[
https://issues.apache.org/jira/browse/SOLR-16322?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17574861#comment-17574861
]
Uwe Schindler commented on SOLR-16322:
--------------------------------------
The summary task seems to use an outdated seed, not the one of the individual
task. If you scroll up you should see the correct seed with every failed test.
I noticed this already when working on JVM bugs (I was beasting a lot), but I
never look at the summary output so I was not taking care. The correct seed is
printed in test failure, so it wasn't an issue to me. In general the summary
output only shows the seed of the never executed main "test" task, but not of
the virtual "text_xy" tasks created by the beaster.
Could you please also open bug in Lucene, as this is code inherited from there,
it is not specific for Solr.
> Beast task does not print correct seeds
> ---------------------------------------
>
> Key: SOLR-16322
> URL: https://issues.apache.org/jira/browse/SOLR-16322
> Project: Solr
> Issue Type: Bug
> Reporter: Mike Drob
> Priority: Major
>
> When there are multiple failing seeds, the beast task will not print all of
> them in the summary. Instead I have to go to the individual failing test
> outputs to find all of the seeds.
> For example, the summary output at the end will show:
> {noformat}
> ERROR: The following test(s) have failed:
> -
> org.apache.solr.uninverting.TestUninvertingReader.testSortedSetIntegerManyValues
> (:solr:core)
> Test output:
> /Users/mdrob/code/solr/solr/core/build/test-results/test_16/outputs/OUTPUT-org.apache.solr.uninverting.TestUninvertingReader.txt
> Reproduce with: gradlew :solr:core:test --tests
> "org.apache.solr.uninverting.TestUninvertingReader.testSortedSetIntegerManyValues"
> -Ptests.jvms=16 "-Ptests.jvmargs=-XX:TieredStopAtLevel=1
> -XX:ReservedCodeCacheSize=120m" -Ptests.seed=98CEE6B610F6FC3A
> -Ptests.badapples=true -Ptests.file.encoding=UTF-8 -Ptests.dups=50
> -
> org.apache.solr.uninverting.TestUninvertingReader.testSortedSetIntegerManyValues
> (:solr:core)
> Test output:
> /Users/mdrob/code/solr/solr/core/build/test-results/test_1/outputs/OUTPUT-org.apache.solr.uninverting.TestUninvertingReader.txt
> Reproduce with: gradlew :solr:core:test --tests
> "org.apache.solr.uninverting.TestUninvertingReader.testSortedSetIntegerManyValues"
> -Ptests.jvms=16 "-Ptests.jvmargs=-XX:TieredStopAtLevel=1
> -XX:ReservedCodeCacheSize=120m" -Ptests.seed=98CEE6B610F6FC3A
> -Ptests.badapples=true -Ptests.file.encoding=UTF-8 -Ptests.dups=50
> -
> org.apache.solr.uninverting.TestUninvertingReader.testSortedSetIntegerManyValues
> (:solr:core)
> Test output:
> /Users/mdrob/code/solr/solr/core/build/test-results/test_2/outputs/OUTPUT-org.apache.solr.uninverting.TestUninvertingReader.txt
> Reproduce with: gradlew :solr:core:test --tests
> "org.apache.solr.uninverting.TestUninvertingReader.testSortedSetIntegerManyValues"
> -Ptests.jvms=16 "-Ptests.jvmargs=-XX:TieredStopAtLevel=1
> -XX:ReservedCodeCacheSize=120m" -Ptests.seed=98CEE6B610F6FC3A
> -Ptests.badapples=true -Ptests.file.encoding=UTF-8 -Ptests.dups=50
> -
> org.apache.solr.uninverting.TestUninvertingReader.testSortedSetIntegerManyValues
> (:solr:core)
> Test output:
> /Users/mdrob/code/solr/solr/core/build/test-results/test_14/outputs/OUTPUT-org.apache.solr.uninverting.TestUninvertingReader.txt
> Reproduce with: gradlew :solr:core:test --tests
> "org.apache.solr.uninverting.TestUninvertingReader.testSortedSetIntegerManyValues"
> -Ptests.jvms=16 "-Ptests.jvmargs=-XX:TieredStopAtLevel=1
> -XX:ReservedCodeCacheSize=120m" -Ptests.seed=98CEE6B610F6FC3A
> -Ptests.badapples=true -Ptests.file.encoding=UTF-8 -Ptests.dups=50
> {noformat}
> We can see that here every seed is {{98CEE6B610F6FC3A}}. However, if I go to
> the individual test outputs, I see a bunch of different seeds (none of which
> are actually this seed!)
> I was seeing this on my fork
> https://github.com/madrob/solr/commit/314716a681d71d95d27640e06f7aeb422504287c
> in case anybody wants to try reproduce it, or you could introduce your own
> randomly failing test.
> cc [~uschindler] [~dweiss]
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
