Uploaded image for project: 'Couchbase Server'
  1. Couchbase Server
  2. MB-39548

[Jepsen] KeyEexists received from DCP Consumer for DCP_COMMIT

    XMLWordPrintable

Details

    • Bug
    • Status: Closed
    • Critical
    • Resolution: Duplicate
    • 6.6.0
    • 7.0.0
    • couchbase-bucket
    • Ubuntu 1804
      Couchbase Server 6.6.0-7686-1

    Description

      Couchbase error found due to unexpected status code received for DCP_COMMIT during kv-engine-mad-hatter-jepsen-nightly-530. While running test lein trampoline run test --nodes-file ./nodes --username root --password couchbase --workload=rebalance --scenario=sequential-rebalance-out-in --replicas=2 --no-autofailover --disrupt-count=1 --rate=0 --durability=0:100:0:0 --bucket-type=ephemeral --cas --enable-tcp-capture --hashdump --enable-memcached-debug-log-level --net-interface=eth0

      Error message:

      ./20200523T212123.000-0700/172.23.104.123/cbcollect_info_ns_1@172.23.104.123_20200524-042422/memcached.log:2020-05-23T21:23:05.630665-07:00 ERROR 58: (default) DCP (Producer) eq_dcpq:replication:ns_1@172.23.104.123->ns_1@172.23.106.35:default - DcpProducer::handleResponse disconnecting, received unexpected response:{"bodylen":0,"cas":0,"datatype":"raw","extlen":0,"keylen":0,"magic":"ClientResponse","opaque":130,"opcode":"DCP_COMMIT","status":"Data exists for key"}
      

      Attachments

        Issue Links

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

          Activity

            owend Daniel Owen added a comment -

            Have also seen the "KeyEexists received from DCP Consumer for DCP_COMMIT" error in MB-40053.
            Dave Rigby Could you take a look, as I'm unsure how we should proceed on this one?

            1) Should we be disconnecting (which results in rebalance failure) when KeyEexists is received from DCP Consumer for DCP_COMMIT?
            2) If yes, then does this mean we have a bug with durability?

            owend Daniel Owen added a comment - Have also seen the "KeyEexists received from DCP Consumer for DCP_COMMIT" error in MB-40053 . Dave Rigby Could you take a look, as I'm unsure how we should proceed on this one? 1) Should we be disconnecting (which results in rebalance failure) when KeyEexists is received from DCP Consumer for DCP_COMMIT? 2) If yes, then does this mean we have a bug with durability?

            Daniel Owen Please see MB-37009 and http://review.couchbase.org/c/kv_engine/+/130046, current behaviour of 6.6.0 is that we won't disconnect when a DCP Producer receives a KeyEexists status from a DCP Consumer for DCP_COMMIT as o build: 6.6.0-7789 and 7.0.0-2356. The build used in MB-40053 is 7.0.0-2351. Think we can mark these MBs as resolved.

            richard.demellow Richard deMellow added a comment - Daniel Owen Please see MB-37009 and http://review.couchbase.org/c/kv_engine/+/130046 , current behaviour of 6.6.0 is that we won't disconnect when a DCP Producer receives a KeyEexists status from a DCP Consumer for DCP_COMMIT as o build: 6.6.0-7789 and 7.0.0-2356. The build used in MB-40053 is 7.0.0-2351. Think we can mark these MBs as resolved.
            owend Daniel Owen added a comment -

            Based on Richard deMellow feedback marking as resolved.

            owend Daniel Owen added a comment - Based on Richard deMellow feedback marking as resolved.
            owend Daniel Owen added a comment -

            Marking as a duplicate of MB-37009

            owend Daniel Owen added a comment - Marking as a duplicate of MB-37009

            Bulk closing invalid, duplicate and won't fix bugs

            raju Raju Suravarjjala added a comment - Bulk closing invalid, duplicate and won't fix bugs

            People

              owend Daniel Owen
              richard.demellow Richard deMellow
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved:

                Gerrit Reviews

                  There are no open Gerrit changes

                  PagerDuty