Thanks for you reply,
Yes, this is indeed an option. But I was more after a config option to handle
that scenario. If the HA metadata points to a checkpoint that is obviously not
present (err 404 in the S3 case) there is little value in retrying. The HA data
are obviously worthless in that scenario.
But maybe there isn't any.
Best regards
JM
________________________________
From: Zhanghao Chen <zhanghao.c...@outlook.com>
Sent: Tuesday, June 11, 2024 03:56
To: Jean-Marc Paulin <j...@uk.ibm.com>; user@flink.apache.org
<user@flink.apache.org>
Subject: [EXTERNAL] Re: Failed to resume from HA when the checkpoint has been
deleted.
Hi, In this case, you could cancel the job using the flink stop command, which
will clean up Flink HA metadata, and resubmit the job. Best, Zhanghao Chen
From: Jean-Marc Paulin <jmp@ uk. ibm. com> Sent: Monday, June 10, 2024 18: 53
To: user@ flink. apache. org
Hi,
In this case, you could cancel the job using the flink stop command, which
will clean up Flink HA metadata, and resubmit the job.
Best,
Zhanghao Chen
________________________________
From: Jean-Marc Paulin <j...@uk.ibm.com>
Sent: Monday, June 10, 2024 18:53
To: user@flink.apache.org <user@flink.apache.org>
Subject: Failed to resume from HA when the checkpoint has been deleted.
Hi,
We have a 1.19 Flink streaming job, with HA enabled (ZooKeeper),
checkpoint/savepoint in S3. We had an outage and now the jobmanager keeps
restarting. We think it because it read the job id to be restarted from
ZooKeeper, but because we lost our S3 Storage as part of the outage it cannot
find the checkpoint to restart from, and dies.
```
Found 1 checkpoints in
ZooKeeperStateHandleStore{namespace='flink/aiops/ir-lifecycle/jobs/2512c6153c7ae16fa6da6d64772d75c5/checkpoints'
Trying to fetch 1 checkpoints from storage.
Trying to retrieve checkpoint 50417.
exception: JobMaster for job 2512c6153c7ae16fa6da6d64772d75c5 failed.
Caused by: org.apache.flink.runtime.client.JobInitializationException: Could
not start the JobMaster.
Caused by: java.util.concurrent.CompletionException:
java.lang.RuntimeException:
org.apache.flink.runtime.client.JobExecutionException: Failed to initialize
high-availability completed checkpoint store
...
Caused by: org.apache.flink.util.FlinkException: Could not retrieve checkpoint
50417 from state handle under /0000000000000050417. This indicates that the
retrieved state handle is broken. Try cleaning the state handle store.
Caused by:
com.facebook.presto.hive.s3.PrestoS3FileSystem$UnrecoverableS3OperationException:
com.amazonaws.services.s3.model.AmazonS3Exception: The specified key does not
exist. (Service: Amazon S3; Status Code: 404; Error Code: NoSuchKey; Request
ID: 17D7166A4D756355; S3 Extended Request ID:
fe09be003d6379d952fad9de241c370b5f7ac43631c02fdfbc9dda9c4398d6df; Proxy: null),
S3 Extended Request ID:
fe09be003d6379d952fad9de241c370b5f7ac43631c02fdfbc9dda9c4398d6df (Path:
s3://test/high-availability/flink-job/completedCheckpoint64d901465702)
Fatal error occurred in the cluster entrypoint.
```
Is there an option we can use to configure the job to ignore this error?
Kind regards
JM
Unless otherwise stated above:
IBM United Kingdom Limited
Registered in England and Wales with number 741598
Registered office: PO Box 41, North Harbour, Portsmouth, Hants. PO6 3AU
Unless otherwise stated above:
IBM United Kingdom Limited
Registered in England and Wales with number 741598
Registered office: PO Box 41, North Harbour, Portsmouth, Hants. PO6 3AU
