Quanlong Huang has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/21376 )

Change subject: IMPALA-13047: Support restarting/killing specified impalads in 
start-impala-cluster.py
......................................................................


Patch Set 2:

(1 comment)

http://gerrit.cloudera.org:8080/#/c/21376/1/bin/start-impala-cluster.py
File bin/start-impala-cluster.py:

http://gerrit.cloudera.org:8080/#/c/21376/1/bin/start-impala-cluster.py@1118
PS1, Line 1118:   existing_cluster_size = 
cluster_ops.get_cluster().get_max_impalad_id() + 1
We currently don't have a way to remember the cluster size. So if the last 
impalad is killed, we will see the size shrink. I think that's ok. This check 
is added to avoid abusing this feature since I haven't test it with all 
combinations of other flags. This script is only used by developers (in test 
codes, we already have a way to restart any process). We can add support for 
more complex use cases if we really need.

> For example start 3, kill number 2, then restart number 0, it runs on the 
> assertion again.

In this case, when restarting number 0, we need "-s 2" as a workaround.

> Isn't it possible to use the length of cluster_ops.get_cluster().__impalads?

This list consists of the running impalads. So if some were killed, the length 
shrinks as well.



--
To view, visit http://gerrit.cloudera.org:8080/21376
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Id4a863853341959eb6870b662d82188e7d570971
Gerrit-Change-Number: 21376
Gerrit-PatchSet: 2
Gerrit-Owner: Quanlong Huang <[email protected]>
Gerrit-Reviewer: Daniel Becker <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Quanlong Huang <[email protected]>
Gerrit-Comment-Date: Mon, 06 May 2024 20:56:58 +0000
Gerrit-HasComments: Yes

Reply via email to