[
https://issues.apache.org/jira/browse/HDDS-15676?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ryan Blough resolved HDDS-15676.
--------------------------------
Resolution: Invalid
The cluster where the problem was observed turned out to be version 1.4.0 +
cherry-picks, so the issue was already solved in the 2.X line, but the fixes
were not done in time for 1.4.0. Several related rocksDB fixes went into 1.4.1,
but the correct approach to solving the problem is to resolve cherry-picking
for them.
> Fix Recon rocksDB lifecycle management
> --------------------------------------
>
> Key: HDDS-15676
> URL: https://issues.apache.org/jira/browse/HDDS-15676
> Project: Apache Ozone
> Issue Type: Bug
> Components: Ozone Recon
> Affects Versions: 2.2.0
> Reporter: Ryan Blough
> Assignee: Ryan Blough
> Priority: Major
>
> The lifecycle management of rocksDB in Recon is different from and weaker
> than that used by OM and SCM roles in their management of rocksDB. This
> leaves Recon vulnerable to problems such as ongoing "Rocks Database is
> closed" errors that persist through restarts.
> We should adopt the methods employed on OM/SCM for Recon, in order to close
> these vulnerabilities.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
