ADSM-L

Re: TSM and Shark disk

2004-01-28 05:48:18
Subject: Re: TSM and Shark disk
From: Rodney clark <Rodney.Clark AT INGBANK DOT COM>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Wed, 28 Jan 2004 11:17:55 +0100
I think the main reason for this freezing is two fold.
1. maxperm minperm is at the default setting of 75 25
2. Io pacing is turned off
quick win is lower maxperm to say 24 man min to 12
turning i/o pacing on will also help but wil lower perfomance sometimes.


-----Original Message-----
From: Roger Deschner [mailto:rogerd AT UIC DOT EDU]
Sent: Wednesday, January 28, 2004 7:52 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: TSM and Shark disk


I have observed this on JBOD disks, if you run dsmfmt on the same
Physical Volume as any AIX paging area, the system will freeze. I
suspect in your Shark, you are seeing the same effect, except on a
larger and more complicated scale. Is there any AIX paging into your
Shark? The dsmfmt program is memory hungry for its very large I/O
buffers, and if paging its huge I/O buffers in and out conflicts with
the I/O itself, you can have a deadly embrace situation.

Paul Ripke's idea to use Raw Logical Volumes is sound, and is what I
have done to bypass the numerous problems with dsmfmt. It's possible to
shoot yourself in the foot with Raw LVs, by enlarging one for instance.
The TSM manuals are filled with warnings not to use them for this
reason. You can protect yourself somewhat against that by setting
"MAXIMUM NUMBER of LOGICAL PARTITIONS" to the same as "Number of LOGICAL
PARTITIONS" when you allocate it in smit. At least then it takes two
separate steps to fire that gun aimed at your foot.

Roger Deschner      University of Illinois at Chicago     rogerd AT uic DOT edu
==== Four-wheel-drive just means getting stuck in deeper snowdrifts. ===


On Wed, 28 Jan 2004, Paul Ripke wrote:

>On Tuesday, Jan 27, 2004, at 06:45 Australia/Sydney, Bill Boyer wrote:
>
>> 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.
>
>JFS or JFS2? Do you have minperm, maxperm, maxclient and/or
>strict_maxperm
>tuned? What was numperm and numclient when the system started responding
>again?
>
>My recommendation is to either use raw devices, or if you don't like
>that idea,
>mount the given filesytems with the dio mount option set.
>
>Cheers,
>--
>Paul Ripke
>Unix/OpenVMS/TSM/DBA
>I love deadlines. I like the whooshing sound they make as they fly by.
>-- Douglas Adams
>


-----------------------------------------------------------------
ATTENTION:
The information in this electronic mail message is private and
confidential, and only intended for the addressee. Should you
receive this message by mistake, you are hereby notified that
any disclosure, reproduction, distribution or use of this
message is strictly prohibited. Please inform the sender by
reply transmission and delete the message without copying or
opening it.

Messages and attachments are scanned for all viruses known.
If this message contains password-protected attachments, the
files have NOT been scanned for viruses by the ING mail domain.
Always scan attachments before opening them.
-----------------------------------------------------------------

<Prev in Thread] Current Thread [Next in Thread>