NOTE: Double-Fault. Red Hat does not generally support this kind of situation. Behavior in event of loss of IP-tie-breaker is undefined. Actual results: Undefined/Unhandled behavior. Expected results: Undefined. Additional info: Consider monitoring the IP tie-breaker more proactively when a half-split occurs in the cluster; handle in a similar manner to loss of disk-tie-breaker (ie, loss of quorum at that point).
CVS build of clumanager: http://people.redhat.com/lhh/.testing/clumanager-1.2.5-0.1.89.2.8.i386.rpm http://people.redhat.com/lhh/.testing/clumanager-1.2.5-0.1.89.2.8.src.rpm Fixes: #103721, #107274, #107276
Cluster handles loss of quorum when tie breaker IP is no longer available. However, since this is a double-fault, the cluster requires either: (a) a majority of real members to return to normal operation afterward (simply bring the tiebreaker back online is not sufficient), or (b) restarting the members after the tiebreaker IP comes back online.
PER BUG 108148 ------ Additional Comments From lhh 2003-10-28 10:26 ------- CVS build of clumanager: http://people.redhat.com/lhh/.testing/clumanager-1.2.5-0.1.89.2.8.i386.rpm http://people.redhat.com/lhh/.testing/clumanager-1.2.5-0.1.89.2.8.src.rpm Fixes: #103721, #106465, #107274, #107276, #108148
Tested, closing as errata.
Fixing product name. Clumanager on RHEL3 was part of RHCS3, not RHEL3