Hello everybody,
IHAC about to implement SFHA 5.0MP3 (on Solaris 10) with SAN based boot disks.
They already implemented SFHA 4.1MP1 (on Solaris 8) which had a VCS related issue: VCS when losing visibility of the attached storage would result in the faulted node to stall, hence VCS didn’t have the ability of properly managing the situation. The expected behavior would have been the faulted node to panic with the standby node to take over.
A local PS engineered a workaround that basically changed the Monitor DiskGroup resource script in way that –a situation like the one just described– would trigger a "uadmin" command from /tmp to make sure that the command be executed in RAM the moment the system would lose visibility of the OS disks.
At the time it was said that such workaround would have been followed by a private patch first and (possibly) this issue as a whole would have fixed in the next releases of SFHA. Anyone can tell me whether we still have such a problem and if there's any specific patch to sort this?
Thanks for listening
