[
https://issues.apache.org/jira/browse/HADOOP-16363?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16861942#comment-16861942
]
Steve Loughran edited comment on HADOOP-16363 at 6/12/19 9:37 AM:
------------------------------------------------------------------
Fixed in HADOOP-15183 with a new exception {{TableDeleteTimeoutException
extends PathIOException}}; IllegalArgumentException on teardown is mapped to
this.
Just realised I need to catch this in the S3GuardTool.Destroy as it's not a
failure, I'm going to downgrade to warn. Scripts will still succeed, they just
may fail if there is a new init immediately after -and that failure will be
caught and reported.
was (Author: [email protected]):
Fixed in HADOOP-15183 with a new exception {{TableDeleteTimeoutException
extends PathIOException}}; IllegalArgumentException on teardown is mapped to
this.
Just realised I need to catch this in the S3GuardTool.Destroy as it's not a
failure, I'm going to downgrade to warn. Scripts will still succeed, they just
may fail on the following init
> S3Guard DDB store prune() doesn't translate AWS exceptions to IOEs
> ------------------------------------------------------------------
>
> Key: HADOOP-16363
> URL: https://issues.apache.org/jira/browse/HADOOP-16363
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/s3
> Affects Versions: 3.2.0
> Reporter: Steve Loughran
> Assignee: Steve Loughran
> Priority: Minor
>
> Fixing in HADOOP-15183: if you call prune() against a nonexist DDB table, the
> exception isn't being translated into an IOE.
> This is interesting as the codepath is going through retry(), it's just that
> where the IO is taking place is happening inside the iterator, and we don't
> have checks there.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
