[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17438518#comment-17438518
]
Berenguer Blasi commented on CASSANDRA-17070:
-
Yep wall of green :-)
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17438204#comment-17438204
]
Andres de la Peña commented on CASSANDRA-17070:
---
The PR for trunk looks good to me. Here are 200 runs of {{ViewComplex*Test}} in
the PR for trunk:
*
[j8|https://app.circleci.com/pipelines/github/adelapena/cassandra/1114/workflows/2a6ea982-7109-4cf2-b5af-7f06188f2d4f]
*
[j11|https://app.circleci.com/pipelines/github/adelapena/cassandra/1114/workflows/12ab59cb-24d5-4795-a309-21b7eb6b187c]
If those succeed I think we'll be definitively ready to merge.
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17437991#comment-17437991
]
Berenguer Blasi commented on CASSANDRA-17070:
-
I added the trunk PR with CI. There's no rush so this will give you time to
take a look at it. Let me know if you're still ok I merge.
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17437409#comment-17437409
]
Andres de la Peña commented on CASSANDRA-17070:
---
I don't think we should put the duplicated cleanup logic back, in principle not
reusing MV names should be enough.
I mentioned the query timeout to indicate that the failure is not related to
what we are addressing here, so I think the run can be considered successful.
By the way,
[here|https://app.circleci.com/pipelines/github/adelapena/cassandra/1101/workflows/63014008-3d52-4d07-a5aa-748fad389ec7]
is the repeated run for j11, all green. I think we are ready to merge.
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17437207#comment-17437207
]
Berenguer Blasi commented on CASSANDRA-17070:
-
Right. But it's on view creation, not the cleanup logic. Seems like a legit
timeout. I've had the feeling timeouts have to be raised for some time. I still
think we're good to merge. Or are you suggesting we put the cleanup back in?
{noformat}
[junit-timeout] Testcase:
testUpdateWithColumnTimestampSmallerThanPkWithFlush[3](org.apache.cassandra.cql3.ViewComplexUpdatesTest):
Caused an ERROR
[junit-timeout] [localhost/127.0.0.1:43739] Timed out waiting for server
response
[junit-timeout] com.datastax.driver.core.exceptions.OperationTimedOutException:
[localhost/127.0.0.1:43739] Timed out waiting for server response
[junit-timeout] at
com.datastax.driver.core.exceptions.OperationTimedOutException.copy(OperationTimedOutException.java:43)
[junit-timeout] at
com.datastax.driver.core.exceptions.OperationTimedOutException.copy(OperationTimedOutException.java:25)
[junit-timeout] at
com.datastax.driver.core.DriverThrowables.propagateCause(DriverThrowables.java:35)
[junit-timeout] at
com.datastax.driver.core.DefaultResultSetFuture.getUninterruptibly(DefaultResultSetFuture.java:293)
[junit-timeout] at
com.datastax.driver.core.AbstractSession.execute(AbstractSession.java:58)
[junit-timeout] at
com.datastax.driver.core.AbstractSession.execute(AbstractSession.java:45)
[junit-timeout] at
org.apache.cassandra.cql3.CQLTester.executeNet(CQLTester.java:972)
[junit-timeout] at
org.apache.cassandra.cql3.ViewComplexTester.createView(ViewComplexTester.java:83)
[junit-timeout] at
org.apache.cassandra.cql3.ViewComplexUpdatesTest.testUpdateWithColumnTimestampSmallerThanPk(ViewComplexUpdatesTest.java:224)
[junit-timeout] at
org.apache.cassandra.cql3.ViewComplexUpdatesTest.testUpdateWithColumnTimestampSmallerThanPkWithFlush(ViewComplexUpdatesTest.java:207)
[junit-timeout] Caused by:
com.datastax.driver.core.exceptions.OperationTimedOutException:
[localhost/127.0.0.1:43739] Timed out waiting for server response
[junit-timeout] at
com.datastax.driver.core.RequestHandler$SpeculativeExecution.onTimeout(RequestHandler.java:979)
[junit-timeout] at
com.datastax.driver.core.Connection$ResponseHandler$1.run(Connection.java:1636)
[junit-timeout] at
com.datastax.shaded.netty.util.HashedWheelTimer$HashedWheelTimeout.expire(HashedWheelTimer.java:663)
[junit-timeout] at
com.datastax.shaded.netty.util.HashedWheelTimer$HashedWheelBucket.expireTimeouts(HashedWheelTimer.java:738)
[junit-timeout] at
com.datastax.shaded.netty.util.HashedWheelTimer$Worker.run(HashedWheelTimer.java:466)
[junit-timeout] at
com.datastax.shaded.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30)
[junit-timeout] at java.lang.Thread.run(Thread.java:748)
{noformat}
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17436023#comment-17436023
]
Andres de la Peña commented on CASSANDRA-17070:
---
[Here|https://app.circleci.com/pipelines/github/adelapena/cassandra/1101/workflows/c335ffa0-1513-40a1-b925-292e0d03cee3/jobs/10214]
are other 500 rounds of the tests without the duplicated cleanup. There is a
single error in
{{ViewComplexUpdatesTest.testUpdateWithColumnTimestampSmallerThanPk}},
apparently due to a query timeout (see the bottom of the
[stdout|https://circle-production-customer-artifacts.s3.amazonaws.com/picard/52e11c3bf7d0f78c41388745/617bdf2aa30bc2444e1c8dda-15-build/artifacts/stdout/fails/05/test-ViewComplex%2ATest.txt?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20211029T144933Z&X-Amz-SignedHeaders=host&X-Amz-Expires=60&X-Amz-Credential=AKIAJR3Q6CR467H7Z55A%2F20211029%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Signature=033813c3f1bc4bd9485bd06dad36099d61616fc505b8ae03d71cba4e41cf7844]
file).
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17435921#comment-17435921
]
Berenguer Blasi commented on CASSANDRA-17070:
-
They didn't fail sort to speak... They show the 13200 passed tests with no
_test_ failures. It failed the run bc of the wilcard. It is easy to fix it by
switching to {{test}} instead of {{testsome}}.
Removing the cleanup logic breaks the {{CQLTester}} teardown when it tries to
drop tables that have attached MVs. So I want to give that a thought...
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17435885#comment-17435885
]
Andres de la Peña commented on CASSANDRA-17070:
---
It seems that the multiplexer runs have failed due to the wildcard. I think it
would work if we use {{ant test}} instead of {{ant testsome}}, with the
unqualified class name:
{code}
.circleci/generate.sh -m \
-e REPEATED_UTEST_TARGET=test \
-e REPEATED_UTEST_COUNT=500 \
-e REPEATED_UTEST_CLASS=ViewComplex*Test
{code}
The wildcard is a great idea, it would have saved me time with the above runs :)
I think we can try some more runs without the duplicated cleanup logic and see
how it goes, wdyt?
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17435840#comment-17435840
]
Berenguer Blasi commented on CASSANDRA-17070:
-
[~adelapena] the reason to run that code twice was that the test keeps a list
of the views. So if a cleanup failed the next one would pick up from the
failing point
But +1 to the abstract class and unique MV name. I ported your changes to my PR
but: I deleted the extra cleanup which is not needed anymore, I changed the
config to run {{ViewComplex*Test}} instead of having one commit per repeat test
and I attached results to the PR.
I am only on the fence about the cleanup logic, which we could remove now there
are unique MV names, but on the other hand it's nice to have it. I think I am
+1 on how things are right now. Wdyt?
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17435630#comment-17435630
]
Ekaterina Dimitrova commented on CASSANDRA-17070:
-
+1 on the proposed change to use a super class similar to #16777, it was on my
list for some time but it was a lower priority. Honestly, we should have done
it at first place like that. Thanks [~adelapena]!
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17435564#comment-17435564
]
Andres de la Peña commented on CASSANDRA-17070:
---
I forgot to mention that I found an unrelated mistake
[here|https://github.com/apache/cassandra/blob/trunk/test/unit/org/apache/cassandra/cql3/ViewComplexTest.java#L142-L148],
where two tests are supposed to run a scenario with and without flushing the
table, but both of them run the scenario with flush. It's fixed in the commit
extracting a superclass,
[here|https://github.com/adelapena/cassandra/blob/204f3bbe69aae9165c0570224bfd811c8bbd4d07/test/unit/org/apache/cassandra/cql3/ViewComplexTest.java#L55-L61].
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17435553#comment-17435553
]
Andres de la Peña commented on CASSANDRA-17070:
---
I'm not sure I understand why the MV cleanup can fail. Would the second round
of cleanup guarantee that the MV cleanups succeed, or would it reduce the
chances of having leftovers because it's unlikely to have the same error twice?
Maybe it's just giving more time for the first async cleanup to happen?
In any case, here are 500 runs of each ViewComplex test:
*
[ViewComplexDeletionsTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1090/workflows/4362eefe-a13a-4a8c-a8eb-9e543a74623a/jobs/10187]
*
[ViewComplexLivenessTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1091/workflows/a9185c89-ceff-4ddd-90c0-960281233fe3/jobs/10189]
*
[ViewComplexTTLTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1092/workflows/1f74edb4-3a41-47fd-b95c-0edc1d514a20/jobs/10191]
*
[ViewComplexTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1093/workflows/f2e69851-19ea-4ab6-9315-5af9d721ec3d/jobs/10193]
*
[ViewComplexUpdatesTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1094/workflows/0d25bf72-7c90-4287-92d6-21a751082e21/jobs/10196]
>From this only a test runner for {{ViewComplexTest}} is failing due to
>timeouts.
It's not strictly related to the fix, but it seems there is some repeated code
among the five tests in which {{ViewComplexTest}} was split by CASSANDRA-16670.
Probably we should do the same thing that we did with {{ViewTest}} in
CASSANDRA-16777 and extract a superclass to avoid code duplication. This would
changes like this one a bit easier to do. I gave it a go in [this
commit|https://github.com/adelapena/cassandra/commit/204f3bbe69aae9165c0570224bfd811c8bbd4d07],
and it saves us some 176 lines of code, which I think is always nice.
Here are some repeated runs for the test using a common superclass, with no
failures so far:
*
[ViewComplexDeletionsTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1096/workflows/ded0dfe1-e736-468a-babb-0c007a6bf3c9/jobs/10201]
*
[ViewComplexLivenessTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1097/workflows/544b429d-b712-428c-b97d-edd50076992a/jobs/10203]
*
[ViewComplexTTLTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1098/workflows/c062d815-2e26-4a75-aa64-656bfc48c8cc/jobs/10205]
*
[ViewComplexTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1099/workflows/1f834191-b30b-4380-87ce-da257fac731e/jobs/10207]
*
[ViewComplexUpdatesTest|https://app.circleci.com/pipelines/github/adelapena/cassandra/1095/workflows/3df8138d-1d7b-4283-be3c-a23cb2c0bc40]
An alternative/complementary approach to avoid problems with leftovers from
previous test runs is making sure that all MVs are created with a new, unique
name. This way the tests don't collide with any surviving MVs if the cleanup
has failed or if it hasn't finished. {{CQLTester}} does this with keyspace,
table, function and aggregate names.
I gave it a go to this generation of unique names
[here|https://github.com/adelapena/cassandra/commit/e1928447e4139f1ad4126d9ed34e3de9d3bceb10].
It is just for {{ViewComplexTest}}, but we may consider having a more generic
and reusable solution for creating/cleaning MVs in {{CQLTester}}, although I
think we don't have to it here. wdyt?
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
[jira] [Commented] (CASSANDRA-17070) ViewComplexTest hardening
[
https://issues.apache.org/jira/browse/CASSANDRA-17070?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17435174#comment-17435174
]
Berenguer Blasi commented on CASSANDRA-17070:
-
Hardening by trying to rerun the cleanup logic on any leftovers in case the
first attempt failed. Trunk PR will be identical.
> ViewComplexTest hardening
> -
>
> Key: CASSANDRA-17070
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17070
> Project: Cassandra
> Issue Type: Bug
> Components: Test/unit
>Reporter: Berenguer Blasi
>Assignee: Berenguer Blasi
>Priority: Normal
> Fix For: 4.0.x, 4.x
>
>
> I have seen a number of times already the {{ViewComplexTest}} family timeout
> on test method teardown. This leaves a dirty env behind triggering the
> following test methods to fail on it. This ticket aims at hardening them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
