[
https://issues.apache.org/jira/browse/SOLR-5652?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13880659#comment-13880659
]
Steve Rowe commented on SOLR-5652:
----------------------------------
In the trial #129 failure, the second (failing) page is sorted wrong - the sort
field is long_dv_first, but multiple docs with missing values (including the
"walk already seen" doc: id=53) are not sorted first. Below is the output of
{{egrep 'SOLR-5652: \[\^(]|id=53'}} over {{129.log}} - note that the first page
has the id=53 doc sorted properly:
{noformat}
[junit4] 2> 21647 T10 oasc.DistribCursorPagingTest.assertFullWalkNoDups
SOLR-5652:
({params(q=*%3A*&fl=id%2Clong_dv_first&rows=24&sort=long_dv_first+desc%2C+id+desc),defaults(cursorMark=*)})
gave us these docs: {id=53}; {id=25, long_dv_first=9028609258260341288};
{id=1, long_dv_first=8079898934015269594}; {id=21,
long_dv_first=7851701882515113803}; {id=62, long_dv_first=7838548354468740764};
{id=15, long_dv_first=7703783121293450688}; {id=77,
long_dv_first=7259374194906229471}; {id=59, long_dv_first=7109412730326436803};
{id=29, long_dv_first=6983226518504627711}; {id=69,
long_dv_first=6894002903936490262}; {id=108,
long_dv_first=6425678483825374576}; {id=48, long_dv_first=6162850160620829991};
{id=18, long_dv_first=6050872288145443429}; {id=10,
long_dv_first=5639159091440117036}; {id=49, long_dv_first=5585109074321704897};
{id=121, long_dv_first=5564641483351446975}; {id=73,
long_dv_first=5557697725656376716}; {id=124,
long_dv_first=5515795054987553907}; {id=91, long_dv_first=5446634527267601359};
{id=66, long_dv_first=5412453097925579481}; {id=57,
long_dv_first=5391575838257268802}; {id=67, long_dv_first=5285225733925278855};
{id=76, long_dv_first=5196111855800982134}; {id=118,
long_dv_first=4962434967015612730};
[junit4] 2> 21673 T10 oasc.DistribCursorPagingTest.assertFullWalkNoDups
SOLR-5652:
({params(q=*%3A*&fl=id%2Clong_dv_first&rows=24&sort=long_dv_first+desc%2C+id+desc),defaults(cursorMark=AoJ%2FvAJZAg%3D%3D)})
gave us these docs: {id=94, long_dv_first=5064}; {id=74, long_dv_first=5045};
{id=100, long_dv_first=5038}; {id=95, long_dv_first=5024}; {id=71,
long_dv_first=5010}; {id=117}; {id=53}; {id=46}; {id=99,
long_dv_first=-57033495908952210}; {id=84, long_dv_first=-369682029177105515};
{id=51, long_dv_first=-571619284359424369}; {id=33,
long_dv_first=-777789016127771765}; {id=89, long_dv_first=-827568794981027330};
{id=27, long_dv_first=-874200373569477439}; {id=32,
long_dv_first=-897758339419136464}; {id=37,
long_dv_first=-1097686452252701951}; {id=20,
long_dv_first=-1142352694920116441}; {id=60,
long_dv_first=-1237791953129111727}; {id=78,
long_dv_first=-1381169508523349434}; {id=58,
long_dv_first=-1427953974149921646}; {id=104,
long_dv_first=-1433781337878484605}; {id=8,
long_dv_first=-1778050783003838576}; {id=55,
long_dv_first=-1879894088937473264}; {id=38,
long_dv_first=-1955827817437289663};
{noformat}
> 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,
> jenkins.thetaphi.de_Lucene-Solr-4.x-MacOSX_1200.log.txt,
> jenkins.thetaphi.de_Lucene-Solr-4.x-MacOSX_1217.log.txt
>
>
> Twice 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 (in 3 failures):
> * So far only seen on http://jenkins.thetaphi.de & sarowe's mac
> * So far only seen on MacOSX
> * So far only seen on branch 4x
> * So far seen on both Java6 and Java7
> * 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 both when doing a "desc" sorting on one of the "\*_dv_last" or
> "\*_dv_first" fields (docValues=true, either sortMissingLast=true OR
> sortMissingFirst=true)
> ** sort on same field asc has always 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: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
