-
```
Hi
I have a mysql cluster with one master node and two hot standby slave nodes.
All reads and writes go to master. The three node cluster is managed by
corosync pacemaker that has a VIP for the…
-
### Tracking of effort to include pcs-0.10.7 into ebuild
This request came via email on 2020/12/06 KST.
Challenges/Current progress:
- Gentoo recently moved to newer python3 versions after long t…
-
We've observed that the `nnf-dm` and `clientmountd` daemons generate a surprising amount of system noise on the computes even when they should be idle. For reference, over the last 2 weeks they've b…
-
We are currently using LINSTOR controllers in a HA configuration accessible from a floating IP managed by Corosync+Pacemaker.
Our `cinder.conf` uses the following configuration (organization specif…
-
```
Hi
I have a mysql cluster with one master node and two hot standby slave nodes.
All reads and writes go to master. The three node cluster is managed by
corosync pacemaker that has a VIP for the…
-
```
Hi
I have a mysql cluster with one master node and two hot standby slave nodes.
All reads and writes go to master. The three node cluster is managed by
corosync pacemaker that has a VIP for the…
-
```
Hi
I have a mysql cluster with one master node and two hot standby slave nodes.
All reads and writes go to master. The three node cluster is managed by
corosync pacemaker that has a VIP for the…
-
This is not an issue with the current design. Possibly label as enhancement?
In particular, due to the documented issue "RHEL 7 High Availability and Resilient Storage Pacemaker cluster experiences…
-
I built a HA-solution based on etcd (with test-and-set), which basically works great and as expected.
I'm struggling right now with, that there does not seem to be an option to convert a running ma…
-
When promoting a (new) shadow to master, i'm hitting the error:
```
Attempted shadow->master transition without metadata - aborting
```
isn't possible to promote a brand new server as master?
…