[
https://issues.apache.org/jira/browse/IGNITE-7506?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16337142#comment-16337142
]
Sergey Chugunov commented on IGNITE-7506:
-----------------------------------------
Patch is ready, TC status looks acceptable (test failures count matches with
the same in master branch):
https://ci.ignite.apache.org/viewQueued.html?itemId=1057419&tab=queuedBuildOverviewTab
> GridClusterStateProcessor#compatibilityMode flag never gets reset to false
> --------------------------------------------------------------------------
>
> Key: IGNITE-7506
> URL: https://issues.apache.org/jira/browse/IGNITE-7506
> Project: Ignite
> Issue Type: Bug
> Reporter: Sergey Chugunov
> Assignee: Sergey Chugunov
> Priority: Major
>
> When Ignite node of version 2.4+ joins cluster with older versions of Ignite
> running it sets *compatibilityMode* flag to handle **cluster state in a
> special way (excluding BaselineTopology from discovery exchange process and
> so on).
> But when all old nodes are turned off there is no way except full cluster
> restart to enable setting very first **BaselineTopology.
> This should be addressed and some way to exit compatibilityMode without full
> cluster restart should be provided.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
