You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ozone.apache.org by "Jackson Yao (Jira)" <ji...@apache.org> on 2021/09/03 06:41:00 UTC
[jira] [Updated] (HDDS-5710) initialize sequenceIdToLastIdMap when
SequenceIdGenerator#StateManager reinitializes
[ https://issues.apache.org/jira/browse/HDDS-5710?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jackson Yao updated HDDS-5710:
------------------------------
Description: 2021-09-03 12:50:38,632 [1114efaf-c975-4ea0-bc8b-c3309a57cc8d@group-58D600BFB107-StateMachineUpdater] WARN org.apache.hadoop.hdds.scm.ha.SequenceIdGenerator: Failed to allocate a batch for localId, expected lastId is 0, actual lastId is 107544261438510000. (was: in GrpcLogAppender.java, `synchronized` is used for two purposes:
1 protect `wait()`
2 protect variables like `pendingRequests`, `pending` , etc.
i think it is better off add a separate RWLock for the second case ， and keep `synchronized` only for wait and notify, thus any slow IO option will not block notify)
> initialize sequenceIdToLastIdMap when SequenceIdGenerator#StateManager reinitializes
> ------------------------------------------------------------------------------------
>
> Key: HDDS-5710
> URL: https://issues.apache.org/jira/browse/HDDS-5710
> Project: Apache Ozone
> Issue Type: Improvement
> Reporter: Jackson Yao
> Assignee: Jackson Yao
> Priority: Major
>
> 2021-09-03 12:50:38,632 [1114efaf-c975-4ea0-bc8b-c3309a57cc8d@group-58D600BFB107-StateMachineUpdater] WARN org.apache.hadoop.hdds.scm.ha.SequenceIdGenerator: Failed to allocate a batch for localId, expected lastId is 0, actual lastId is 107544261438510000.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org