ADSM-L

Re: AIX Server - What's Stable?

2001-03-26 13:35:22
Subject: Re: AIX Server - What's Stable?
From: Kathleen M Hallahan <Kathleen_Hallahan AT FREDDIEMAC DOT COM>
Date: Mon, 26 Mar 2001 13:35:17 -0500
Has anyone noticed anything like this happening with regular backups?  We
recently upgraded our server code to 4.1.2, and while the vast majority of our
backups are running faster, there are a couple of Sun Solaris clients out there
whose backups are running much more slowly.  None of the clients have been
upgraded yet; the two that stand out as being especially slow are on 3.1.0.5 and
3.1.0.8.  Most of our other clients are also on these code levels.

It would seem on the surface that there must be something different with these
particular nodes; discussions with the sysadmins haven't turned up anything as
yet, though.  In addition, the change in speed coincides exactly with our server
upgrade.  Does this match your experience at all (or anyone else's, for that
matter)?

Thanks!

Kathleen



|--------+---------------------------->
|        |          Tab Trepagnier    |
|        |          <Tab.Trepagnier@LA|
|        |          ITRAM.COM>        |
|        |                            |
|        |          03/26/2001 12:29  |
|        |          PM                |
|        |          Please respond to |
|        |          "ADSM: Dist Stor  |
|        |          Manager"          |
|        |                            |
|--------+---------------------------->
  >----------------------------------------------------------------------------|
  |                                                                            |
  |       To:     ADSM-L AT VM.MARIST DOT EDU                                   
      |
  |       cc:     (bcc: Kathleen M Hallahan/ISS/HQ/FHLMC)                      |
  |       Subject:     Re: AIX Server - What's Stable?                         |
  >----------------------------------------------------------------------------|






David,

You stated:

"One archive job we did usually ran for 2 hours. When we went to 4.1.2 it
took over 15 hours... Not acceptable of course! So we went back to 4.1.0
and
the archive ran okay.  IBM Support is still working on this.  I seems that
no one else in the ADSM-L AT VM.MARIST DOT EDU <mailto:ADSM-L AT VM.MARIST DOT 
EDU>
land
was having this trouble."

The statement about no one else having that problem is not accurate.

We have a large CAD server whose archive job normally runs for a couple of
hours and sends about 30 GB.

After upgrading to TSM 4.1.2, the last run of that job had sent only 1 GB
after running for over 30 hours.  Also, the DB cache hit percentage which
is normally in the 98.8%-99.1% range was driven down to less than 97.5%.

There is a problem there and it is on my list to call that in also.

Tab Trepagnier
TSM Administrator
Laitram Corporation
<Prev in Thread] Current Thread [Next in Thread>