[
https://issues.apache.org/jira/browse/IMPALA-7532?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16619890#comment-16619890
]
Tianyi Wang commented on IMPALA-7532:
-------------------------------------
[~vukercegovac] Do you know the details about the problem? AFAIK the main RPC
used by fetch-from-catalogd is GetPartialCatalogObject, which has retrial
mechanisms in catalog-op-executor.cc
> Add retry/back-off to fetch-from-catalog RPCs
> ---------------------------------------------
>
> Key: IMPALA-7532
> URL: https://issues.apache.org/jira/browse/IMPALA-7532
> Project: IMPALA
> Issue Type: Sub-task
> Reporter: Todd Lipcon
> Assignee: Tianyi Wang
> Priority: Major
>
> Currently if there is an error connecting to the catalog server, the 'fetch
> from catalog' implementation will retry with no apparent backoff. We should
> retry for some period of time with backoff in between the attempts, so that
> impala can ride over short interruptions of the catalog service.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
