On our 5.1a cluster, I was trying to remove a (non-default) cluster
alias
using sysman, and in the course of this I restarted the clu_alias daemon
several times because the alias, while gone from system files, is still
visible to "cluamgr -s all". Shortly thereafter users reported they
couldn't get to the node. Investigation showed that the default routes
on both nodes changed from the assigned gateway, as defined in
/etc/routes, to the memory channel interconnects! Once I deleted this
new bogus default route and re-added original one, all seems well.
Anyone ever seen anything like this?
TIA!
Judith Reed
jreed_at_navisite.com
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
Benjamin Franklin
Received on Tue Sep 23 2003 - 14:35:27 NZST