ADSM-L

an frustrated ADSMer

1997-08-01 05:42:08
Subject: an frustrated ADSMer
From: Matthias Brasch <M_Brasch AT COMPUSERVE DOT COM>
Date: Fri, 1 Aug 1997 05:42:08 -0400
Hi ADSM'ers,

many thanks to those how answered so immediately to my 
'restore-question' from yesterday.
Although the answers where a little bit frustrating for me
I've to give an great compliment to this 'listserv-group'.
Meanwhile I begin to think about my decision to use ADSM as our
backup-solution.
There are many good features in ADSM but almost every task
I want to do with ADSM runs on problems...

Lets give me some examples :

I use ADSM to backup our Netware4.11-servers and have to handle with
up to 2.000.000 files per server.
My ADSM-server ( 2.xx ) is running on MVS and I've got 3490 tapes 
with an robot system.
Whenever files get inactive it is really a pain to work with ADSM. 
The netware-ADSM-Client is running on the netware-system-console.
When you are on this console you have to know very good what you are 
  doing... 
  So the restore-task has to be done by people who know about Netware.
  ( when unloading the ADSM-client in the wrong moment your server 
    crashes, interrupting an task is only possible by canceling the session
    with an admin-client which isn't available for netware )
The client is character-based ( I'm not an gui-fan, I know that an
  commandline does often work better ) and requires some knowhow.
You can work with the -pick option to get a list of files.
  When the files are inactive I've to wait up to 60 minutes (!) to
  get the list of files.
  The pathnames in netware are very long (normally more than 50 chars ) 
  so it happens that I make an error in giving the name   
  -> once again...
  When everything works and I get the list it is a pain to look through
  it ( long names on netware-screen with 80 chars means scrolling a lot ) 

Could this really be the way to work with an product which is 
sold as 'almost the best backup solution...' ???

other negativ points for me are :

* As seen from my question yesterday there is no possibilty to
  make an "point-in-time-restore", without doing a lot of own work.

* you can't delete files when they have been backed up without doing
  a lot of work ( rebinding and after that excluding... ) 

* When an mass-restore-fails because of a single problem the whole
  restore fails and you have to start from the beginning...
  ( I'm afraid what would happen when one of my servers crashes completely
)
* The database has an unbelievable size ( 5.2 GB for 6.000.000 files )
* The database has an unbelievable size ( 5.2 GB for 6.000.000 files )

* The performance for server-tasks ( backup stgpool, migration, move data,
  query ...) is very poor, but the cpu-usage is very high

* I've got a lot of other points but I don't think that anyone want's to
  read so much text....


If I read the IBM announcements everything sounds so fine, but looking
behind the front shows an big 'building site'.
Now I'm told that all my problems will disappear with V3, 
can I believe this ???

As for me I'll begin to look for alternatives to ADSM.
What do the others think ??
Are you satisfied with ADSM ??


best regards from an frustrated ADSM'er,

Matthias Brasch, HEW-Hamburg, m_brasch AT compuserve DOT com


 


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