From: Andrew Martin (andrew.martin@cameo-consultants.com)
Date: Tue Aug 12 2003 - 06:03:46 EDT
Hello Folks,
I did not get too many responses on this. Many thanks to Jan Mark Holzer and Dr.
Thomas P. Blinn.
Jan said that this seems to be a known problem and has already been reported by
other customers. He suggested I open a call with the support center and have
them refer to case an exiting case (it's against 5.1B but the same problem
applies to 5.1A and has just been 'submitted' to the patch kit pools).
Original problem:
We have a two node TruCluster V5.1A configuration. the max_aliasid setting is 8
on both members. We have 8 cluster aliases defined, and try and add one more, and get
the message "no more aliases available".
We then decide to increase the number of cluster aliases to 32. We do this as
follows on both cluster members (its a dynamic kernel param that can be changed
in the running kernel):
# sysconfig -r clua max_aliasid=32
Both systems accept it without any complaints.
We then try and add the alias:
# cluamgr -a ........
Then the other cluster member crashes (not the one we issued the cluamgr -a cmd
on).
Looking at the crash-data file, I see that it was running clua code
(update_member_info in clua_kch.c).
Any idea why this happened (systems have patch kit 3 installed)?
-- Best regards, Andrew mailto:andrew.martin@cameo-consultants.com
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:49:31 EDT