You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues-all@impala.apache.org by "ASF subversion and git services (Jira)" <ji...@apache.org> on 2024/03/07 03:40:00 UTC
[jira] [Commented] (IMPALA-12849) Standby catalogd should reject requests from coordinators
[ https://issues.apache.org/jira/browse/IMPALA-12849?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17824241#comment-17824241 ]
ASF subversion and git services commented on IMPALA-12849:
----------------------------------------------------------
Commit 28bc2bdb32ce8e9cd08e4ae21a8a067a513c714c in impala's branch refs/heads/master from wzhou-code
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=28bc2bdb3 ]
IMPALA-12849: Standby catalogd should reject requests from coordinators
In a catalog HA enabled cluster, it's possible that the standby catalogd
could receive requests from coordinators in a short window after catalog
fail-over is triggered since the coordinators may receive the fail-over
notification from statestore with delayed time. In this scenarios,
the standby catalogd should reject all requests from coordinators so
that only one catalogd serving catalog service for the cluster.
This patch checks if the catalog server is active when handling request
from clients.
Testing:
- Added end-to-end unit-test.
- Passed core tests.
Change-Id: Iea38bdf4f207af657e71670a572efc7c0a0ba807
Reviewed-on: http://gerrit.cloudera.org:8080/21086
Reviewed-by: Riza Suminto <ri...@cloudera.com>
Tested-by: Impala Public Jenkins <im...@cloudera.com>
> Standby catalogd should reject requests from coordinators
> ---------------------------------------------------------
>
> Key: IMPALA-12849
> URL: https://issues.apache.org/jira/browse/IMPALA-12849
> Project: IMPALA
> Issue Type: Bug
> Components: Backend
> Affects Versions: Impala 4.3.0, Impala 4.4.0
> Reporter: Wenzhe Zhou
> Assignee: Wenzhe Zhou
> Priority: Major
> Fix For: Impala 4.4.0
>
>
> In the catalog HA enabled cluster, a standby catalogd could receive requests from coordinators in a short window after a manual fail-over is triggered since the coordinators may receive fail-over notification with delay time. In this corner case, the standby catalogd should reject all requests from coordinators.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-all-unsubscribe@impala.apache.org
For additional commands, e-mail: issues-all-help@impala.apache.org