[
https://issues.apache.org/jira/browse/SOLR-8129?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14972759#comment-14972759
]
Mark Miller edited comment on SOLR-8129 at 10/24/15 5:35 PM:
-------------------------------------------------------------
Seems the above happens because we only have one Solr instance and nothing to
kill. The logic to make sure we have at least 2 is is not correct and needs to
be fixed.
{code}
// we make sure that there's at least one shard with more than one replica
// so that the ChaosMonkey has something to kill
numShards = sliceCount + random().nextInt(TEST_NIGHTLY ? 12 : 2) + 1;
{code}
should be
{code}
// we make sure that there's at least one shard with more than one replica
// so that the ChaosMonkey has something to kill
numShards = sliceCount + random().nextInt(TEST_NIGHTLY ? 12 : 2) + 2;
{code}
* That's actually probably not the right fix it looks - but something like this.
was (Author: [email protected]):
Seems the above happens because we only have one Solr instance and nothing to
kill. The logic to make sure we have at least 2 is is not correct and needs to
be fixed.
{code}
// we make sure that there's at least one shard with more than one replica
// so that the ChaosMonkey has something to kill
numShards = sliceCount + random().nextInt(TEST_NIGHTLY ? 12 : 2) + 1;
{code}
should be
{code}
// we make sure that there's at least one shard with more than one replica
// so that the ChaosMonkey has something to kill
numShards = sliceCount + random().nextInt(TEST_NIGHTLY ? 12 : 2) + 2;
{code}
> HdfsChaosMonkeyNothingIsSafeTest failures
> -----------------------------------------
>
> Key: SOLR-8129
> URL: https://issues.apache.org/jira/browse/SOLR-8129
> Project: Solr
> Issue Type: Bug
> Reporter: Yonik Seeley
> Attachments: fail.151005_064958, fail.151005_080319
>
>
> New HDFS chaos test in SOLR-8123 hits a number of types of failures,
> including shard inconsistency.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
