[
https://issues.apache.org/jira/browse/HADOOP-13664?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15526487#comment-15526487
]
Steve Loughran commented on HADOOP-13664:
-----------------------------------------
There's something else to consider: would we ever want to have a retry policy
for all the FS level operations (GET, HEAD), or should the code fail fast. I
think we'd need to get more traces of in-the-field failures to see how easy it
would be to distinguish transient network errors from unrecoverable problems.
Presumably, any HTTP error code would be taken as unrecoverable; I don't know
how the rest would present themselves or already get handled in the AWS
stack...we know there is some retry logic there.
> S3AInputStream to use a retry policy on read failures
> -----------------------------------------------------
>
> Key: HADOOP-13664
> URL: https://issues.apache.org/jira/browse/HADOOP-13664
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/s3
> Affects Versions: 2.8.0
> Reporter: Steve Loughran
> Priority: Minor
>
> {{S3AInputStream}} has some retry logic to handle failures on a read: log and
> retry. We should move this over to a (possibly hard coded RetryPolicy with
> some sleep logic, so that longer-than-just-transient read failures can be
> handled.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
