Details
-
Bug
-
Resolution: Cannot Reproduce
-
Critical
-
2.1.0
-
Security Level: Public
-
SLES11 SP2, Xen virtual machines with a single processor and 1.5GB RAM
Description
Servers M1 and M2 are each running couchbase (C1 and C2 respectively, both added to the same cluster via the web interface) and an application (A1 and A2 respectively). A1 is configured to primarily access C2 and to fail over to C1 if C2 becomes unavailable. A2 is similarly configured, only accessing C1 primarily and failing over to C2.
When accessing A1, Couchbase reports that the value is stored on C2 and has replicated to C1, howerver while accessing A2, Couchbase reports that the value is stored on C1 and has not replicated to C2. Additionally, when attempting to manually rebalance the cluster, the rebalance fails (the log output is the attached image). After consulting the IRC channel, user ingenthr suggested I file this as an issue.
I've also attached the output of cbcollect_info in the hopes that it might contain something useful. If you need more information, do not hesitate to ask.
Attachments
Issue Links
- relates to
-
MB-9596 [system test] rebalance failed with error {{{badmatch,[{<0.6316.374>,noproc}]}, [{misc,sync_shutdown_many_i_am_trapping_exits, 1},
- Closed
For Gerrit Dashboard: MB-7943 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
25710,2 | MB-7943: Log too many connections as warning level | branch-20 | memcached | Status: MERGED | +2 | +1 |
25718,1 | Merge remote-tracking branch 'membase/branch-20' | master | memcached | Status: MERGED | +2 | +1 |