Couchbase Server

Performance Status

04/30/2014

General News


  • Mobile performance testing automation

NEW tickets

  • MB-10943: Erlang memory usage goes up to 60GB in XDCR setups [was: Source node auto failed over during initial data load with XDCR]
  • MB-10956: Rebalance-in with views takes 10 hours (used to be 70 minutes)
  • MB-10959: Rebalance exited {noproc, {gen_server,call, [{'janitor_agent-bucket-1','ns_1@172.23.96.16'}, {if_rebalance,<0.26548.32>, {update_vbucket_state,933,active,undefined, undefined}}, infinity]}}
  • MB-11004: Rebalance with views fails due to "wait_checkpoint_persisted_failed"

STILL NOT CLOSED


  • MB-10908:  beam.smp RSS grows to 50GB during delta recovery causing OOM killer invocation and rebalance failure
  • MB-10533: Compaction doesn't make any progress in spite of high fragmentation
  • MB-10437: XDCR replication rate drops almost to zero in presence of light write workload on src side
  • MB-10370:  ep-engine deadlock in write-heavy DGM cases 
  • MB-10273:  View compaction doesn't catch up in basic non-DGM tests with view queries

2.X DEBT


  • MB-9620: (or just kill mccouch) multi-tenancy: beam.smp memory usage optimization for non-views cases
  • MB-9930: regression in memory fragmentation in tcmalloc with appends ops(2.5.0v2.2.0)
  • MB-9676: Dramatically increasing latency of SET operations during rebalance tests
  • MB-9637: Rebalance-In 5 nodes on empty 18 buckets is very slow per bucket

2.X DEBT


  • MB-9822: One of nodes is too slow during indexing
  • MB-9825: Rebalance exited with reason bad_replicas
  • MB-9461: In heavy-DGM (<5%) scenarios with views and high cache miss rate rebalance/client operations can fail due to timed out requests to memcached/ep-engine
  • MB-10191: CouchDB crashed due to 'Cannot allocate 467078560 bytes of memory (of type "heap").'
  • MB-7907: Issues when scaling XDCR on single node
  • n1ql issues


    • MB-10948: cbq-engine crashes under very light load

    BACKLOG


    • [ongoing] Query performance testing

    BLOCKERS


    • MB-10472:  Handle multiple snapshots properly in view engine
    • MB-10875:  Flusher queue doesn't get flushed
    • CBD-1239:  missing Windows builds
    • CBIT-1182: client machines for regression tests
    • CBIT-1158: RAID 10 drives for performance testing

    Performance Status

    By Pavel Paulau

    Performance Status

    • 209