Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-18637

Measure initial lock time in operation latency

    • Type: Icon: Bug Bug
    • Resolution: Duplicate
    • Priority: Icon: Major - P3 Major - P3
    • None
    • Affects Version/s: None
    • Component/s: Admin, Diagnostics
    • None
    • ALL
    • Query 2017-06-19

      Based on this thread, v3.0 should include initial locking time in total operation latency: https://groups.google.com/forum/#!searchin/mongodb-dev/mark$20callaghan%7Csort:date/mongodb-dev/l2fR4Xr-Bxo/ibYPp7BB4-cJ

      However, this is not true. As an example, here is a log line of an operation that claims latency of 2ms, but it waited 1.5sec for a lock:

       query ... query: { .... } planSummary: IXSCAN { venueId: 1, timestamp: 1, activeCommand: 1 } ntoreturn:100 ntoskip:0 nscanned:31 nscannedObjects:31 keyUpdates:0 writeConflicts:0 numYields:1 nreturned:31 reslen:6475 locks:{ Global: { acquireCount: { r: 4 } }, Database: { acquireCount: { r: 2 }, acquireWaitCount: { r: 1 }, timeAcquiringMicros: { r: 1468984 } }, Collection: { acquireCount: { r: 2 } } } 2ms
      

            Assignee:
            david.storch@mongodb.com David Storch
            Reporter:
            igor Igor Canadi
            Votes:
            2 Vote for this issue
            Watchers:
            17 Start watching this issue

              Created:
              Updated:
              Resolved: