none
Indexing halt on FSIS multi node installation RRS feed

  • Question

  • Hi All,

    I have deployed the search solution to the customer based on the FSIS multinode environment.
    I used the content feeder API to feed the content into indexer. However, I faced with the issue that the indexing process was halt while indexing.

    I have checked the following criteria
    • Average disk queue is less than 0.1 most of the time
    • CPU utilization is less than 10% most of the time
    • The TCPv4:Connection Error / TCPv4:Connection Active in performance monitoring is in range 0.3 – 0.8
    • The machine has physical memory 96 GB
    So, the disk and CPU issue should not be the root cause.

    And when the indexing is halted, there is no error log in the var/log. 
    So, I have enabled the indexer log using 'indexeradmin enabledebug' command. And here is the log I have (which might be helpful.. Hopefully).

    [2012-01-17 23:41:26.228] DEBUG      indexer index_producer (0): Waiting for index completion (43%)

    And if I trace back to previous line where this log occurs, it should be followed by simple_index_producer_callback as below:

    [2012-01-17 23:40:20.538] DEBUG      indexer index_producer (0): Waiting for index completion (0%)
    [2012-01-17 23:40:25.331] DEBUG      indexer simple_index_producer_callback (0): Indexing completed (directly fed 0-0). 6792 OK docs, 0 failed docs, and 0 errors

    So, it seems the indexer is waiting for the simple_index_producer_callback which is never called so that the indexing was halt.

    Have you ever experienced with this issue?
    Best Regards,
    Andy
    Wednesday, January 18, 2012 6:48 AM

Answers

  • Hi Rob,

    I used the ESP 5.3 slipstream on Windows Server 2008 R2.

    Then I get recommendation that I should install the service pack 4 with the searchengine.patch09 and it seems the error is gone.

     

    So, I think the ESP 5.3 slipstream might  not compatible with Windows Server 2008 R2. And Update the pack is the solution.

    Thanks for your interesting :)

     

    Best Regards,

    Andy

    Tuesday, January 24, 2012 2:49 AM

All replies

  • Hi Andy,

     

    Are you using FSIS with CTS flows to feed content into ESP, or are you just directly feeding content using the content API into ESP.  Maybe you are doing the latter.

     

    If you are using CTS flows to feed content into ESP,  can you provide an error on the FSIS side from the ContentEngineNode logs?  It might tell us some information about the problem, such as the contentdistributor is refusing connections, or other issue.  Let us know your findings.

     

    Thanks!

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

    Customer Service & Support                          Microsoft | Services

    Monday, January 23, 2012 8:32 PM
    Moderator
  • Hi Rob,

    I used the ESP 5.3 slipstream on Windows Server 2008 R2.

    Then I get recommendation that I should install the service pack 4 with the searchengine.patch09 and it seems the error is gone.

     

    So, I think the ESP 5.3 slipstream might  not compatible with Windows Server 2008 R2. And Update the pack is the solution.

    Thanks for your interesting :)

     

    Best Regards,

    Andy

    Tuesday, January 24, 2012 2:49 AM