Summarising what we know so far and adding a few more details: 1. We understand why mongo was continually restarting. jujud currently restarts mongo every time it starts so every time jujud panicked (~ every 10s), upstart would restart jujud and jujud would restart mongo. This explains the observed mongo restarts and upstart config rewrites. This isn't ideal behaviour and is being fixed right now (bug #1349969)
2. As per @niemeyer, the txn collection is definitely not capped. txn.log is capped but in the dump provided by Kapil above it only has 15,671 documents in it. The cap is 10,000,000 so there's plenty of breathing room. Capped collections is unlikely to be at play here. 3. @sabdfl, @adam-collard and @james-page have all seen this issue in separate environments. There may be a scale/load component to this (although Adam says he's seen it with a small scale Juju environment). Does anyone have any ideas on what to look at from here? -- You received this bug notification because you are a member of Ubuntu Server Team, which is subscribed to juju-core in Ubuntu. https://bugs.launchpad.net/bugs/1318366 Title: jujud on state server panic misses transaction in queue To manage notifications about this bug go to: https://bugs.launchpad.net/juju-core/+bug/1318366/+subscriptions -- Ubuntu-server-bugs mailing list Ubuntu-server-bugs@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-server-bugs
