Group resources with VIPs
Bug #1952753 reported by
Gabriel Cocenza
This bug affects 1 person
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| charm-interface-hacluster |
New
|
Undecided
|
Unassigned | ||
Bug Description
Right now when a systemd resource is added, by default it will be cloned to run on every node in the cluster. If for some unknown reason the service starts to fail in one node that also has a VIP resource, Pacemaker won't try to locate the VIP to another node.
In the OpenStack use case, for example, there would be an outage if the haproxy service was down on the node that the vip was runnung on. IMHO, grouping the VIP with the resource is a good way of having those resources together and resilient to failures.
| affects: | charm-hacluster → charm-interface-hacluster |
To post a comment you must log in.
