Details
-
Bug
-
Resolution: Fixed
-
Blocker
-
4.5.1
-
None
-
Untriaged
-
Centos 64-bit
-
Yes
Description
Build: 4.5.1-2829.
Steps:
1. Create a cluster with 2 nodes.
2. Create an empty non-default bucket with SASL auth.
{
|
"authType": "sasl",
|
"bucketType": "membase",
|
"evictionPolicy": "valueOnly",
|
"flushEnabled": 1,
|
"name": "bucket-1",
|
"ramQuotaMB": 51200,
|
"replicaIndex": 0,
|
"replicaNumber": 1,
|
"saslPassword": "password"
|
}
|
3. Restart one of the servers.
# time service couchbase-server restart
|
Restarting couchbase-server (via systemctl): [ OK ]
|
|
real 5m0.471s
|
user 0m0.036s
|
sys 0m0.057s
|
It's not really OK. In fact, the stop task just timed out:
# systemctl show couchbase-server.service | grep Timeout
|
TimeoutStartUSec=5min
|
TimeoutStopUSec=5min
|
Because of that the "start" part of the "restart" task had no affect:
# service couchbase-server status
|
couchbase-server is not running
|
Attachments
Issue Links
For Gerrit Dashboard: MB-20716 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
67169,4 | MB-20716: Ensure DCP connections in EWOULDBLOCK are unpaused on bucket delete | watson | ep-engine | Status: MERGED | +2 | +1 |
67195,2 | Merge remote-tracking branch 'couchbase/watson' into master | master | ep-engine | Status: MERGED | +2 | +1 |
67252,4 | MB-20751: Fix lock cycle (deadlock) during bucket delete & client disconnect | watson | ep-engine | Status: MERGED | +2 | +1 |
67344,2 | MB-20751: Fix lock cycle (deadlock) during bucket delete & client disconnect | master | ep-engine | Status: ABANDONED | -1 | +1 |
67375,6 | MB-20716: Ensure DCP consumers in EWOULDBLOCK are unpaused on bucket delete | watson | ep-engine | Status: MERGED | +2 | +1 |
67433,2 | Merge remote-tracking branch 'couchbase/watson' into master | master | ep-engine | Status: MERGED | +2 | +1 |