[jira] [Resolved] (SOLR-17437) Shard splits no longer check for sufficient disk space

["Matthew Biscocho (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/SOLR-17437?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Matthew Biscocho resolved SOLR-17437.
-------------------------------------
    Fix Version/s: 10.1
       Resolution: Fixed

> Shard splits no longer check for sufficient disk space
> ------------------------------------------------------
>
>                 Key: SOLR-17437
>                 URL: https://issues.apache.org/jira/browse/SOLR-17437
>             Project: Solr
>          Issue Type: Bug
>          Components: SolrCloud
>    Affects Versions: 9.3
>            Reporter: David Smiley
>            Priority: Minor
>              Labels: newdev
>             Fix For: 10.1
>
>
> SolrCloud shard splits check for sufficient disk space. After the SOLR-16507 refactoring, the metrics request is no longer routed to the pertinent node; instead it's evaluated at some arbitrary node or maybe the Overseer. It doesn't find the size because it can't find the replica on that node (unless it actually is there). Then logs a warning: "cannot verify information for parent shard leader". This behavior can be verified by modifying SplitShardTest to use a nodeCount of 2 and then look for this warning.
> Maybe we should add an "assert false" here in addition to the fix. IMO doesn't need a dedicated test but SplitShardTest could use an additional node for nightly runs.
> -When this is fixed, probably use SolrCloudManager after SOLR-16503. If that has not been done yet (as of this writing is not), then the current way is to call SolrRequest.setBasePath.-



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@solr.apache.org
For additional commands, e-mail: issues-help@solr.apache.org