Unstable test: validateDeployment – com.linagora.tmail.deployment.DistributedImapAndSmtpTest

[hungphan227]

Error
Location: /com/linagora/tmail/deployment/scripts/ValidateDeployment.test:404
LastClientMsg: s0001 SELECT anothermailbox
Expected: '* OK [HIGHESTMODSEQ 0] Highest'
Actual : '* OK [HIGHESTMODSEQ 1] Highest'
Stacktrace
org.apache.james.mpt.protocol.ProtocolSession$InvalidServerResponseException:
Location: /com/linagora/tmail/deployment/scripts/ValidateDeployment.test:404
LastClientMsg: s0001 SELECT anothermailbox
Expected: '* OK [HIGHESTMODSEQ 0] Highest'
Actual : '* OK [HIGHESTMODSEQ 1] Highest'
at org.apache.james.mpt.protocol.ProtocolSession.handleFailure(ProtocolSession.java:508)
at org.apache.james.mpt.protocol.ProtocolSession$ServerResponse.checkResponse(ProtocolSession.java:354)
at org.apache.james.mpt.protocol.ProtocolSession$ServerResponse.testProtocol(ProtocolSession.java:340)
at org.apache.james.mpt.protocol.ProtocolSession.runSessions(ProtocolSession.java:105)
at org.apache.james.mpt.script.GenericSimpleScriptedTestProtocol.runSessions(GenericSimpleScriptedTestProtocol.java:178)
at org.apache.james.mpt.script.GenericSimpleScriptedTestProtocol.run(GenericSimpleScriptedTestProtocol.java:131)
at com.linagora.tmail.deployment.TMailDeploymentValidation.validateDeployment(TMailDeploymentValidation.java:79)
at java.base/java.lang.reflect.Method.invoke(Unknown Source)
at java.base/java.util.ArrayList.forEach(Unknown Source)
at java.base/java.util.ArrayList.forEach(Unknown Source)

https://james-jenkins.lin-saas.com/blue/organizations/jenkins/Tmail%20build/detail/PR-1566/1/tests/

[Arsnael]

They do break sometimes cause they are declared in james but not tested there anymore, while we still use it for tmail deployment tests on our side.

So you need to fix it on james while testing on tmail to make it pass

[Arsnael]

Except if flaky maybe... did you try to run this locally on your PR? green or not? If breaks with your changes on tmail though, you might want to check your PR again

[hungphan227]

Tested in local. It is flaky

[Arsnael]

Tested in local. It is flaky

In master branch or your branch?

[hungphan227]

In my branch. I am only able to reproduce it once. However, I don't think my code change has anything to do with this test. My best guess is that the mailbox is changed by something after being created (one possible root cause is recent change by Tung for MailboxACL in PropagateLookupRightListener). Therefore, I suggest removing checking HIGHESTMODSEQ in the test script:

C: s0001 SELECT anothermailbox
S: \* OK \[MAILBOXID \(.+\)\] Ok
S: \* FLAGS \(\\Answered \\Deleted \\Draft \\Flagged \\Seen\)
S: \* 0 EXISTS
S: \* 0 RECENT
S: \* OK \[UIDVALIDITY .*\] UIDs valid
S: \* OK \[PERMANENTFLAGS \(\\Answered \\Deleted \\Draft \\Flagged \\Seen \\\*\)\] Limited
S: \* OK \[HIGHESTMODSEQ 0\] Highest
S: \* OK \[UIDNEXT 1\] Predicted next UID
S: s0001 OK \[READ-WRITE\] SELECT completed.

[hungphan227]

#1580