Uploaded image for project: 'Fabric'
  1. Fabric
  2. FAB-14499

RAFT OSN1 does not receive consensusRequest for 6 seconds

    XMLWordPrintable

Details

    • Bug
    • Status: Closed
    • Medium
    • Resolution: Invalid: Environment Issue
    • None
    • None
    • fabric-orderer
    • (Please add steps to reproduce)

    Description

      Under the stress test (FAB-14350), the followers OSN1 does not receive ConsensusRequest from leader OSN2 for 6 seconds and then initiates leader election.

      at 21:44.394, both OSN1 and OSN3 respond to leader OSN2:
      Feb 27 22:21:44 orderer1st-ordererorg orderer1st-ordererorg 2019-02-28 03:21:44.394 UTC [orderer.common.cluster.step] sendMessage -> DEBU 37a20fc Send of ConsensusRequest for channel testorgschannel139 with payload of size 28 to orderer2nd-ordererorg.ordererorg(orderer2nd-ordererorg:7050) took 16.181µs
      Feb 27 22:21:44 orderer1st-ordererorg orderer1st-ordererorg 2019-02-28 03:21:44.433 UTC [orderer.common.cluster.step] sendMessage -> DEBU 37a23c9 Send of ConsensusRequest for channel testorgschannel139 with payload of size 28 to orderer2nd-ordererorg.ordererorg(orderer2nd-ordererorg:7050) took 27.121µs

      OSN2 keeps sending sonsensusRequest every 500ms:
      Feb 27 22:21:48 orderer2nd-ordererorg orderer2nd-ordererorg 2019-02-28 03:21:48.483 UTC [orderer.common.cluster.step] sendMessage -> DEBU 3d027f0 Send of ConsensusRequest for channel testorgschannel139 with payload of size 28 to orderer3rd-ordererorg.ordererorg(orderer3rd-ordererorg:7050) took 23.266µs
      Feb 27 22:21:48 orderer2nd-ordererorg orderer2nd-ordererorg 2019-02-28 03:21:48.483 UTC [orderer.common.cluster.step] sendMessage -> DEBU 3d027f1 Send of ConsensusRequest for channel testorgschannel139 with payload of size 28 to orderer1st-ordererorg.ordererorg(orderer1st-ordererorg:7050) took 21.426µs

      OSN1 does not receive consensusRequest from OSN2 for more than 5 seconds. As a result, OSN1 initiates leader election at 21:50.079:
      Feb 27 22:21:50 orderer1st-ordererorg orderer1st-ordererorg 2019-02-28 03:21:50.079 UTC [orderer.consensus.etcdraft] Step -> INFO 37a2e6f 1 is starting a new election at term 11 channel=testorgschannel139 node=1

      The log does not show the reason why OSN1 does not receive consensusRequest from OSN2 for 6 seconds.

      The log is attached to the bug.

      The logDNA is available in LogDNA-community-quality-lab, with the following filter:
      02/27 10:21pm to 02/27 11pm and "testorgschannel139"

      Attachments

        Issue Links

          Activity

            People

              Unassigned Unassigned
              dongming Dongming Hwang
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: