[
https://issues.apache.org/jira/browse/SOLR-5652?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hoss Man updated SOLR-5652:
---------------------------
Attachment: SOLR-5652.nocommit.patch
Ok, this new patch has the following...
* new {{\*_dv}} fields in the schema for all the various types w/o using any of
the sort missing options
* tweaked the simple testing in both the single node and distrib test so that:
** one doc is missing an int value
** we randomly pick either "int" or "int_dv" as a field to use in explicit sorts
*** currently a nocommit in place to force this to be "int_dv"
** we explicitly sort on all 3 "missing" sub-variants ("", "_first", "_last")
and check the doc order exactly matches our expectations
* includes everything from SOLR-5652.codec.skip.dv.patch...
** ...but there is a nocommit bypassing hte codec check so docvalues are always
used.
With this patch, and these nocommits, it's pretty trivial to reliably reproduce
failing seeds that pop up when running...
{code}
ant test -Dtests.class=\*Cursor\* -Dtests.codec=Lucene40
{code}
...and likewise, my limted testing so far hasn't seen any failures when running
this patch with Lucene45 codec...
{code}
ant test -Dtests.class=\*Cursor\* -Dtests.codec=Lucene45
{code}
> 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.codec.skip.dv.patch,
> SOLR-5652.nocommit.patch,
> jenkins.thetaphi.de_Lucene-Solr-4.x-MacOSX_1200.log.txt,
> jenkins.thetaphi.de_Lucene-Solr-4.x-MacOSX_1217.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]
