• State Not Answered
  • Locked Locked
  • Replies 5 replies
  • Subscribers 59 subscribers
  • Views 5033 views
  • Users 0 members are here
Options
Related
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Foglight traces in SQL error log- how do you stop these?

chubbsm
over 5 years ago

 I tried setting the Deadlock collection for every 60 minutes but still showing as every 15 minutes. What process/collection controls this? Can it be modified Enterprise wide, not every agent?

  • 0 over 5 years ago
    Hi Chubbsm,

    The deadlock collection will reset (stop and start) the trace thread if one of the following options take place:
    A) No event occurred in at least X minute time period (default 30 minutes)
    B). The trace ID does not appear in the SQL Server instance view 'sys.fn_trace_geteventinfo'

    Knowledgebase article 229591 contains a script to change the default from 30 minutes to once per day. Our Support team can also assist with providing a script to making the change for all of the SQL Server agents.

    John
  • 0 over 5 years ago
    The KB script didn't do anything for me. Still kept running every 15 minutes. I tried updating the collections as instructed by support; no change. The only thing that stops it is to disable the deadlock collection entirely which isn't really what I want. Is there anything else to look at or update? Can I get a copy of the script to update all agents? Maybe that would work.
  • 0 over 5 years ago
    Hi Chubbsm,

    I recommend that you open a Service Request with our Support team for this. There have been some changes to the trace collection in recent cartridges and it is difficult to provide general suggestions when the specific environment information (including the cartridge version) isn't available.

    We recently encountered an issue similar to what you described where it was determined that another process (external to Foglight) was killing the trace every fifteen minutes and a new trace was being opened using the same trace id.

    Another potential cause we have seen is when two different SQL Server agents monitor the same database instance.

    The script to update all of the agents is not included in the KB article because of the broader change that it makes to all database agents. A Support Engineer assigned to the Service Request can also provide the script for updating all of the agents.

    John
  • 0 over 5 years ago
    Hi John - I had opened a request previously and got the same info you provided. The only way I got it to stop short of disabling it was to provide crazy sized numbers for the collection parameters. It doesn't look like updating the script had any affect.
  • 0 over 5 years ago
    Hi Chubbsm,

    If you could reopen the previous Service Request or open a new Service Request, our Support Team can reengage to review some of the items suggested previously, and engage the cartridge developers as well.

    John