[
https://issues.apache.org/jira/browse/SOLR-8611?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15529332#comment-15529332
]
Stephan Lagraulet edited comment on SOLR-8611 at 9/28/16 11:31 AM:
-------------------------------------------------------------------
This only occured on our production node when a massive recovery was forced on
all nodes of one shard.
Unfortunately we did not have a simple case to reproduce the problem.
Maybe a static analysis of the code should provide a hint to where to find the
bug?
was (Author: stephlag):
This only occured on our production node when a massive recovery (for all
nodes) was forced on all our nodes for one shard.
Unfortunately we did not have a simple case to reproduce the problem.
Maybe a static analysis of the code should provide a hint to where to find the
bug?
> Incorrect logging in ZkController
> ---------------------------------
>
> Key: SOLR-8611
> URL: https://issues.apache.org/jira/browse/SOLR-8611
> Project: Solr
> Issue Type: Bug
> Components: SolrCloud
> Affects Versions: 5.4
> Reporter: Stephan Lagraulet
> Priority: Minor
> Labels: logging
>
> When a new Zookeeper Session is created all cores publish a "down" status.
> Then a call to "waitForLeaderToSeeDownState" is made.
> Unfortunately, the logged info inside this method is not correct, keeping the
> last published core:
> {code}
> 2016-01-28 10:19:36.296 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [ ]
> o.a.s.c.ZkController ZooKeeper session re-connected ... refreshing core
> states after session expiration.
> 2016-01-28 10:19:36.296 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr)
> [c:offers_suggest_marketing s:shard1 r:core_node3
> x:offers_suggest_marketing_shard1_replica5] o.a.s.c.ZkController publishing
> state=down
> 2016-01-28 10:19:36.300 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr)
> [c:offers_suggest_topsearch s:shard1 r:core_node4
> x:offers_suggest_topsearch_shard1_replica2] o.a.s.c.ZkController publishing
> state=down
> 2016-01-28 10:19:36.301 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [c:offers_storage
> s:shard3 r:core_node8 x:offers_storage_shard3_replica2] o.a.s.c.ZkController
> publishing state=down
> 2016-01-28 10:19:36.302 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [c:offers_lean
> s:shard3 r:core_node6 x:offers_lean_shard3_replica2] o.a.s.c.ZkController
> publishing state=down
> 2016-01-28 10:19:36.305 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [c:offers_lean
> s:shard3 r:core_node6 x:offers_lean_shard3_replica2] o.a.s.c.ZkController
> Replica core_node3 NOT in leader-initiated recovery, need to wait for leader
> to see down state.
> 2016-01-28 10:19:36.313 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [c:offers_lean
> s:shard3 r:core_node6 x:offers_lean_shard3_replica2] o.a.s.c.ZkController
> Replica core_node4 NOT in leader-initiated recovery, need to wait for leader
> to see down state.
> 2016-01-28 10:19:36.317 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [c:offers_lean
> s:shard3 r:core_node6 x:offers_lean_shard3_replica2] o.a.s.c.ZkController
> Replica core_node8 NOT in leader-initiated recovery, need to wait for leader
> to see down state.
> {code}
> It should be
> {code}
> 2016-01-28 10:19:36.296 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [ ]
> o.a.s.c.ZkController ZooKeeper session re-connected ... refreshing core
> states after session expiration.
> 2016-01-28 10:19:36.296 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr)
> [c:offers_suggest_marketing s:shard1 r:core_node3
> x:offers_suggest_marketing_shard1_replica5] o.a.s.c.ZkController publishing
> state=down
> 2016-01-28 10:19:36.300 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr)
> [c:offers_suggest_topsearch s:shard1 r:core_node4
> x:offers_suggest_topsearch_shard1_replica2] o.a.s.c.ZkController publishing
> state=down
> 2016-01-28 10:19:36.301 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [c:offers_storage
> s:shard3 r:core_node8 x:offers_storage_shard3_replica2] o.a.s.c.ZkController
> publishing state=down
> 2016-01-28 10:19:36.302 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [c:offers_lean
> s:shard3 r:core_node6 x:offers_lean_shard3_replica2] o.a.s.c.ZkController
> publishing state=down
> 2016-01-28 10:19:36.305 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr)
> [c:offers_suggest_marketing s:shard1 r:core_node3
> x:offers_suggest_marketing_shard1_replica5] o.a.s.c.ZkController Replica
> core_node3 NOT in leader-initiated recovery, need to wait for leader to see
> down state.
> 2016-01-28 10:19:36.313 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr)
> [c:offers_suggest_topsearch s:shard1 r:core_node4
> x:offers_suggest_topsearch_shard1_replica2] o.a.s.c.ZkController Replica
> core_node4 NOT in leader-initiated recovery, need to wait for leader to see
> down state.
> 2016-01-28 10:19:36.317 INFO
> (zkCallback-3-thread-37-processing-n:node054:8983_solr) [c:offers_storage
> s:shard3 r:core_node8 x:offers_storage_shard3_replica2] o.a.s.c.ZkController
> Replica core_node8 NOT in leader-initiated recovery, need to wait for leader
> to see down state.
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
