We're trying to work with Websphere and Oracle 9i as Linux guests (one
for each) on an MP 3000 H30 running z/VM 4.4 as a hypervisor.
Admittedly, this is a small box (1 GB of storage, and one CPU that's not
even running at full-power for an MP 3000), but we're seeing horrible
inconsistent response times in the SLES 8 images we're installing the
products in. We've added paging volumes to spread the paging load out
and are cautiously giving the guests more and more virtual storage, but
since we're running the CPU at 100%, we don't think the problem is all
page waits, although there have been some kernel messages from the
Websphere machine in particular that memory allocations for Java
processes (tasks?) are failing.

Most disturbing have been a couple of episodes in which the Websphere
guest just simply stops responding to terminal inputs. Our terminal
sessions are via ssh, but these sessions seem to just hang. New
connections time out, and logged in users can't kill running tasks and
so forth. As I say, the Linux images are getting (taking) a fair amount
of the CPU, so it's not that the whole Linux images are just dormant.

Clearly I need to take the time to put instrumentation in place to
monitor and manage this carefully, but we can't really afford the time,
and the users are getting disgusted.

Are there some tell-tale signs I should looking for to explain why
suddenly interactions with the guest are suspended for minutes at a
time? Is there a list of stupid things I should make sure I haven't done
that have quick fixes?

(Too much, too quickly, without any coordination: a lousy way to run a
pilot project....)

Thanks,
Nick

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to