[
https://issues.apache.org/jira/browse/SOLR-18002?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18054226#comment-18054226
]
ASF subversion and git services commented on SOLR-18002:
--------------------------------------------------------
Commit b6983892ba3b6d129a766f07b5377c24b523de56 in solr's branch
refs/heads/branch_10x from jvanneman
[ https://gitbox.apache.org/repos/asf?p=solr.git;h=b6983892ba3 ]
SOLR-18002: idle timeouts should cause servers to be added to the zombie list
(#3891)
CloudSolrClient/LBSolrClient should consider a retry-able request that times
out as another condition to internally mark that replica as a "zombie".
Previously, unresponsive servers continued to receive traffic and high client
latencies as the idle timeout is consistently triggered on every request to
that replica.
> LBHttp2SolrClient doesn't add unresponsive servers to the zombie list
> ---------------------------------------------------------------------
>
> Key: SOLR-18002
> URL: https://issues.apache.org/jira/browse/SOLR-18002
> Project: Solr
> Issue Type: Bug
> Components: SearchComponents - other
> Affects Versions: 9.10
> Reporter: James Vanneman
> Priority: Major
> Labels: pull-request-available
> Time Spent: 2h 50m
> Remaining Estimate: 0h
>
> The exception that's triggered when the idle timeout is exceeded is a
> TimeoutException which does not inherit from IOException and fails the
> IOException check which would add the unresponsive server to the zombie list.
> The result is that unresponsive servers continue to receive requests instead
> of being added to the zombie list and skipped.
>
> The expected behavior is that unresponsive servers that exceed the idle
> timeout should be marked unhealthy and added to the zombie list.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
