[
https://issues.apache.org/jira/browse/SOLR-5652?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Steve Rowe updated SOLR-5652:
-----------------------------
Attachment:
SOLR-5652-dont-sort-on-any-dv-fields-when-codec-doesnt-support-missing-dvs.patch
str_dv.post-patch.log.txt
I've attached a log from another failure (trunk, Linux, trial #90), with a
complex sort with {{str_dv}} as the primary, where the "walk already seen" doc
is missing its {{str_dv}} value, and the codec is Lucene40.
Hoss's {{bin_dv}} sort failure was on a doc missing its {{bin_dv}} value, and
the codec is Lucene42.
I'm pretty sure both these failures are evidence of another test bug.
Hoss set up the full walk tests to skip sorting on {{dv_first}} and {{dv_last}}
fields when the codec doesn't support missing docvalues, but {{dv}} fields
should not be sorted on in that case either, since they too may have missing
values (and in both the above cases, that's what caused the failures).
I've attached another patch that also skips sorting on {{dv}} fields.
> Heisenbug in DistribCursorPagingTest: "walk already seen ..."
> -------------------------------------------------------------
>
> Key: SOLR-5652
> URL: https://issues.apache.org/jira/browse/SOLR-5652
> Project: Solr
> Issue Type: Bug
> Reporter: Hoss Man
> Assignee: Hoss Man
> Attachments: 129.log, 372.log,
> SOLR-5652-dont-sort-on-any-dv-fields-when-codec-doesnt-support-missing-dvs.patch,
> SOLR-5652.codec.skip.dv.patch, SOLR-5652.nocommit.patch, SOLR-5652.patch,
> bin_dv.post-patch.log.txt,
> jenkins.thetaphi.de_Lucene-Solr-4.x-MacOSX_1200.log.txt,
> jenkins.thetaphi.de_Lucene-Solr-4.x-MacOSX_1217.log.txt,
> str_dv.post-patch.log.txt
>
>
> Several times now, Uwe's jenkins has encountered a "walk already seen ..."
> assertion failure from DistribCursorPagingTest that I've been unable to
> fathom, let alone reproduce (although sarowe was able to trigger a similar,
> non-reproducible seed, failure on his machine)
> Using this as a tracking issue to try and make sense of it.
> Summary of things noticed so far:
> * So far only seen on http://jenkins.thetaphi.de & sarowe's mac
> * So far seen on MacOSX and Linux
> * So far seen on branch 4x and trunk
> * So far seen on Java6, Java7, and Java8
> * fails occured in first block of randomized testing:
> ** we've indexed a small number of randomized docs
> ** we're explicitly looping over every field and sorting in both directions
> * fails were sorting on one of the "\*_dv_last" or "\*_dv_first" fields
> (docValues=true, either sortMissingLast=true OR sortMissingFirst=true)
> ** for desc sorts, sort on same field asc has worked fine just before this
> (fields are in arbitrary order, but "asc" always tried before "desc")
> ** sorting on some other random fields has sometimes been tried before this
> and worked
> (specifics of each failure seen in the wild recorded in comments)
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
