ADSM has positioned itself in the market as the Enterprise backup solution -
designed to support large systems for installations that are serious about
protecting Enterprise data.

One of the biggest selling points I used for ADSM V2 was that it was
reliable, the code really works, and it got remarkably good technical
support.

ADSM V3 has indeed been a disappointment.  This installation is being forced
to move to ADSM V3 (no more support for the AIX server) at a significant
upgrade cost, but we can't roll out the V3 Windows client because it is
still too flaky for mass distribution.  Doesn't play well with the customer
here, I can tell you.

It would be a shame indeed if IBM management risked ADSM's position in the
market by allowing quality to decline for whatever reason.

I hope the ADSM developers will take the comments from people on this list
not as criticism, but as AMMUNITION - and let your management know that many
of us are serious indeed about protecting our data, and our users.

Thanks for listening.
Wanda Prather

Opinions expressed here are my own, entirely.







> ----------
> From:         Richard Sims[SMTP:rbs AT bu DOT edu]
> Sent:         Thursday, August 20, 1998 8:12 AM
> To:   ADSM-L AT vm.marist DOT edu
> Subject:      Re: Simple Question for Client Developers...
>
> >Why was a new PTF released to the public with so many "known" problems
> >and bugs. I thought that these releases are supposed to be tested
> >before release. Nobody even took a look at the dsmsched.log to see
> >that it was putting in blank lines during testing?
> >The extra lines are a problem in the Unix versions also.
> >Are we the beta testers? I'd actually expect this kinda thing from
> >Microsoft before IBM.
> >
> >Brian Rowan
> >Dec Consultant for Compaq Services
>
> Brian's frustration is one that many of us share.  We systems people look
> at the defects which end up in formal maintenance levels and shake our
> heads in dismay.  So many of the problems have such obvious manifestations
> that we have to conclude that there is no rigorous testing suite as there
> should be.  We as customers are getting the impression that testing of
> ADSM code occurs in an ad hoc manner, as individuals think of various
> things to try.  The result is defects making it out which range from the
> annoying ("..." not working in excludes) to the disastrous (small file
> aggregation data loss).
>
> ADSM is supposed to be a flagship, enterprise product whose quality should
> not have to be questioned, particularly by large businesses with huge
> investments in the data they entrust to this product.  But from what we're
> seeing in chronic programming errors making it to market, we do feel like
> beta testers.  IBM certainly can and should be doing better.
>
>     Richard Sims, Boston University OIT (and former corporate systems
> programmer)
>