Couchbase Server

Performance Status

07/23/2014

Bug Snapshot

Last Wednesday:

Blockers: 3
Critical: 6
Major: 2

Today:

Test Blockers: 1
Blockers: 7
Critical: 5
Major: 2

KV

MB-11801: It takes almost 2x more time to rebalance 10 empty buckets

MB-11799: Bucket compaction causes massive slowness of UPR consumers

MB-11794: Creating 10 buckets causes memcached segmentation fault

MB-11779: Memory underflow in updates-only scenario with 5 buckets

MB-11768: movement of 27 empty replica vbuckets gets stuck in seqnoWaiting

KV

MB-11738: Evaluate GIO CPU utilization on systems with 16 vCPU

MB-11731: Persistence to disk suffers from bucket compaction

MB-11729: CPU utilization by memcached suddenly increases (~2-3x) after couple hours, IO operations affected

MB-11675: 40-50% performance degradation on append-heavy workload compared to 2.5.1

MB-11642: Intra-replication falling far behind under moderate-heavy workload

KV

MB-11582: tmp_oom errors due to spike in memory usage during delta recovery

MB-11405: Shared thread pool: high CPU overhead due to OS level context switches / scheduling

Views

MB-11752: Rebalance from 3 to 4 nodes [with views] fails due to "{unexpected_exit, ..., {bulk_set_vbucket_state_failed..."

MB-11728: Improve stale=update_after query latency

XDCR

MB-11770: Re-investigate impact of xdcrMaxConcurrentReps on XDCR latency (LAN/WAN) and choose safer value for 3.0
MB-11643: Incoming workload suffers when XDCR enabled
MB-11435: 1500-2000% CPU utilization by beam.smp on source nodes in XDCR scenarios (was ~500% in 2.5.x)
MB-11434: 600-800% CPU utilization by memcached on source nodes in XDCR scenarios (was <100% in 2.5.x)

MISC

MB-11798: Operations to create user and create session in Sync Gateway are slower with 3.0 Beta

Couchbase Server