Philip Gwyn wrote:
Hello, my first question :
Looking at apcmaster.c and apcmastersnmp.c, it seems that STONITH will only
reboot a node. Seems to me this is exactly what one doesn't want. If a node
goes bad, it needs to be turned off completely. Then "put up on blocks" by a
human to find out what went wrong. Only then may it rejoin the cluster.
Looking at apcmaster.c I see apcmaster_onoff() called for ST_POWERON and
ST_POWEROFF and MSReset() called for ST_GENERIC_RESET, so that is fully
functional; looking at apcmastersnmp.c it does reboot the node
regardless of the ST_* request.
Am I missing something?
You can decide what happens to a node when it needs to be fenced by
setting stonith-action to reboot or poweroff. Of course, if the STONITH
device you choose doesn't support poweroff...
-Philip
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
