Uploaded image for project: 'controller'
  1. controller
  2. CONTROLLER-954

Clustering: ShardWriteTransaction instances don't get garbage collected

    XMLWordPrintable

Details

    • Bug
    • Status: Resolved
    • Resolution: Done
    • Helium
    • None
    • mdsal
    • None
    • Operating System: All
      Platform: All

    • 2252

    Description

      Looking at a heap dump showed stale Shard(Read)WriteTransaction instances that aren't getting garbage collected.

      This was caused by recent changes that removed the ThreePhaseCommitCohort actor class however ThreePhaseCommitCohort was the one who terminated the ShardWriteTransaction actor after forwarding the commit to the Shard.

      Solution:

      After forwarding the ReadyTransaction message to the shard, the ShardWriteTransaction actor is no longer needed so it can self-destruct (i.e. send PoisonPill to itelf).

      Attachments

        No reviews matched the request. Check your Options in the drop-down menu of this sections header.

        Activity

          People

            tpantelis Tom Pantelis
            tpantelis Tom Pantelis
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: