ns-diag-20120821005704.txt.gz
[MB-6342] User-triggered compaction of view `default/_design/dev_view4` failed: shutdown, when rebalancing out a server node Created: 21/Aug/12 Updated: 09/Jan/13 Resolved: 23/Aug/12 |
|
| Status: | Closed |
| Project: | Couchbase Server |
| Component/s: | ns_server, view-engine |
| Affects Version/s: | 2.0-beta |
| Fix Version/s: | 2.0 |
| Security Level: | Public |
| Type: | Bug | Priority: | Minor |
| Reporter: | Karan Kumar | Assignee: | Aliaksey Artamonau |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: | CentOS 64 bit | ||
| Attachments: |
ns-diag-20120821005704.txt.gz
|
| Description |
|
1) Swap rebalance one node in and out 2) Trigger compaction for production view We see the following error on all the nodes in the cluster:- User-triggered compaction of view `default/_design/dev_view4` failed: shutdown. See logs for detailed reason. We should probably ignore bucket shutdown error. Along the same lines as http://www.couchbase.com/issues/browse/MB-6278 [ns_server:warn,2012-08-21T0:57:56.407,ns_1@10.3.121.16:<0.15895.9>:compaction_daemon:do_chain_compactors:520]Compactor for view `default-s/_design/dev_view1` (pid [{type,view}, {name, <<"default-s/_design/dev_view1">>}, {important,false}, {fa, {#Fun<compaction_daemon.21.91655267>, [<<"default-s">>, <<"_design/dev_view1">>, {config, {30, 18446744073709551616}, {30, 18446744073709551616}, undefined,false, {daemon_config,30, 131072}}, false,bucket]}}]) terminated unexpectedly (ignoring this): {updater_died, {updater_error, shutdown}} [ns_server:warn,2012-08-21T0:17:30.366,ns_1@10.3.121.16:<0.15369.9>:compaction_daemon:do_chain_compactors:515]Compactor for view `default/_design/dev_view4/main` (pid [{type,view}, {important,true}, {name, <<"default/_design/dev_view4/main">>}, {fa, {#Fun<compaction_daemon.16.5288869>, [<<"default">>, <<"_design/dev_view4">>, main, {config, {30, 18446744073709551616}, {30, 18446744073709551616}, undefined,false, {daemon_config, 30,131072}}, true, {view, <<"_design/dev_view4">>}]}}]) terminated unexpectedly: shutdown |
| Comments |
| Comment by Filipe Manana [ 21/Aug/12 ] |
| This is normal to happen during rebalance and can safelly be ignored. |
| Comment by Filipe Manana [ 21/Aug/12 ] |
| At The very best, ns_ server compaction daemon can ignore this particular exit reason and retry immediately or leave it for next iteration. |
| Comment by Aliaksey Artamonau [ 21/Aug/12 ] |
| The problem with user-triggered compactions here is that in case something goes wrong we show the error in the UI log. So it might be confusing for the user to see this kind of errors. But it's not trivial to ignore those because the actual error can look differently. For instance, in one case above it's just 'shutdown', in the other {updater_died, {updater_error, shutdown}}. There can be other variants probably. |
| Comment by Filipe Manana [ 21/Aug/12 ] |
|
Aliaksey, the only expected, harmless exit reasons, are:
shutdown {updater_died, shutdown} {updater_died, noproc} |
| Comment by Karan Kumar [ 21/Aug/12 ] |
| Thanks. I would believe these would still get logged in ns_server logs and be not UI visible? |
| Comment by Thuan Nguyen [ 22/Aug/12 ] |
|
Integrated in github-ns-server-2-0 #448 (See [http://qa.hq.northscale.net/job/github-ns-server-2-0/448/]) Result = SUCCESS Aliaksey Kandratsenka : Files : * src/compaction_daemon.erl |