HA improvements of pacemaker and corosync
Current problems with corosync/pacemaker
* Creating and committing shadows not only adds constant pain with dependencies and unneeded complexity but also rewrites cluster attributes and even other changes if you mess up with ordering and it’s really hard to debug it.
* Restarting resources by Puppet’s pacemaker service provider restarts them even if they are running on other nodes and it sometimes impacts the cluster.
* An attempt to stop or restart corosync service brings down a lot of resources and probably will fail and bring down the entire deployment.
* Controllers other the the first download configured cib an immediate start all cloned resources before they are configured so they have to be cleaned up later.
https:/
Manage/unmanage in corosync init scripts
https:/
Blueprint information
- Status:
- Complete
- Approver:
- Sergey Vasilenko
- Priority:
- Essential
- Drafter:
- Dmitry Ilyin
- Direction:
- Approved
- Assignee:
- Dmitry Ilyin
- Definition:
- Approved
- Series goal:
- Accepted for 5.1.x
- Implementation:
-
Implemented
- Milestone target:
-
5.1
- Started by
- Vladimir Kuklin
- Completed by
- Vladimir Kuklin
Related branches
Related bugs
Whiteboard
First, we should discuss proposed improvements and choose which of them we would like to see in our product.
Gerrit topic: https:/
Addressed by: https:/
remove unnided CIB definition for some OCF resources (part #1)
Gerrit topic: https:/
Addressed by: https:/
remove unnided CIB definition for some OCF resources (part#2)
Addressed by: https:/
Corosync service improvements
Addressed by: https:/
Make simple resouurces to be unbaned everywhere
Gerrit topic: https:/
Work Items
Work items:
Discuss what changes are we going to implement: INPROGRESS
Dependency tree
* Blueprints in grey have been implemented.
