This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

File Age Monitor Issue - odd sawtooth pattern...

I have a few file age monitors all of which demonstrate the same odd behavior.  This is making my alerts less than accurate.  Anyone seen anything like this?

NPM 9.5.0 SP4

APM 3.1

  • Hello mstolle,
    what files do you monitor? Your chart is really strange but file age can be modified so maybe that files you are monitoring are used by some software that does strange things with file's date.

    If you are sure that file modification time doesn't change this way, please open a support ticket for this issue.

  • Jiri,

    They are text files.  They are created on a separate system at the end of a particular daily processing job, and are only created if the processing job was successful.  This job only runs once a day each week day.  Once the job successfully completes, they are FTPd to the server where I am monitoring their age.  So effectively they are files that are simply overwritten once a day and are not accessed dynamically.

    I'll open a ticket and see where that takes me.

    Thanks for your help.

  • We are seeing the same issue here.  I've tested on different servers with different files and they all jump by 24 hours between 6PM-Midnight every day.  However, if I test it against the server running Orion it does not exhibit those symptoms.

    I opened a ticket several days ago but do not have a resolution yet.

  • Case # 148751 Opened 17-Mar.

  • I am experiencing the same EXACT problem with the file age monitor. Between 6pm and midnight all my file age monitors across multiple servers, jumps up 24 hours, and comes back down after midnight.

    I have SW case 148522 opened, and the technician just told me that this has been identified as a system wide issue and is being looked at by the developers now.

    I am eagerly waiting for a patch:)

     

    For anyone who is using this component as a check to make sure a file is being constantly updated, i do have a suggestion to stop the false alarms.

    I am alarming if a file hasn't been updated within an hour, so I just put in the alerting rule to not alert if the file age goes over 24 hours. Before doing this, I would get a false alarm every day at 6pm

  • I am seeing the same bug, even after specifically upgrading to rev 3.1 for this functionality.

    I saw several posts from the Solarwinds developers about this crucial component that many of us need to use, and many of them were cheesy one-liners "Fixed in 3.1".  That is rubbish.

    The SolarWinds developers need to take the time to properly QA the service packs and there response posts.  Instead of throwing out half=baked code and inaccurate statements about their fixes.

    Regardless of these issues, this function needs to be fixed asap because serious users of this product want and need the functionality to work, period.  Isn't that why we pay a premium for maintenance.

  • I just received a message from Solarwinds Support:

    "Development has made a Fix for the File Age Monitor Issue. I have Tested and Verified that the issue has been resolved. After you Download, You will need to stop all Solarwinds Services, then you will need to put it in this location:

    C:\Program Files\Solarwinds\Orion\APM

    Then start all Solarwinds Services.

     

    Hopefully this will fix the issue.  I'll post my results after the patch.

  • This did the trick.  With the new DLL in place, the sawtooth does not repeat.  Looks like a winner.

  • mstolle, we really appreciate you sharing the results of your testing with the community, but the "test" patch was only intended for support distribution.   While it may have fixed your specific issue, it's too early to tell whether it will fix everyone's issue.   I apologize if expectations weren't set appropriately.

    For folks experiencing this problem, please open a ticket with support and we'll include you in the testing process as appropriate.

  • So what was the result of this thread?  Was the fix for this issue every released to the general populace or is it still for "support" only?

    What is the fix?