[
https://issues.apache.org/jira/browse/SOLR-4401?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13588947#comment-13588947
]
Erick Erickson commented on SOLR-4401:
--------------------------------------
I'm going to try to put this into a junit test over the weekend. It can run
really long or really short. I'm thinking of providing three methods, one that
runs all the time and takes 30 seconds? a minute? Label it as "slow" either way.
A nightly that runs for 10 minutes
A weekly that runs for an hour
The background here is that I see issues after 1-2 hours running on my machine,
occasionally as little as 30 minutes. I suppose if it runs more often it'd have
a better chance of hitting at shorter intervals. I'm open to suggestions, so
let me know.... I really don't have a sense of how onerous a really
long-running test would be....
Mark:
When I mentioned "prior art" it was mostly a shorthand to myself that there
seemed to be code I could use as a model in TestReplicationHandler in terms of
starting up servers and all that rot. But I'll look at JettySolrRunner as you
indicated, thanks for the pointer.
> Move the stress test in SOLR-4196 into a junit test
> ---------------------------------------------------
>
> Key: SOLR-4401
> URL: https://issues.apache.org/jira/browse/SOLR-4401
> Project: Solr
> Issue Type: Test
> Affects Versions: 4.2, 5.0
> Reporter: Erick Erickson
> Assignee: Erick Erickson
> Priority: Minor
> Fix For: 4.2, 5.0
>
> Attachments: StressTest.zip
>
>
> As part of SOLR-4196, I created a stress test proces for rapidly opening and
> closing cores. It'd probably be useful to make this into a junit test that
> ran nightly (it needs some time in order to show anything, as in minutes).
> Typical failures are 20 minutes into the run, but occasionally they're faster.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
