[olug] Fwd: faulty Microsoft patch affects Exchange and DC

[Will Langford]

On a similar note...

There's been many times I'd update a Linux kernel and find that some
drivers that worked on the previous one no longer work for me in the
updated one.  Worked for some other people, but not me. Sure, a
version or three down the line, it's been fixed...  and it's annoying
to have to wait... but I don't particularly ever recall a ChangeLog
saying "fixed the bug that So-and-So introduced when they did foo" ...
so they don't really admit it either.

-Will


On 5/18/05, Sean Kelly <smkelly at zombie.org> wrote:
> On Wed, May 18, 2005 at 10:31:21AM -0500, Rod Hurley wrote:
> > Just another predictable Micro$oft snowjob.
> 
> That article is overhyped. We have many HP-UX systems and HP does this sort
> of thing all the time. A patch may come out that is critical, and later
> there will be a critical patch to replace said critical patch due to
> unforseen problems with the first one. This is normal. I would only have a
> problem if Microsoft didn't address the problem at all. This is not the
> case. They're providing hotfixes, and they're redeploying a fixed patch.
> What more do you want? Do you want 100% functionality all the time? This is
> not possible.
> 
> The article makes one good point:
> 
>    "When we are dealing with Microsoft updates, one thing we always
>    reiterate, then reiterate some more, is to test before deploying. The
>    guidance is always to download, test, then deploy the patches.  With
>    Microsoft, the test section of our guidance has gotten larger and
>    larger."
> 
> We test all patchsets before applying them. We pick the least important
> machine (or a test environment one) and apply the set. We do this for every
> general configuration we have, be it HP-UX 11i, HP-UX 11, or Linux. This is
> just good practice. And guess what? Our testing doesn't always find
> problems with patches that we're later snagged by when we deploy it to all
> systems. Just like Microsoft can't guarentee perfection, neither can our
> testing procedures.
> 
> --
> Sean Kelly         | PGP KeyID: D2E5E296
> smkelly at zombie.org | http://www.zombie.org
> _______________________________________________
> OLUG mailing list
> OLUG at olug.org
> http://lists.olug.org/mailman/listinfo/olug
>