[ 
https://issues.apache.org/jira/browse/KUDU-3011?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17010231#comment-17010231
 ] 

ASF subversion and git services commented on KUDU-3011:
-------------------------------------------------------

Commit 4619473b9a83dce2b6a60a79802fde33c8686238 in kudu's branch 
refs/heads/master from Andrew Wong
[ https://gitbox.apache.org/repos/asf?p=kudu.git;h=4619473 ]

KUDU-3011 p3: mechanism to quiesce scans

This prevents new scans from being started on a tablet server when it is
quiescing. The scans are retried elsewhere.

I added some tests to check the behavior with various non-fault-tolerant
scenarios to ensure we can rely on quiescing to move scans away from
specific tablet servers without failing read workloads. To do this, I
added some additional configuration to TestWorkload.

Change-Id: Idedca29f40b0a8576245be0f7cfad2be29db4135
Reviewed-on: http://gerrit.cloudera.org:8080/14986
Reviewed-by: Adar Dembo <a...@cloudera.com>
Tested-by: Kudu Jenkins


> Support for smooth maintenance window
> -------------------------------------
>
>                 Key: KUDU-3011
>                 URL: https://issues.apache.org/jira/browse/KUDU-3011
>             Project: Kudu
>          Issue Type: New Feature
>            Reporter: LiFu He
>            Assignee: Andrew Wong
>            Priority: Major
>
> A scan corresponding to a tablet failure causes the entire SQL to fail on the 
> common query engines, such as Impala. Though we have the fault-tolerant 
> feature by "SetFaultTolerant()", Impala doesn't use it right now since that 
> will make lower throughput. Thus, lots of SQL that are running will fail when 
> we shutdown/reboot/upgrade the tserver. That can be scary.
> Maybe we can do some improvement in this area, for example, the tablets are 
> not allowed to be scanned after the tserver is in maintenance mode 
> (KUDU-2069). And for the LEADER_ONLY mode scanning, the leader role needs to 
> be shifted from the maintenance tserver. Then we can shutdown the tserver 
> smoothly after all the existing SQL are completed.
>  



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to