[
https://issues.apache.org/jira/browse/SOLR-15892?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17468865#comment-17468865
]
Ankur Gupta commented on SOLR-15892:
------------------------------------
Yes, the current code is selecting a fixed shard (the first in the list) for
the delete and commit requests to a RoutedAlias. and in my solution code, I am
suggesting randomizing the shard selection to fix the "fixed" shard selection
for such requests.
Yes, you are right, we've observed that the current code is overburdening one
particular node when there's a large number of delete requests.
> RoutedAliasUpdateProcessor not distributing delete and commit requests.
> -----------------------------------------------------------------------
>
> Key: SOLR-15892
> URL: https://issues.apache.org/jira/browse/SOLR-15892
> Project: Solr
> Issue Type: Bug
> Security Level: Public(Default Security Level. Issues are Public)
> Affects Versions: 8.1, 8.2, 8.3, 8.4, 8.5, 8.6, 8.7, 8.8, main (9.0), 8.9,
> 8.10, 8.11
> Reporter: Ankur Gupta
> Priority: Minor
> Fix For: main (9.0), 8.11
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Currently, the RoutedAliasUpdateProcessor is not distributing and forwarding
> all delete and commit requests for a RoutedAlias to a fixed shard leader.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
