I've noticed this problem on my lab systems, RS/6000 F40's and
43p-150's, using JBOD.  This happens on AIX 5.1 ML5 with TSM 5.2.0.0,
but it's unfortunate to see that nothing's changed on newer versions.

In my experience, the unresponsiveness lasts as long as the format
takes to complete.  I just try to avoid these operations when I can,
and suffer through them when I can't.

-JD.

At 2:53 PM -0500 1/26/04, Justin Case wrote:

We use raw logical volumes not dsmfmt disks for TSM from the ESS (shark).
Justin




Bill Boyer <bill.boyer AT VERIZON DOT NET>@VM.MARIST.EDU> on 01/26/2004 02:45:52
PM

Please respond to "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>

Sent by:    "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>


To:    ADSM-L AT VM.MARIST DOT EDU
cc:

Subject:    TSM and Shark disk

p630 server running AIX 5.2 ML2
TSM 5.2.2.0
2109-S16 fibre switches

When running dsmfmt on the volumes on Shark disk, the system runs at very
high wait cpu %. at times it will jump over 90% and the system will become
unresponsive for several seconds. When the dsmfmt completes, topas will
still show 5-8mb/sec on the Shark paths for almost another minute. Nothing
is running.

When I do the same formats within TSM (formatsize=x) on the def vol command
the system doesn't seem to push the Shark paths as high and we never see
the
unresponsive times, but still the system sits at 45-50% wait cpu time.

Anyone had strange performance issues using Shark and TSM on pServers?

Bill Boyer
"Some days you are the bug, some days you are the windshield." - ??