[
https://issues.apache.org/jira/browse/SOLR-6099?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18132153#comment-18132153
]
Nick Shanin commented on SOLR-6099:
-----------------------------------
>From code reading against apache/main b73f03976fa, DELETESHARD now accepts
>CONSTRUCTION and RECOVERY sub-shards
>(solr/core/src/java/org/apache/solr/cloud/api/collections/DeleteShardCmd.java:84-88),
> but a sub-shard in RECOVERY_FAILED, which split sets on failure
>(solr/core/src/java/org/apache/solr/cloud/api/collections/SplitShardCmd.java:666),
> is still rejected there. A re-split after a failed split therefore still
>appears to fail. Is this still a problem on a current release?
🤖 AI-generated 🤖
> Fix cleanup mechanism for a previous (failed) SPLITSHARD
> --------------------------------------------------------
>
> Key: SOLR-6099
> URL: https://issues.apache.org/jira/browse/SOLR-6099
> Project: Solr
> Issue Type: Bug
> Components: SolrCloud
> Reporter: Anshum Gupta
> Priority: Major
>
> Right now, SPLITSHARD tries to cleanup an under-construction/recovery shard
> using the deleteshard API if it already exists. DELETESHARD on the other hand
> will never delete a shard that's not INACTIVE/uses implicit routing.
> This would just raise an exception and never really delete a shard right now.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
