[
https://issues.apache.org/jira/browse/SAMZA-168?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Martin Kleppmann updated SAMZA-168:
-----------------------------------
Attachment: SAMZA-168.patch
Attaching patch which adds exponential backoff. RB:
https://reviews.apache.org/r/18760/
While I was at it, I also tidied up ExponentialThreadSleepStrategy and moved it
to samza-core.
> Getting Kafka system metadata should back off on retry
> ------------------------------------------------------
>
> Key: SAMZA-168
> URL: https://issues.apache.org/jira/browse/SAMZA-168
> Project: Samza
> Issue Type: Bug
> Affects Versions: 0.6.0
> Reporter: Martin Kleppmann
> Assignee: Martin Kleppmann
> Fix For: 0.7.0
>
> Attachments: SAMZA-168.patch
>
>
> If an exception occurs in KafkaSystemAdmin.getSystemStreamMetadata (eg
> because the broker returns an error for the partitions metadata request), we
> currently retry in a tight loop. This puts unnecessary load on Kafka, and
> especially risks hammering Zookeeper.
> There should be an exponential back-off in the retry loop.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
