[
https://issues.apache.org/jira/browse/ARTEMIS-2716?focusedWorklogId=605853&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-605853
]
ASF GitHub Bot logged work on ARTEMIS-2716:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 03/Jun/21 11:31
Start Date: 03/Jun/21 11:31
Worklog Time Spent: 10m
Work Description: michaelpearce-gain edited a comment on pull request
#3555:
URL: https://github.com/apache/activemq-artemis/pull/3555#issuecomment-853782345
I disagree failback is very important, feature relied on by many, i dont
think its something we can simply just drop, e.g. in case of full broker
shutdown, you dont just want any broker on full cluster restart to become the
active leader as it may not be the one with the right/latest state. Having
master failback makes this relative simple to reason with, where by you simply
ensure master is the last to go down and first to come up. You could remove
this, if you added additional state to be held in ZK about which node was last
primary so that on full cluster start up, the last primary is the one that is
given back leadership (you'd give some configurable timeout), but thats quite a
bit more than is currently implemented, with a simplistic lock.
I dont fully understand the issue here though, why do we have a loop? the
slave should always be wanting to give back control, it can have some sort of
backoff so its not going into a tight spin lock, but still it should never give
up trying to give it back.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
Issue Time Tracking
-------------------
Worklog Id: (was: 605853)
Time Spent: 9h 20m (was: 9h 10m)
> Implements pluggable Quorum Vote
> --------------------------------
>
> Key: ARTEMIS-2716
> URL: https://issues.apache.org/jira/browse/ARTEMIS-2716
> Project: ActiveMQ Artemis
> Issue Type: New Feature
> Reporter: Francesco Nigro
> Assignee: Francesco Nigro
> Priority: Major
> Attachments: backup.png, primary.png
>
> Time Spent: 9h 20m
> Remaining Estimate: 0h
>
> This task aim to ideliver a new Quorum Vote mechanism for artemis with the
> objectives:
> # to make it pluggable
> # to cleanly separate the election phase and the cluster member states
> # to simplify most common setups in both amount of configuration and
> requirements (eg "witness" nodes could be implemented to support single
> master-slave pairs)
> Post-actions to help people adopt it, but need to be thought upfront:
> # a clean upgrade path for current HA replication users
> # deprecate or integrate the current HA replication into the new version
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
