[
https://issues.apache.org/jira/browse/FLINK-22889?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17410856#comment-17410856
]
Roman Khachatryan commented on FLINK-22889:
-------------------------------------------
In the latest logs it's clearly seen that MySQL "XA PREPARE" statements hang
for >8s, for example:
{code:java}
---TRANSACTION 2139, ACTIVE 8 sec
1 lock struct(s), heap size 1136, 0 row lock(s), undo log entries 7
MySQL thread id 43, OS thread handle 140332354893568, query id 715 172.17.0.1
test starting
XA PREPARE
0xb508d2a2741da30eb384acef037ce93c000000000000000000000000,0x6d33a077,0xc9
{code}
I don't see any obvious issues in the reported InnoDB state (in particular,
there are no deadlocks between transactions or IO pressure).
This causes timeout in Flink in snapshotState and task cancellation.
However, after usually 10s the command succeeds (but the cluster is already
shut down).
As the issue seems to be not related to Flink but rather MySQL I'm going
increase timeouts on the Flink side from 2 to 15s so that test can proceed.
> JdbcExactlyOnceSinkE2eTest.testInsert hangs on azure
> ----------------------------------------------------
>
> Key: FLINK-22889
> URL: https://issues.apache.org/jira/browse/FLINK-22889
> Project: Flink
> Issue Type: Bug
> Components: Connectors / JDBC
> Affects Versions: 1.14.0, 1.13.1
> Reporter: Dawid Wysakowicz
> Assignee: Roman Khachatryan
> Priority: Critical
> Labels: pull-request-available, test-stability
> Fix For: 1.14.0
>
>
> https://dev.azure.com/apache-flink/apache-flink/_build/results?buildId=18690&view=logs&j=ba53eb01-1462-56a3-8e98-0dd97fbcaab5&t=bfbc6239-57a0-5db0-63f3-41551b4f7d51&l=16658
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
