ADSM-L

Re: summary-table

2003-02-07 19:57:43
Subject: Re: summary-table
From: "Seay, Paul" <seay_pd AT NAPTHEON DOT COM>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Fri, 7 Feb 2003 19:56:12 -0500
The definition of compatibility may not mean that no features get broken
that are not directly related to the integrity of the system from a Tivoli
point of view.  Unfortunately, this stuff has been broken since about
4.2.1.11.  And for the most part, the summary table contents are not
necessary for any TSM externals to work properly.  Customers use it for
reporting via SQL.

After about 6 months of trying to figure out what the cause of the lost data
was, Tivoli Development was about to go crazy.  Luckily, I some reports that
I was running that triggered a thought that the cause of the problem was
because the client session would time out during a backup over the session
timeout length.  At the end of the backup the client was reforming the
session to record the statistics, but the information delivery was to the
wrong session and everything was getting zeroed out.  To fix the problem
required them to make a fix to both the client and the server.  The database
even had to be updated to provide some additional capability based on what I
heard.

Yes, it is a really nasty problem.  How did it get introduced.  I am
guessing they had to rearchitect some things to fix some serious SAN client
issues.  During the process apparently the designs of the client and the
server became incompatible.

Tivoli was very thankful that I was able to point them in the right
direction.  But, as you say, I doubt sessions longer than the session
timeout were tested significantly in their regression testing and I doubt
they did any summary table data verification because it is not critical to
TSM operation nor most of the commands.  It is a customer thing.

There were a lot bigger fish to fry than fixing summary table issues.  The
only reason Tiovli put their heads together and made it a priority was when
I explained that this had broken many customer billing systems for TSM
usage.  That was a real financial integrity issue for customers.  That got
their attention.

Now that Tivoli knows that this is a critical compatibility item, I hope
they will actually test to see if it is working correctly from release to
release and compatibility of client to server.  There are no guarantees.

Paul D. Seay, Jr.
Technical Specialist
Northrop Grumman Information Technology
757-688-8180


-----Original Message-----
From: Allen Barth [mailto:allen.barth AT DB DOT COM]
Sent: Friday, February 07, 2003 4:48 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: summary-table


It's stuff like this that makes my shake my head in disbelief...

The summary table was working fine will ALL my clients, regardless of
version, when the server was 4.1.5.  After upgrading ONLY the server to
5.1.5. the byte counts in the summary table are Zeroes.

Hmmm, compatability tested?  I think not!   If so, where are the notices
and gotcha's in the readme's?

--
Al




Matt Simpson <msimpson AT UKY DOT EDU>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> 02/07/03 03:05 
PM
Please respond to "ADSM: Dist Stor Manager"


        To:     ADSM-L AT VM.MARIST DOT EDU
        cc:
        Subject:        Re: summary-table


At 3:19 PM -0500 2/7/03, Seay, Paul wrote:
>Apparently, the
>client needs to be updated as well to get a complete fix.  I do not
>know
if
>there is a client level for 4.2.3.2 and higher that completely
>eliminates the problem.

According to the info I got from TSM support, for what I think is the same
problem:

>The APARs that describe
>this are IC33840 for the client and IC34693 for the TSM server. Both a
>client fix and a server fix need to be applied. `
>The client fixtests are in ....>
>  Client fixtests 5.1.5.2 and 4.2.3.1.
>go to...>
>ftp://service.boulder.ibm.com/storage/tivoli-storage-management/
>patches/client/v4r2/
>`
>and then platform and the v423 directory or folder.
>For the V4.2 Win98/Me platform only, the fixtest is 4.2.3.2.
>These fixtests must be used in conjunction with the server fix.


I haven't tried to apply any of the indicated fixes.


--


Matt Simpson --  OS/390 Support
219 McVey Hall  -- (859) 257-2900 x300
University Of Kentucky, Lexington, KY 40506 <mailto:msimpson AT uky DOT edu>
mainframe --   An obsolete device still used by thousands of obsolete
companies serving billions of obsolete customers and making huge obsolete
profits for their obsolete shareholders.  And this year's run twice as fast
as last year's.

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