---------------------------- Top of message ----------------------------
>>--> 01-05-00  10:14  S.SHEPPARD     (SHS)    Re: TSM server on OS/390
>>--> 01-05-00  10:14  S.SHEPPARD     (SHS)    Re: TSM server on OS/390

APAR number is PQ31998.

Sam Sheppard
San Diego Data Processing Corp.
-----------------------------------------------------------------------`
---------------------------- Top of message ----------------------------
---------------------------- Top of message ----------------------------
>>--> 01-05-00  09:59  ..NETMAIL     (001)     Re: TSM server on OS/390
>>--> 01-05-00  09:59  ..NETMAIL     (001)     Re: TSM server on OS/390
Date: Wed, 5 Jan 2000 12:55:32 -0500
From: "William Boyer" <wboyer AT PTD DOT NET>
Subject: Re: TSM server on OS/390 FIXEDIOBUFSIZE
To: ADSM-L AT VM.MARIST DOT EDU
_________________________________Top_of_Message_________________________________

Don't use it!! When I initially installed TSM 3.7.1.0, I specified this
option. After some period of time the database backups started failing
indicating a read error. Turns out there is an APAR (don't have the number
right here) that addresses this. The logic they used to pagefix and use the
real addresses had an error if that page was reclaimed. The workaround is to
leave it at 0. The developer actually called me in reference to the PMR I
had opened to explain how this 'bug' slipped by. It's an error that creeps
up after a period of inactivity.

Bill Boyer
DSS, Inc.

> -----Original Message-----
> From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU]On Behalf Of
> Julie Phinney
> Sent: Wednesday, January 05, 2000 12:36 PM
> To: ADSM-L AT VM.MARIST DOT EDU
> Subject: TSM server on OS/390 FIXEDIOBUFSIZE
>
>
> Does anyone have a recommended FIXEDIOBUFSIZE value for the Tivoli Storage
> Manager server running on OS/390?
> It's a new option, I've found it referenced in the Administrator's
> Reference and the TSM Technical Guide (redbook), but can't find a
> recommended value.  The default is 0, but since this is supposed to be a
> new performance-enhancer option, I'm sure that's not the
> recommended value.
> Thanks,
> Julie
>

-----------------------------------------------------------------------`
=========================================================================