[
https://issues.apache.org/jira/browse/SOLR-8933?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15235264#comment-15235264
]
Mike Drob commented on SOLR-8933:
---------------------------------
So it occurred to me that we _might_ be able to get away with wrapping the
{{HttpRequestContentStream}} in a {{CloseShieldInputStream}} for minimal
changes, assuming that everything else in the code base behaves consistently by
reading the content stream and not the stream from the request, and then let
various handlers/parsers/etc. still try to close the stream when they are done.
> SolrDispatchFilter::consumeInput logs "Stream Closed" IOException
> -----------------------------------------------------------------
>
> Key: SOLR-8933
> URL: https://issues.apache.org/jira/browse/SOLR-8933
> Project: Solr
> Issue Type: Bug
> Affects Versions: 4.10.3
> Reporter: Mike Drob
> Assignee: Mark Miller
> Attachments: SOLR-8933.patch, SOLR-8933.patch, SOLR-8933.patch,
> SOLR-8933.patch, 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]
