[
https://issues.apache.org/jira/browse/TINKERPOP-2248?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16866816#comment-16866816
]
ASF GitHub Bot commented on TINKERPOP-2248:
-------------------------------------------
spmallette commented on pull request #1146: TINKERPOP-2248 Force replacement of
connections on certain errors
URL: https://github.com/apache/tinkerpop/pull/1146
https://issues.apache.org/jira/browse/TINKERPOP-2248
Relying only on the `Connection.isDead()` check to determine connection
replacement seemed to introduce a regression as of
2cd84ff1f3944d67d44cbba5bb032a1c57377975 The `shouldBlockRequestWhenTooBig` was
failing somewhat randomly since that change. Not sure if there is a better way
to make `isDead()` be the only check needed for connection replacement in this
context, but this change brings the logic back that checks exception types that
were forcing replacement prior to this change.
@divijvaidya would you mind having a look at this change? perhaps you are
aware of better solutions or perhaps I misinterpreted something in my analysis
that you could correct?
All tests pass with `docker/build.sh -t -i`
VOTE +1
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Instability of driver for blocked requests
> ------------------------------------------
>
> Key: TINKERPOP-2248
> URL: https://issues.apache.org/jira/browse/TINKERPOP-2248
> Project: TinkerPop
> Issue Type: Bug
> Components: driver
> Affects Versions: 3.3.7
> Reporter: stephen mallette
> Assignee: stephen mallette
> Priority: Critical
>
> I'm not sure what's causing this, because there haven't been changes to this
> area of code that would likely have an ill effect, but the
> {{shouldBlockRequestWhenTooBig}} integration test is failing with some
> consistency which is making travis unhelpful.
> https://api.travis-ci.org/v3/job/546837334/log.txt
> I'm thinking that it might be related to:
> https://github.com/apache/tinkerpop/commit/2cd84ff1f3944d67d44cbba5bb032a1c57377975
> as a blocked request of the type in that test will throw a
> {{CorruptedFrameException}} and this commit has something to do with that
> specifically.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)