none
Process indexer terminated by signal SIGSEGV RRS feed

  • Question

  • We have a indexer node that dies suddenly. The ADMIN node receives the message "Process indexer terminated by signal SIGSEGV" for this indexer. There is no smoking gun in the indexer.log file, even in debug mode.

    Prior to the error occurring, the indexer is suspended while we load only document updates. As we are loading, I can see new fixml directories being created but the total document count does not go up. The size of the data_index/blacklist.txt file steadily grows over time.

    We can only recover this indexer after running a defragfixml command. The same node then dies in the same fashion a few days later.

    We are running ESP 5.3 SP4, searchengine_patch07.  We have tried moving the hard drives from the problem server to a proven good server, so I doubt that h/w is the problem.

    Any ideas?

    Thursday, November 3, 2011 9:48 PM

Answers

  • Hello,

     

    Thanks for this post.  I believe that this was addressed in esp.5.3.SP4.searchengine.patch08.  One of the reasons you would want to apply patch08, per the readme, is “if you experience sudden crashes in indexer that prevent further feeding until a defrag operation is run.” I would recommend downloading this patch from MSConnect (connect.microsoft.com) and test this in your environment to confirm resolution.  Let us know your results.

     

    Thanks!

    Rob Vazzana | Sr Support Escalation Engineer | US Customer Service & Support

    Customer Service & Support                          Microsoft | Services

    Monday, November 7, 2011 3:03 PM
    Moderator