Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-7488

Database lockup during map/reduce job

    XMLWordPrintableJSON

Details

    • Icon: Bug Bug
    • Resolution: Done
    • Icon: Critical - P2 Critical - P2
    • None
    • 2.2.0
    • MapReduce
    • None
    • Linux mpac-fe-dev.private.atlassian.com 2.6.32-220.el6.x86_64 #1 SMP Tue Dec 6 19:48:22 GMT 2011 x86_64 x86_64 x86_64 GNU/Linux
      16GB RAM
    • Linux

    Description

      We saw this morning that our staging server running MongoDB 2.2.0 was locked up. Several DB requests were queued up and would not complete. A long-running map/reduce job had just begun and didn't go more than a minute before the updates ceased.

      mongod backtrace via gdb:

      #0  0x0000003f1bae0d03 in select () from /lib64/libc.so.6
      #1  0x00000000006a4a5a in mongo::Listener::initAndListen() ()
      #2  0x0000000000559154 in mongo::listen(int) ()
      #3  0x000000000055c86a in mongo::_initAndListen(int) ()
      #4  0x000000000055d2dd in mongo::initAndListen(int) ()
      #5  0x0000000000563dbe in ?? ()
      #6  0x0000000000565399 in main ()

      See an attachment for db.currentOp() output.

      Attachments

        1. currentOp.json
          13 kB
        2. currentOp-2.json
          8 kB
        3. downloads.json
          10 kB
        4. gdb-2.out
          56 kB
        5. mongodb.log
          4.77 MB

        Activity

          People

            tad Tad Marshall
            arun Arun Bhalla
            Votes:
            1 Vote for this issue
            Watchers:
            6 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: