[
https://issues.apache.org/jira/browse/SOLR-8933?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15228467#comment-15228467
]
Mike Drob commented on SOLR-8933:
---------------------------------
bq. Have we ever actually seen that happen with Jetty or on trunk, or have we
still only seen it with a version of 4.10.3 and Tomcat?
I've only seen it on 4.10.3 and Tomcat. As [~thetaphi] alludes to later, Jetty
might be taking care of this for us.
bq. Also, are we sure this patch catches all the spots this can happen on
trunk? How did you track down each spot the streams are closed early?
I wrapped the request and intercepted requests to close() and logged them with
stack traces. I had removed this for the patch because I wasn't confident in
the performance impact that I would be causing, but can easily add it back.
> SolrDispatchFilter::consumeInput logs "Stream Closed" IOException
> -----------------------------------------------------------------
>
> Key: SOLR-8933
> URL: https://issues.apache.org/jira/browse/SOLR-8933
> Project: Solr
> Issue Type: Bug
> Reporter: Mike Drob
> Assignee: Mark Miller
> Attachments: SOLR-8933.patch, SOLR-8933.patch
>
>
> After SOLR-8453 we started seeing some IOExceptions coming out of
> SolrDispatchFilter with "Stream Closed" messages.
> It looks like we are indeed closing the request stream in several places when
> we really need to be letting the web container handle their life cycle. I've
> got a preliminary patch ready and am working on testing it to make sure there
> are no regressions.
> A very strange piece of this is that I have been entirely unable to reproduce
> it on a unit test, but have seen it on cluster deployment quite consistently.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
