It happens only on one node and whenever I do any of the 4 points you
mentioned. If I take any resource offline/online or if I try to make a
switchover with any of the existing service groups, even if I try to
switch over the ClusterService service group. It does not matter if I
change to or from the node on which the error appears. The error also
appeard after a "hastop -local -force" command but I did not try with other
commands.
I try to monitro with vmstat 1 the statuses of the processes and even
during the VCS operation (online/offline/switchover) the use of the
processes does not go above 20-30% but usually it is lower then 20%.
I also tried to increase the gab timeout but did not helped.
BR,
Laszlo
----- Original Message ----
From: "Cronin, John S" <[EMAIL PROTECTED]>
To: Kiss László - Károly <[EMAIL PROTECTED]>; Stuart McDermid <[EMAIL
PROTECTED]>; veritas-ha@mailman.eng.auburn.edu
Sent: Friday, 10 August, 2007 5:16:35 PM
Subject: RE: [Veritas-ha] Fw: gab restarts had
DIV {
MARGIN:0px;}
Please define "every time I do something with
VERITAS"? Surely, running something like "hastatus -sum"
doesn't cause this?
If this happens every time you:
1) Bring the service group online
2) Take the service group offline
3) Switch the service group from one node to
another
4) Bring a specific resource online or
offline
Also, is this happening on every node in the cluster
simultaneously, or only on the node where the application is going offline or
online?
If it is happening when you take the service group online
or offline, then the problem is almost certainly that the application is
putting
a very heavy load on the system when it starts up or shuts down, and this is
causing VCS to be starved of resources.
--
John Cronin
678-480-6266
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Kiss
László - Károly
Sent: Friday, August 10, 2007 5:42 AM
To:
Stuart McDermid; veritas-ha@mailman.eng.auburn.edu
Subject: Re:
[Veritas-ha] Fw: gab restarts had
Hi,
Thanks
for the answer.
You're right, in this case the load was high, bu tthis happen
every time I do something with Veritas and I can find a lot of logs where the
load is high at all.
For example:
number of
cpu:
2
physical
memory:
4072832 K
free
memory:
825296 K
average free memory in 5 sec: 825272
K
average free memory in 30 sec: 822712
K
number of
processes:
199
load average in 1
min: 0.16
load
average in 5 min:
0.18
load average in 15
min: 0.18
pagein
rate:
0
pageout
rate:
0
Port h: client process failure: killing
process
BR,
Laszlo
-----
Original Message ----
From: Stuart McDermid
<[EMAIL PROTECTED]>
To:
veritas-ha@mailman.eng.auburn.edu
Sent: Thursday, 9 August, 2007 3:38:28
PM
Subject: Re: [Veritas-ha] Fw: gab restarts had
Sorry for my late entry in this conversation, meant to reply
sooner.
Jim you are correct it probably is load. In the messages file
that
Laszlo attached we have the load averages:
> number of
cpu:
2
> number of
processes:
192
> load average in 1
min: 1.86
>
load average in 5 min:
0.66
> load average in 15
min: 0.30
So assuming that the
previous 5/15 minutes prior to the problem where
"normal" (15/30% load) then
previous minute certainly wasn't (93%+
load). Laszlo what you need to do now
is find out what is causing the
load.
If you are on a production
machine. Stop and back out any changes that
have been made. Then apply those
changes to a development box, if you
have one, and test.
If you are on
a non-production cluster then run the utility "top", if it
is not installed I
believe it is on the Solaris 9 Companion CD/DVD, or
any other utility that
will show you the load in near real time. Then
try what ever VCS task you are
trying to do, top should show you the
process that is causing the load. For
example: Is it a single process
that is causing the load, or hundreds of
small processes. Then you have
just got to work out why it is causing the
load, does it do the same if
you run the start/stop script from the command
line etc.
Sorry not an Oracle person any more so I won't be able to help
you
there, but hopefully this helps you track down what is going wrong.
Stuart
______________________________________________________________________
This
e-mail (and any attachments) are confidential and may be legally privileged. If
you are not the intended recipient do not copy, use, or rely on it and please
delete all copies. Please notify me ASAP if this message has been
misdirected.
This e-mail is from: Biffa Waste Services Limited.
Registered in England and Wales No. 946107. Registered Office: Accuray House
Coronation Road, Cressex, High Wycombe, Bucks,HP12
3TZ
______________________________________________________________________
This
email has been scanned by the Biffa MessageLabs Email Security System.
For
more information please visit http://www.messagelabs.com/email
______________________________________________________________________
_______________________________________________
Veritas-ha
maillist - Veritas-ha@mailman.eng.auburn.edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-ha
Yahoo! Answers - Get better answers from someone who knows. Try
it now.*****
The information transmitted is intended only for the person or entity to which
it is addressed and may contain confidential, proprietary, and/or privileged
material. Any review, retransmission, dissemination or other use of, or taking
of any action in reliance upon this information by persons or entities other
than the intended recipient is prohibited. If you received this in error,
please contact the sender and delete the material from all computers. GA621
___________________________________________________________
Yahoo! Answers - Got a question? Someone out there knows the answer. Try it
now.
http://uk.answers.yahoo.com/ _______________________________________________
Veritas-ha maillist - Veritas-ha@mailman.eng.auburn.edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-ha
