Apache
/
lucene
mirror of https://github.com/apache/lucene.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

SOLR-6511: Better handling of ZooKeeper related exceptions when deciding to start the leader-initiated recovery nag thread 

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1629720 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Timothy Potter 2014-10-06 17:50:49 +00:00
parent ac7621e607
commit d308344af8
1 changed files with 9 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
solr/core/src/java/org/apache/solr/update/processor/DistributedUpdateProcessor.java
View File

@ -868,17 +868,16 @@ public class DistributedUpdateProcessor extends UpdateRequestProcessor {
if (sendRecoveryCommand) {
maxTries = 120;
} // else the node is no longer "live" so no need to send any recovery command
} catch (KeeperException.SessionExpiredException see) {
} catch (Exception exc) {
Throwable setLirZnodeFailedCause = SolrException.getRootCause(exc);
log.error("Leader failed to set replica " +
error.req.node.getUrl() + " state to DOWN due to: " + see, see);
// our session is expired, which means our state is suspect, so don't go
// putting other replicas in recovery (see SOLR-6511)
sendRecoveryCommand = false;
} catch (Exception e) {
log.error("Leader failed to set replica " +
error.req.node.getUrl() + " state to DOWN due to: " + e, e);
// will go ahead and try to send the recovery command once after this error
error.req.node.getUrl() + " state to DOWN due to: " + setLirZnodeFailedCause, setLirZnodeFailedCause);
if (setLirZnodeFailedCause instanceof KeeperException.SessionExpiredException ||
setLirZnodeFailedCause instanceof KeeperException.ConnectionLossException) {
// our session is expired, which means our state is suspect, so don't go
// putting other replicas in recovery (see SOLR-6511)
sendRecoveryCommand = false;
} // else will go ahead and try to send the recovery command once after this error
}
} else {
// not the leader anymore maybe or the error'd node is not my replica?