Currently, whenever someone makes a simple change in Yast that makes Yast believe (vs. it really being necessary) it needs to restart the network support, all secondary cluster ip addresses on that node get nuked while the ressources keep running. Something as simple as adding a static route in yast will trigger this.
Of course, a manual "rcnetwork restart" will do the same.
Novell back in the 90s of the last century was able to fix this in Netware after a short while. I'm sure it can be done in OES today too.
by: Massimo R. | over a year ago | Cluster Services (NCS)
Comments
Great idea. Also Since the cluster resources seems to keep losing comms to iManager, have a simple reset cluster master IP reset instead of forcing a failover.
I believe that if you have secondary IP addresses in SLES they don't get nuked, so what ever mechanisom they are using should be reusable for this.