[
https://issues.apache.org/jira/browse/SOLR-5652?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13880735#comment-13880735
]
Steve Rowe commented on SOLR-5652:
----------------------------------
Uh, the trial #129 failure isn't just sorted wrong on the second page, the
whole thing is sorted wrong: all the missing-value docs should be on the first
page. On the second page, the three missing value docs are sorted as if they
had the value 0.
The same thing happened on trial #25 - on the second page, missing value doc
id=53 is sorted as if its int_dv_last value were zero, rather than missing, and
then it shows up again as the final doc on the third page, where it belongs:
{noformat}
[junit4] 2> 19917 T10 oasc.DistribCursorPagingTest.assertFullWalkNoDups
SOLR-5652:
({params(q=*%3A*&fl=id%2Cint_dv_last&rows=50&sort=int_dv_last+desc%2C+id+desc),defaults(cursorMark=*)})
gave us these docs: {id=115, int_dv_last=2091290595}; {id=113,
int_dv_last=2037908394}; {id=41, int_dv_last=2010364276}; {id=55,
int_dv_last=1998540267}; {id=61, int_dv_last=1928174044}; {id=10,
int_dv_last=1926002179}; {id=6, int_dv_last=1860315269}; {id=68,
int_dv_last=1813333298}; {id=72, int_dv_last=1788709893}; {id=94,
int_dv_last=1775138667}; {id=103, int_dv_last=1759722924}; {id=19,
int_dv_last=1708818408}; {id=62, int_dv_last=1702707874}; {id=95,
int_dv_last=1575599582}; {id=18, int_dv_last=1539900646}; {id=3,
int_dv_last=1534894477}; {id=16, int_dv_last=1513487396}; {id=122,
int_dv_last=1512368828}; {id=91, int_dv_last=1490726417}; {id=100,
int_dv_last=1369044665}; {id=129, int_dv_last=1366930639}; {id=2,
int_dv_last=1335797630}; {id=12, int_dv_last=1311171495}; {id=17,
int_dv_last=1197587669}; {id=71, int_dv_last=1075041470}; {id=14,
int_dv_last=1062057736}; {id=87, int_dv_last=1062050843}; {id=98,
int_dv_last=1026760294}; {id=32, int_dv_last=987606417}; {id=48,
int_dv_last=971706716}; {id=25, int_dv_last=860420389}; {id=58,
int_dv_last=844256726}; {id=120, int_dv_last=784821778}; {id=26,
int_dv_last=778495105}; {id=89, int_dv_last=759861936}; {id=42,
int_dv_last=731134129}; {id=38, int_dv_last=507728088}; {id=66,
int_dv_last=504369368}; {id=88, int_dv_last=461655530}; {id=127,
int_dv_last=443109475}; {id=28, int_dv_last=395812574}; {id=76,
int_dv_last=390559531}; {id=27, int_dv_last=388277668}; {id=128,
int_dv_last=312954617}; {id=101, int_dv_last=288358140}; {id=36,
int_dv_last=262776063}; {id=8, int_dv_last=258137179}; {id=49,
int_dv_last=203957872}; {id=104, int_dv_last=193961933}; {id=70,
int_dv_last=177134041};
[junit4] 2> 19936 T10 oasc.DistribCursorPagingTest.assertFullWalkNoDups
SOLR-5652:
({params(q=*%3A*&fl=id%2Cint_dv_last&rows=50&sort=int_dv_last+desc%2C+id+desc),defaults(cursorMark=AoJZndujBVYE)})
gave us these docs: {id=4, int_dv_last=159428848}; {id=109,
int_dv_last=146424475}; {id=82, int_dv_last=118282228}; {id=54,
int_dv_last=85706438}; {id=5, int_dv_last=53954157}; {id=90,
int_dv_last=34653341}; {id=125, int_dv_last=22280683}; {id=74, int_dv_last=49};
{id=84, int_dv_last=47}; {id=44, int_dv_last=45}; {id=40, int_dv_last=40};
{id=126, int_dv_last=39}; {id=79, int_dv_last=38}; {id=96, int_dv_last=36};
{id=77, int_dv_last=29}; {id=124, int_dv_last=26}; {id=52, int_dv_last=26};
{id=99, int_dv_last=24}; {id=51, int_dv_last=22}; {id=92, int_dv_last=20};
{id=53}; {id=86, int_dv_last=-27969262}; {id=63, int_dv_last=-64632197};
{id=111, int_dv_last=-65674043}; {id=78, int_dv_last=-92357840}; {id=69,
int_dv_last=-128203234}; {id=118, int_dv_last=-174163156}; {id=56,
int_dv_last=-213598056}; {id=64, int_dv_last=-250673657}; {id=116,
int_dv_last=-349965047}; {id=73, int_dv_last=-363194086}; {id=112,
int_dv_last=-365330802}; {id=97, int_dv_last=-378023587}; {id=20,
int_dv_last=-425200925}; {id=22, int_dv_last=-447702352}; {id=110,
int_dv_last=-457782308}; {id=50, int_dv_last=-458057913}; {id=65,
int_dv_last=-464626829}; {id=43, int_dv_last=-479019126}; {id=81,
int_dv_last=-490373572}; {id=46, int_dv_last=-491310745}; {id=35,
int_dv_last=-524855567}; {id=93, int_dv_last=-562980068}; {id=102,
int_dv_last=-603711742}; {id=106, int_dv_last=-664725432}; {id=37,
int_dv_last=-861444535}; {id=30, int_dv_last=-872755266}; {id=108,
int_dv_last=-905452841}; {id=85, int_dv_last=-908362401}; {id=31,
int_dv_last=-909902582};
[junit4] 2> 19950 T10 oasc.DistribCursorPagingTest.assertFullWalkNoDups
SOLR-5652:
({params(q=*%3A*&fl=id%2Cint_dv_last&rows=50&sort=int_dv_last+desc%2C+id+desc),defaults(cursorMark=AoIGycP9Cl8B)})
gave us these docs: {id=39, int_dv_last=-1040495703}; {id=83,
int_dv_last=-1098867850}; {id=60, int_dv_last=-1130244120}; {id=75,
int_dv_last=-1241745669}; {id=9, int_dv_last=-1248082530}; {id=29,
int_dv_last=-1263869241}; {id=121, int_dv_last=-1269818264}; {id=67,
int_dv_last=-1368223478}; {id=24, int_dv_last=-1402120260}; {id=117,
int_dv_last=-1432445507}; {id=21, int_dv_last=-1445804798}; {id=13,
int_dv_last=-1448419734}; {id=114, int_dv_last=-1450149432}; {id=57,
int_dv_last=-1455879499}; {id=45, int_dv_last=-1465395474}; {id=123,
int_dv_last=-1507457930}; {id=1, int_dv_last=-1563035961}; {id=105,
int_dv_last=-1615487083}; {id=34, int_dv_last=-1656779776}; {id=23,
int_dv_last=-1734132255}; {id=15, int_dv_last=-1745083413}; {id=33,
int_dv_last=-1755075260}; {id=107, int_dv_last=-1819329253}; {id=47,
int_dv_last=-1835223018}; {id=119, int_dv_last=-1860312074}; {id=80,
int_dv_last=-1875371069}; {id=7, int_dv_last=-1884862947}; {id=59,
int_dv_last=-1963096931}; {id=11, int_dv_last=-1978568913}; {id=53};
{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
