab.sakaiproject.org>
> *Sent:* Tuesday, May 3, 2011 12:32 AM
> *Subject:* [Deploying Sakai] Search tool and rebuilding index
>
> Hello,
>
> I'd like to enable "Search tool" on both of our Sakai nodes (v.2.6.2)
> running behind load balancer and using the same Oracle DB.
>
> I'm testing it now on the dev. server and it seems that it's not finding
> announcement for example.
>
> I've read https://confluence.sakaiproject.org/display/SEARCH/Home but find
> it a bit non user friendly (i.e. configuration instructions ;).
>
> So I've several questions:
>
> 1) What other parameters should I use for  for my installation (besides
> search.enable=true).
>
> 2) How can I rebuild indexes?
> What command should I run? What parameter should I configure for that in
> sakai.properties?
> Should I configure a path for indexes in sakai.properties? How would you
> sync. those indexes on both nodes?
> If I have 2 nodes running on the same Oracle DB
> (Sorry couldn't find much info on that)
>
>
> Thanks in advance for any help,
> Leon Kolchinsky
>
>
> _______________________________________________
> production mailing list
> production at collab.sakaiproject.org
> http://collab.sakaiproject.org/mailman/listinfo/production
>
> TO UNSUBSCRIBE: send email to
> production-unsubscribe at collab.sakaiproject.org with a subject of
> "unsubscribe"
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://collab.sakaiproject.org/pipermail/production/attachments/20110504/9262de87/attachment.html 