And, as you know that does work well in SBSland. However, when the
scale grows, so do the requirements. IN the Medium to Enterprise
space, the idea is more along the lines of a system or series of
systems pumping this type of information into paging and making
intelligent decisions based on the audit, event, alerts, services,
etc.
Which, is right where MOM 2005 drops into the picture. If it _IS_
the event aggregator, or if it's pushing up to a bigger overall item
such as HP OpenView - that data is available. It's just that instead
of getting an e-mail per server (most admins would just begin to
create a rule to send these to DEV/NUL after a while...) MOM
collects, enforces and reports this same type of information.
Scale makes the problem much tougher, as I'm sure you can imagine....
Rick [msft]
--
Posting is provided "AS IS", and confers no rights or warranties ...
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Susan
Bradley, CPA aka Ebitz - SBS Rocks [MVP]
Sent: Sunday, October 16, 2005 8:33 PM
To: ActiveDir@mail.activedir.org
Subject: Re: [ActiveDir] Knowing when users were deleted.
<here she goes again.. I know ... I'm terrible at lurking>
In SBSland we have a daily monitoring email [well ... I send it daily
anyway, but it's configurable] and it looks at the event logs and
tells daily health status of my server.
Like today my email tells me my server has been running for 6 hours
[just rebooted it last night] and it gives me an overview if auto
services are not running, critical alerts and critical errors in the
event logs.
It tells me memory/disk size, cpu use, top processes, if the backup
ran, and aggregates the alerts from all the log files.
It's a health mon that dumps it's data into a msde database and
builds the email to be sent internally or externally.
What it does now, is only pulls data from the one box, the SBS box.
but I
can go into health mon and build my own monitors and grab those event
logs from other machines [need to so that just haven't gotten around
to it].
Right now if someone [usually me] fat fingers a password, for
example, it gives me an alert in the email of the last time it
occurred and how many occurrances. Basically it's tracking the
critical alerts in all the event logs and summarizing the events
along with the number of events in the email [and showing the last
time the event occurred so you can start your investigation from that
point back]
For SBS ....it's in the box, it's a gui wizard that builds this
pretty little html email that my server builds and hits me every
morning at
6 a.m
and says "Hey here's how I'm doing...how are you?". It's the mid
market that doesn't have this. [and yes, we've told Mothership
Redmond they need to steal this sucker and put it in the mid market
server bundle]
Does it make me more aware of events on my server? Oh you betcha it
does.
Which is why this needs to be ....as you say...native in small and
medium servers....heck I'd strongly argue that no server should be
shipped without some admin somewhere getting an in your face report
on that sucker.
I'll go to Frys and buy bigger harddrives if I need to. But give me
a big fat audit log file and I'm a happy camper.
Al Mulnick wrote:
I'll see your Eurocents and add raise you two. :)
I fully understand where you're coming from Ulf. Adding this
information into the DIT when it is currently possible to get is
something that grates against common sense and common engineering
principles even if you
subscribe
to belts and braces methodologies.
However, I think two things make this a worthwhile request with a big
payoff. First to Laura's point about diminishing returns. I agree, at
some
point there will be diminishing returns. I also believe that as
hardware
gets bigger (i.e. Standard 80 GB hard drives, 1 GB memory in
workstation
machines, etc. [1]) the bar gets raised until we get to the diminishing
return. Since we're targeting 80/20 out of the box [2] it seems
reasonable
that 80% of the deployments would benefit from such a change. The
other 20
would be those that a) don't care or know about such things and b)
those
that can't tolerate the additional overhead and therefore wouldn't
want to
deploy it. I say tough pickles to them. :) Seriously, this could
be on
by
default but configurable (group policy?) to disable it as a performance
issue etc.
Second, I think that the major benefit is the ability to actually get
usable
information native to the product vs. having to invest in a third party
product. Why? Because today in order to get that information I have to
have
something that scrapes the Security logs looking for such
information. Is
this a good idea? I think it is. Is it something that could be
native? I
think it could and should be native if technically feasible.
Making us look in a particular DC's event logs is more difficult
than it
should be without yet another product. That's fine for the really
large
companies that have deeper pockets, and larger needs. For the small to
medium businesses, it should not be so difficult nor should it
*require*
SQL
licensing or expertise.
[1] I'm not saying that the quality has kept up, only that the
hardware is
bigger, faster, stronger and cheaper. [2] I'm making that up, but it
sounds reasonable
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Ulf B.
Simon-Weidner
Sent: Sunday, October 16, 2005 4:42 PM
To: ActiveDir@mail.activedir.org
Subject: RE: [ActiveDir] Knowing when users were deleted.
Hmm.
Do we really want to excuse prior failure of proper auditing by putting
more
data into AD? Wouldn't that lead into every request of non-configured
auditing to requests for extending the AD? Do it right the first way.
I completely agree that we should make the people more auditing
aware, and
it would be great to have a centralized auditing together with some
force
of
configuration instead of the per server events and auditing which is
rearly
configured.
However I'm not sure if I want this kind of data in the AD.
Just my Eurocents.
Ulf
|-----Original Message-----
|From: [EMAIL PROTECTED]
|[mailto:[EMAIL PROTECTED] On Behalf Of Laura |E.
Hunter
|Sent: Sunday, October 16, 2005 10:28 PM
|To: ActiveDir@mail.activedir.org
|Subject: Re: [ActiveDir] Knowing when users were deleted.
|
|Various thoughts from this thread:
|
|[1] I agree with Al and Paul[1] on a desire for that sort of
metadata. |I'm not as convinced of the trade-off value of bloating
the DIT for |full undelete information, particularly in monster big
environments.
|For my teeny-tiny single domain it probably wouldn't be that |bad
of a hit, but I imagine that the laws of diminishing |returns would
quickly set in.
|
|[2] Please finish the thought, Brett, I'm sure I'd find it
|helpful/enlightening/informative even if it's only speaking in
|hypotheticals.
|
|[3] It's Gil and Darren's turn to crack me up today, I guess
|joe is taking a break.
|
|
|[1] *waves* Hi Paul! Glad to see you alive post-Summit.
|
|- L
|List info : http://www.activedir.org/List.aspx
|List FAQ : http://www.activedir.org/ListFAQ.aspx
|List archive:
|http://www.mail-archive.com/activedir%40mail.activedir.org/
|
List info : http://www.activedir.org/List.aspx
List FAQ : http://www.activedir.org/ListFAQ.aspx
List archive:
http://www.mail-archive.com/activedir%40mail.activedir.org/
List info : http://www.activedir.org/List.aspx
List FAQ : http://www.activedir.org/ListFAQ.aspx
List archive:
http://www.mail-archive.com/activedir%40mail.activedir.org/