Werk #9338: Repair broken cluster process startup during firmware update
| Component | Cluster |
| Title | Repair broken cluster process startup during firmware update |
| Date | Jan 25, 2018 |
| Level | Trivial Change |
| Class | Bug Fix |
| Compatibility | Compatible - no manual interaction needed |
| Appliance Version | 1.3.19 |
In earlier firmware versions (<1.3.8) it could happen that a disconnect/connect of a cluster left the cluster management process (corosync) startup during boot disabled. The werk #9280 fixed this issue in 1.3.8.
After applying the update, the situation had to be resolved manually by hand by re-enabling the startup using this command on both nodes: update-rc.d corosync defaults
This change now automatically fixes this issue on cluster nodes during first boot after a firmware update. This is only done on "connected" nodes.