[
https://issues.apache.org/jira/browse/IGNITE-23571?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mirza Aliev updated IGNITE-23571:
---------------------------------
Description:
h3. Motivation
According to
[IEP-131|https://cwiki.apache.org/confluence/display/IGNITE/IEP-131%3A+Partition+Majority+Unavailability+Handling]
{{DisasterRecoveryManager#resetPartitions}} will be the core method for
recovering partition majority availability. Current logic of this method
schedules new rebalance where it tries to restore replica factor after reset.
For the purposes of
[IEP-131|https://cwiki.apache.org/confluence/display/IGNITE/IEP-131%3A+Partition+Majority+Unavailability+Handling]
this method must only reset without scheduling, so
{{DisasterRecoveryManager#resetPartitions}} must be improved with some kind of
flag and if this flag is HA, replica factor should not be recovered.
h3. Definition of Done
* resetPartitions must be improved and it must be possibile to not schedule
replica factor restoration for HA mode
was:
h3. Motivation
According to
[IEP-131|https://cwiki.apache.org/confluence/display/IGNITE/IEP-131%3A+Partition+Majority+Unavailability+Handling]
{{DisasterRecoveryManager#resetPartitions}} will be the core method for
recovering partition majority availability. Current logic of this method
schedules new rebalance where it tries to restore replica factor after reset.
For the purposes of
[IEP-131|https://cwiki.apache.org/confluence/display/IGNITE/IEP-131%3A+Partition+Majority+Unavailability+Handling]
this method must only reset without scheduling, so
{{DisasterRecoveryManager#resetPartitions}} must be improved with som kind of
flag and if this flag is HA, replica factor should not be recovered.
h3. Definition of Done
* resetPartitions must be improved and it must be possibile to not schedule
replica factor restoration for HA mode
> resetPartitions improvements: do not try to recover replica factor after reset
> ------------------------------------------------------------------------------
>
> Key: IGNITE-23571
> URL: https://issues.apache.org/jira/browse/IGNITE-23571
> Project: Ignite
> Issue Type: Improvement
> Reporter: Mirza Aliev
> Assignee: Mirza Aliev
> Priority: Major
> Labels: ignite-3
>
> h3. Motivation
> According to
> [IEP-131|https://cwiki.apache.org/confluence/display/IGNITE/IEP-131%3A+Partition+Majority+Unavailability+Handling]
> {{DisasterRecoveryManager#resetPartitions}} will be the core method for
> recovering partition majority availability. Current logic of this method
> schedules new rebalance where it tries to restore replica factor after reset.
> For the purposes of
> [IEP-131|https://cwiki.apache.org/confluence/display/IGNITE/IEP-131%3A+Partition+Majority+Unavailability+Handling]
> this method must only reset without scheduling, so
> {{DisasterRecoveryManager#resetPartitions}} must be improved with some kind
> of flag and if this flag is HA, replica factor should not be recovered.
> h3. Definition of Done
> * resetPartitions must be improved and it must be possibile to not schedule
> replica factor restoration for HA mode
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
