Details
-
Bug
-
Resolution: Fixed
-
Critical
-
3.0-Beta, 3.0
-
Security Level: Public
-
None
-
CentOS, CB 3.0.0 build 918
-
Untriaged
-
Centos 64-bit
-
Unknown
Description
Created 3-node cluster (1GB RAM, 2 CPUs), with 4 buckets. Put one bucket under modest cbworkloadgen workload (10,000 items, 13k op/s).
Selected "Graceful Failover" for one of the nodes. Failover started, got to ~55% then just paused (see screenshot). Waited for maybe 30mins but no progress.
Logs uploaded (via "Collect Information"
https://s3.amazonaws.com/customers.couchbase.com/daver_graceful_failover_hung/collectinfo-2014-07-11T144818-ns_1%40192.168.73.101.zip
https://s3.amazonaws.com/customers.couchbase.com/daver_graceful_failover_hung/collectinfo-2014-07-11T144818-ns_1%40192.168.73.102.zip
https://s3.amazonaws.com/customers.couchbase.com/daver_graceful_failover_hung/collectinfo-2014-07-11T144818-ns_1%40192.168.73.103.zip
Attachments
Issue Links
For Gerrit Dashboard: MB-11706 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
39348,3 | MB-11706 Fix the race condition during exit | master | couchdb | Status: MERGED | +2 | +1 |
39588,20 | CBQE-2336:: add additional tests for views+graceful failover, MB-11706 | master | testrunner | Status: MERGED | +2 | +1 |