[
https://issues.apache.org/jira/browse/IGNITE-26761?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mirza Aliev updated IGNITE-26761:
---------------------------------
Description:
h3. Motivation
Currently, partition restart with clean up is implemented in the way that such
operation could be performed only on a single specified node. Such limitation
is motivated by the fact that Disaster recovery mechanism is developed in the
way that every DR intent is completed only when such intent is completed on the
initiator node. All other nodes perform such operation in a async manner.
With partition restart with clean up this limitation is crucial because the
operation is dangerous in terms of loosing user data.
We need to implement mechanism that allow to perform such restart cluster wide.
was:
h3. Motivation
Currently, partition restart with clean up is implemented in the way that such
operation could be performed only on a single specified node. Such limitation
is motivated by the fact that Disaster recovery mechanism is developed in the
way that every DR intent is completed only when such intent is completed on the
initiator node. All other nodes perform such operation in a async manner.
With partition restart with clean up this limitation is crucial because the
operation is dangerous in terms of loosing user data.
We need to implement mechanism that allow to perform such restart cluster wide.
> Extend partition restart with clean up so it could be performed not only in a
> single node
> -----------------------------------------------------------------------------------------
>
> Key: IGNITE-26761
> URL: https://issues.apache.org/jira/browse/IGNITE-26761
> Project: Ignite
> Issue Type: Improvement
> Reporter: Mirza Aliev
> Priority: Major
> Labels: ignite-3
>
> h3. Motivation
> Currently, partition restart with clean up is implemented in the way that
> such operation could be performed only on a single specified node. Such
> limitation is motivated by the fact that Disaster recovery mechanism is
> developed in the way that every DR intent is completed only when such intent
> is completed on the initiator node. All other nodes perform such operation in
> a async manner.
> With partition restart with clean up this limitation is crucial because the
> operation is dangerous in terms of loosing user data.
> We need to implement mechanism that allow to perform such restart cluster
> wide.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
