«

Carnations wrote:
> On Thu, 21 Aug 2008 13:12:14 +1200, thing wrote:
>
>> Carnations wrote:
>>> On Tue, 19 Aug 2008 19:08:44 +1200, thingy wrote:
>>>
>>>> If I was based in Europe I might be inclined to swap in SUSE for Red
>>>> Hat on the server, but not in NZ.
>>> Why? (Curious)
>>>
>>> It's a solid system.
>>>
>>>
>>>
>> Yep nothing wrong with SUSE in itself, Novell is a bit of a vampire /
>> leech though...their proprietry stuff on top can price them as bad if
>> not worse than MS in NZ IMHO....
>>
>> Also the good guys at SUSE seem to have all left and been replaced with
>> Novell corporate hacks...clash of cultures does not bode well IMHO.
>>
>> SUSE has a bigger following in Europe than in NZ...so I guess a user
>> base thing...no real specific reason.
>>
>> Also 3rd party support for SUSE was poorer than RH, but that has
>> improved very substantially in the last 4 years.
>>
>> If I had to decide between say Red Hat / SUSE and MS it would be based
>> on what the application was and the experience with that particular
>> version of Linux..
>>
>> eg., we have been looking at some LDAP/identity products and Red Hat is
>> weak in this area....Sun and SUSE/Novell very strong... but from bitter
>> experience I know trying to run Sun's ID system on Red Hat is just a no
>> show they are clueless when it comes to support "well it works on
>> Solaris"...gee thanks...So I would have to seriously consider Novell's
>> solution on top of SUSE....I'd be reluctant to go Novell's ID on top of
>> RH, I know RH well but the Sun experience ikky...or continue go the MS
>> solution route, but we have that now and its <shudder>...
>>
>> So in this scenario I'm pretty much "forced" to go the SUSE route with
>> Novell's ID....Red Hat cant do it in house, Sun's is, well awful and
>> MS's is worse...
>
> eDirectory is a good solution, in my (limited) experience.
>
When I experienced it it was a heap of brown stuff. But I experienced it
as an end user. I also experienced Groupwise as an end user and that was
a heap of fermenting brown stuff. Groupwise was fine if you didn't
mind getting someone else's email, or seeing the same email *thousands*
of times and being unable to delete it, or having GW send to your whole
Address Book instead of the single person you intended. It never got to
the person you intended! Oh yes, GW was *fun* for some bizarre
definition of fun.
>
> But isn't there a simpler reliable solution that uses LDAP?
>
eDirectory (and for that matter, MS Active directory) are LDAP based.
Neither AD not eDirectory to my knowledge exposes a useful LDAP
interface - in other words the LDAP is essentially internal.
>
> On a totally unrelated note, I just found out today that MS Exchange
> craps on itself and will start throwing errors if it is not given semi-
> regular offline defrags of the database. Apparently this because, like
> the MS Windows Registry, it only ever expands in size on disc and never
> reduces, and because after a while the degree of fragmentation gets to
> the point where MS Exchange cannot serve up the requisite data within
> even a prolonged timeframe.
>
Whoever told you that is talking *******s. I speak as a former MVP in
the area.
>
> Apparently this has been fixed in Exchange 2007 by using a MS SQL Server
> as the database for the MS Exchange mail store.
>
Again, this is *******s. The Jet database engine as optimised for
Exchange is one of the most efficient database engine in the field.
>
> Frankly, I think it is rather extreme overkill for a rather stupid and
> very basic fault.
>
Exchange 5.5 has a fault, which in *extreme* cases required an offline
defrag. Exchange 2000 had a limit to the size of the Exchange database
which, if it was allowed to fill up required an offline defrag.

Scarcely 'basic faults' and not requiring 'semi-regular offline defrags.

Cheers,

Cliff

--

"I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.

Carnations wrote:
> On Fri, 22 Aug 2008 22:18:15 +1200, Enkidu wrote:
>
>> Exchange 5.5 has a fault, which in *extreme* cases required an offline
>> defrag. Exchange 2000 had a limit to the size of the Exchange database
>> which, if it was allowed to fill up required an offline defrag.
>>
>> Scarcely 'basic faults' and not requiring 'semi-regular offline defrags.
>
> Those are fundamentally basic database faults, IMO. There is no good
> reason why an email server should need to go offline merely to facilitate
> the reduction in disc size of a mail store.
>
I've had to do it twice, once on an old 5.0 Exchange system that no one
looked after, and once in preparation for going to 2000 from 5.5. It is
NOT a common occurrence. I have once circumvented the need for an
offline defrag by migrating to a new server. It is not something that
should happen in a well maintained system.
>
> BTW, I'm not sure that calling yourself a "MVP" is something to crow
> about, as most of those that I have come across (I am not referring to
> yourself) have been truly clueless.
>
It is an honour that I *earned* by knowing a lot about AD and Exchange
and advising others. I am proud of it.

Cheers,

Cliff

--

"I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.

In article <>,
says...
> On Fri, 22 Aug 2008 22:18:15 +1200, Enkidu wrote:
>
> > Exchange 5.5 has a fault, which in *extreme* cases required an offline
> > defrag. Exchange 2000 had a limit to the size of the Exchange database
> > which, if it was allowed to fill up required an offline defrag.
> >
> > Scarcely 'basic faults' and not requiring 'semi-regular offline defrags.
>
> Those are fundamentally basic database faults, IMO. There is no good
> reason why an email server should need to go offline merely to facilitate
> the reduction in disc size of a mail store.

It doesn't, nor does it need to.

If circular logging is *not* enabled, then you need to perform an
exchange aware backup (otherwise you don't!) - to flush the log files,
you'll eventually fill yer hard disk otherwise.

Short of filling the disk though, Exchange will run just fine. Dunno
where you get your info from, it's wrong.

--
Duncan

In article <>,
says...
> On Fri, 22 Aug 2008 23:26:27 +1200, Dave Doe wrote:
>
> > Short of filling the disk though, Exchange will run just fine. Dunno
> > where you get your info from, it's wrong.
>
> Then I'll challenge the MVP person who said that to me.

Good on yer.

--
Duncan

Dave Doe wrote:
> In article <>,
> says...
>> On Fri, 22 Aug 2008 22:18:15 +1200, Enkidu wrote:
>>
>>> Exchange 5.5 has a fault, which in *extreme* cases required an offline
>>> defrag. Exchange 2000 had a limit to the size of the Exchange database
>>> which, if it was allowed to fill up required an offline defrag.
>>>
>>> Scarcely 'basic faults' and not requiring 'semi-regular offline defrags.
>> Those are fundamentally basic database faults, IMO. There is no good
>> reason why an email server should need to go offline merely to facilitate
>> the reduction in disc size of a mail store.
>
> It doesn't, nor does it need to.
>
> If circular logging is *not* enabled, then you need to perform an
> exchange aware backup (otherwise you don't!) - to flush the log files,
> you'll eventually fill yer hard disk otherwise.
>
> Short of filling the disk though, Exchange will run just fine. Dunno
> where you get your info from, it's wrong.
>
He's possibly talking about pre-Ex2000, such as 5.5 and 5.0. There was a
storage limit in 5.0 and 5.5 of (I think) 2GB. Storage *did* tend not to
be released and the cure for reaching the limit (which crashed the
system) was to do exactly what he said - an offline defrag. Now I come
to think of it, the problem probably went away with 5.5 and 5.0. The
Enterprise Ex2000 went, I think, to 16GB but the Standard was still only
2GB. From memory an offline defrag was possible with Ex2000.

Thanks for reminding me of Exchange aware backups. A restart of the
system with an orderly shutdown also flushed the log files, from memory.

Please feel free to correct any errors above - I'm working purely from
memory!

Cheers,

Cliff

--

"I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.

Carnations wrote:
> On Fri, 22 Aug 2008 23:26:27 +1200, Dave Doe wrote:
>
>> Short of filling the disk though, Exchange will run just fine. Dunno
>> where you get your info from, it's wrong.
>
> Then I'll challenge the MVP person who said that to me.
>
MVP or MSP?

Cheers,

Cliff

--

"I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.

Carnations wrote:
> On Sat, 23 Aug 2008 14:29:36 +1200, Enkidu wrote:
>
>> Carnations wrote:
>>> On Fri, 22 Aug 2008 23:26:27 +1200, Dave Doe wrote:
>>>
>>>> Short of filling the disk though, Exchange will run just fine.
>>>> Dunno where you get your info from, it's wrong.
>>> Then I'll challenge the MVP person who said that to me.
>>>
>> MVP or MSP?
>
> Actually, probably MSP - because to become a MVP you need to spend a
> shitload of time promoting/defending/teaching Micro$oft products
> within community forums, such as Usenet - like what Nathan and Bret
> were doing here.
>
So he picked up from a book. Probably an exchange 5.5 book or some
*really old* usenet postings.

As opposed to actually using it over many years and helping people out
on public forums?

Cheers,

Cliff

--

"I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.

Carnations wrote:
> On Sun, 24 Aug 2008 17:25:40 +1200, Enkidu wrote:
>
>> As opposed to actually using it over many years and helping people
>> out on public forums?
>
> Actually, it is a part of his job to look after that particular
> exchange server.
>
Get someone else.

Cheers,

Cliff

--

"I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.

Carnations wrote:
> On Mon, 25 Aug 2008 23:18:03 +1200, Enkidu wrote:
>
>> Carnations wrote:
>>> On Sun, 24 Aug 2008 17:25:40 +1200, Enkidu wrote:
>>>
>>>> As opposed to actually using it over many years and helping people out
>>>> on public forums?
>>> Actually, it is a part of his job to look after that particular
>>> exchange server.
>>>
>> Get someone else.
>
> Why? He's only looking after Micro$oft shite?
>
> Its not like it's real software on a real OS.
>
That "Micro$oft shite" contains a hell of a lot of the IP of the
company. If he stuffs up he could cost the organisation *millions*,
depending on how big the organisation is.

Cheers,

Cliff

--

"I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.

Carnations wrote:
> On Tue, 26 Aug 2008 22:44:52 +1200, Enkidu wrote:
>
>> That "Micro$oft shite" contains a hell of a lot of the IP of the
>> company.
>
> Really?
>
> I thought that it was Micro$oft's IP that was being used.
>
Intellectual Property.

Cheers,

Cliff

--

"I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.