Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-26548

How can one spot 2.6 serverStatus / executionStats / log Output equivalents in 3.2 WiredTiger?

    XMLWordPrintable

    Details

    • Type: Question
    • Status: Closed
    • Priority: Major - P3
    • Resolution: Done
    • Affects Version/s: 3.2.10
    • Fix Version/s: None
    • Labels:
      None

      Description

      Having recently upgrading our entire Mongo DB replica sets array to 3.2.10,
      We noticed that our monitoring scripts, which used to utilise specific metrics available on 2.6 began breaking due to having these metrics no longer available to them using serverStatus.

      Can you please refer us to the equivalent (whether WiredTiger related or not) metric names for the following metrics?

      memory accessesNotInMemory
      memory pageFaultExceptionsThrown
      memory admin accessesNotInMemory
      memory admin pageFaultExceptionsThrown
      memory local accessesNotInMemory
      memory local pageFaultExceptionsThrown
      memory test accessesNotInMemory
      memory test pageFaultExceptionsThrown
      memory wixExpress accessesNotInMemory
      memory wixExpress pageFaultExceptionsThrown
      flushing flushes
      flushing total_ms
      flushing average_ms
      flushing last_ms
      flushing last_finished
      indexes accesses
      indexes hits
      indexes misses
      indexes resets
      indexes missRatio
      journaling commits
      journaling journaledMB
      journaling writeToDataFilesMB
      journaling compression
      journaling commitsInWriteLock
      journaling earlyCommits
      journaling timeMs dt
      journaling timeMs prepLogBuffer
      journaling timeMs writeToJournal
      journaling timeMs writeToDataFiles
      journaling timeMs remapPrivateView

      plus, we noticed that our 3.2.10 nodes no longer produce the level of detail in terms of queries stats served. i.e, we used to have:

      ntoreturn:20 ntoskip:0 nscanned:45644 nscannedObjects:45644 scanAndOrder:1 keyUpdates:0 writeConflicts:0 numYields:1426 nreturned:1 reslen:6790 locks:{ Global: { acquireCount: { r: 2854 } }, Database: { acquireCount: { r: 1427 } }, Collection: { acquireCount: { r: 1427} } } 1845ms
      

      and now cannot find their equivalent on the new logs...
      is there some sort of backward compatibility flag to be switched on, or a log level one missing here?

      Thanks,
      Avi Kalvo
      WiX.COM

        Attachments

          Activity

            People

            • Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: