[
https://issues.apache.org/jira/browse/HDDS-5258?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17622208#comment-17622208
]
Sumit Agrawal commented on HDDS-5258:
-------------------------------------
Unable to reproduce this with latest ozone with same scenario. This might have
fixed with bug HDDS-5216.
As analysis, log pattern shows similar, where om1 is not able to do failover to
leader SCM. Like for current bug, OM1 is not able to failover to scm3 as leader
identified even logs specify failover log. Same came happened to HDDS-5216 as
per log attached.
Test result with latest ozone is attached:
[^testresult.zip]
> freon randomkeys taking very long time to finish
> ------------------------------------------------
>
> Key: HDDS-5258
> URL: https://issues.apache.org/jira/browse/HDDS-5258
> Project: Apache Ozone
> Issue Type: Bug
> Components: SCM HA, Security
> Reporter: George Huang
> Assignee: Sumit Agrawal
> Priority: Major
> Attachments: testresult.zip
>
>
> on a 3 data node ozonesecure-ha cluster, execute freon randomkeys like this
> takes more than 2 hrs to finish:
> ozone freon randomkeys --numOfVolumes=10 --numOfBuckets 10 --numOfKeys 10
> --replicationType=RATIS --factor=THREE
> Only bring all SCMs up
> Assume SCM1 leader
> Start OM1 so it gets cert from SCM1
> Start DN1 so it gets cert from SCM1
> Change leader by stopping SCM1
> Start OM2 (Get cert from new SCM leader)
> Start DN2 (Get cert from new SCM leader)
> Stop new leader and bring SCM1 up
> Start OM3 (Get cert from new SCM leader)
> Start DN3 (Get cert from new SCM leader)
> Execute ozone freon as following:
> ozone freon randomkeys --numOfVolumes=10 --numOfBuckets 10 --numOfKeys 10
> --replicationType=RATIS --factor=THREE
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
