asume node 1 is down.
Managers
The message above appears in the /var/adm/messages on the remaining
member, if we stop (one member of a 2node DS10L Cluster.
After this message the remaining member is not reachable for connects
(telnet, tnsping, etc). It's like a freeze.
The application client-connect (oracle-net) hangs for more than 1
minute and get a timeout, so that the application must reconnect to the
db.
The local sqlplus-session (selects) hangs too, but does'nt need a
reconnect after the "freeze" of the remaining member.
The above message repeats in /var/adm/messages after 1Min 30sec and
after this we got a lot more messages about some cluster(re)config in
the message-file.
We have nothing found who can explain us this event-entry and/or the
specific behavior of the remaining member.
Any comments and tips who can explain or help to solve this
phenomen/problem are welcome.
(The filesystems are always served from the remaining member).
Thanks
Olaf
Received on Mon Mar 17 2003 - 18:23:48 NZST