[
https://issues.apache.org/jira/browse/LUCENE-10268?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17450115#comment-17450115
]
Robert Muir commented on LUCENE-10268:
--------------------------------------
Don't worry about the loss of seed, I'm running it 100 times and then will
upload a couple of them. Maybe there is an obvious simple pattern.
I "beast" for slow seeds like this:
{noformat}
lucene[main]$ for iteration in {0..100}; do ./gradlew -p lucene/core test
--tests TestFuzzyQuery.testErrorMessage > log$iteration.txt 2>&1; done
...
$ grep testErrorMessage log*.txt
log0.txt: 1.74s TestFuzzyQuery.testErrorMessage (:lucene:core)
log10.txt: 1.73s TestFuzzyQuery.testErrorMessage (:lucene:core)
log11.txt: 1.50s TestFuzzyQuery.testErrorMessage (:lucene:core)
log12.txt: 1.24s TestFuzzyQuery.testErrorMessage (:lucene:core)
log13.txt: 1.19s TestFuzzyQuery.testErrorMessage (:lucene:core)
log1.txt: 3.50s TestFuzzyQuery.testErrorMessage (:lucene:core)
log2.txt: 2.91s TestFuzzyQuery.testErrorMessage (:lucene:core)
log3.txt: 1.97s TestFuzzyQuery.testErrorMessage (:lucene:core)
log4.txt: 1.74s TestFuzzyQuery.testErrorMessage (:lucene:core)
log5.txt: 4.93s TestFuzzyQuery.testErrorMessage (:lucene:core)
log6.txt: 2.59s TestFuzzyQuery.testErrorMessage (:lucene:core)
log7.txt: 48.21s TestFuzzyQuery.testErrorMessage (:lucene:core)
log8.txt: 1.94s TestFuzzyQuery.testErrorMessage (:lucene:core)
log9.txt: 2.13s TestFuzzyQuery.testErrorMessage (:lucene:core)
{noformat}
Already got a 48second run, I'll let all 100 finish.
> TestFuzzyQuery.testErrorMessage can sometimes run for a hundred seconds
> -----------------------------------------------------------------------
>
> Key: LUCENE-10268
> URL: https://issues.apache.org/jira/browse/LUCENE-10268
> Project: Lucene - Core
> Issue Type: Task
> Reporter: Robert Muir
> Priority: Major
>
> I ran 6 test runs, and it happened only in one of them:
> 97.93s TestFuzzyQuery.testErrorMessage (:lucene:core)
> Sorry, I lost the master seed, so I'll have to hunt down a seed to reproduce.
> The test is badly named, but I think the intention is to provoke a
> determinize work limit. Funny because with the lev logic we build a DFA from
> the getgo... so it's just the overhead introduced by UTF32toUTF8.
> Still, this seems to indicate something is wrong, or possible adversary. If I
> can get a seed, we can profile and see what is going on.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]