Home All Groups Group Topic Archive Search About

Method followed for assigning users to a mail store?

Author
1 Jul 2009 1:37 AM
Barkley Bees
Up until now we have only been utilizing one Mail store on our Exchange 2003
server. We have 2 storage groups configured with 2 mail stores in each. We
are planning to increase the mailbox limitations for our users and as such
we will need to spread them across the existing resources to support this.

I would like to know, given all mailbox stores have the same limits
assigned, how do you generally decide what users get assigned to what
store/group? What practices/method do you follow - random, alphabetical, by
department, etc? Is there a 'best practice' for this?

Appreciate any feedback. Thanks.

Author
1 Jul 2009 2:54 AM
Rich Matheisen [MVP]
On Wed, 1 Jul 2009 10:37:56 +0900, "Barkley Bees"
<barkb***@nomail.com> wrote:

>Up until now we have only been utilizing one Mail store on our Exchange 2003
>server. We have 2 storage groups configured with 2 mail stores in each. We
>are planning to increase the mailbox limitations for our users and as such
>we will need to spread them across the existing resources to support this.
>
>I would like to know, given all mailbox stores have the same limits
>assigned, how do you generally decide what users get assigned to what
>store/group? What practices/method do you follow - random, alphabetical, by
>department, etc? Is there a 'best practice' for this?
>
>Appreciate any feedback. Thanks.

We use a pseudo-random dispersion of new mailboxes over all servers in
an AD site (minus a small number of excluded databases). That's
scripted, not done by hand. For one-off new mailboxes we just left the
person creating it pick one they like. :-)

Works well no matter what the number of servers or databases.
---
Rich Matheisen
MCSE+I, Exchange MVP
Author
3 Jul 2009 5:26 AM
Ed Crowley [MVP]
My much smaller customer chose to more or less evenly distribute mailboxes
across their five databases based on the first letter of their name.
--
Ed Crowley MVP
"There are seldom good technological solutions to behavioral problems."
..

Show quoteHide quote
"Rich Matheisen [MVP]" <richnews@rmcons.com.NOSPAM.COM> wrote in message
news:rojl455cbpfhcilkinjoj3grqnu73u81ss@4ax.com...
> On Wed, 1 Jul 2009 10:37:56 +0900, "Barkley Bees"
> <barkb***@nomail.com> wrote:
>
>>Up until now we have only been utilizing one Mail store on our Exchange
>>2003
>>server. We have 2 storage groups configured with 2 mail stores in each. We
>>are planning to increase the mailbox limitations for our users and as such
>>we will need to spread them across the existing resources to support this.
>>
>>I would like to know, given all mailbox stores have the same limits
>>assigned, how do you generally decide what users get assigned to what
>>store/group? What practices/method do you follow - random, alphabetical,
>>by
>>department, etc? Is there a 'best practice' for this?
>>
>>Appreciate any feedback. Thanks.
>
> We use a pseudo-random dispersion of new mailboxes over all servers in
> an AD site (minus a small number of excluded databases). That's
> scripted, not done by hand. For one-off new mailboxes we just left the
> person creating it pick one they like. :-)
>
> Works well no matter what the number of servers or databases.
> ---
> Rich Matheisen
> MCSE+I, Exchange MVP
Author
4 Jul 2009 2:12 AM
jamestechman
Everybody has different approaches no right\wrong way really. I'm
pretty sure there's benefits and disadvantages in each method. I
personally do by quota per store. For 2000/2003 it's a bit
encumbersion to automate mailbox dispersion, 2007 makes it easier with
powershell.


James Chong (MVP)
MCITP | EA | EMA; MCSE | M+, S+
Security+, Project+, ITIL
msexchangetips.blogspot.com

Show quoteHide quote
On Jun 30, 9:37 pm, "Barkley Bees" <barkb***@nomail.com> wrote:
> Up until now we have only been utilizing one Mail store on our Exchange 2003
> server. We have 2 storage groups configured with 2 mail stores in each. We
> are planning to increase the mailbox limitations for our users and as such
> we will need to spread them across the existing resources to support this..
>
> I would like to know, given all mailbox stores have the same limits
> assigned, how do you generally decide what users get assigned to what
> store/group? What practices/method do you follow - random, alphabetical, by
> department, etc? Is there a 'best practice' for this?
>
> Appreciate any feedback. Thanks.
Author
4 Jul 2009 2:38 AM
Ed Crowley [MVP]
I had a customer who assigned mailboxes to different stores based on quota
and named the stores accordingly, e.g., 100MB-1, 100MB-2, 500MB-1, and so
on.  Exchange 2007 diminishes the effectiveness of this method just a bid
because there is no longer the notion of mailbox store policy.  The big
drawback I see to this approach is that to maintain the system, to increase
a user's quota you'd have to move his mailbox.
--
Ed Crowley MVP
"There are seldom good technological solutions to behavioral problems."
..

"jamestechman" <jamestech***@gmail.com> wrote in message
news:8ff8b091-97a8-4ca6-84af-eb8cafdb3cdf@r33g2000yqn.googlegroups.com...
Everybody has different approaches no right\wrong way really. I'm
pretty sure there's benefits and disadvantages in each method. I
personally do by quota per store. For 2000/2003 it's a bit
encumbersion to automate mailbox dispersion, 2007 makes it easier with
powershell.


James Chong (MVP)
MCITP | EA | EMA; MCSE | M+, S+
Security+, Project+, ITIL
msexchangetips.blogspot.com

Show quoteHide quote
On Jun 30, 9:37 pm, "Barkley Bees" <barkb***@nomail.com> wrote:
> Up until now we have only been utilizing one Mail store on our Exchange
> 2003
> server. We have 2 storage groups configured with 2 mail stores in each. We
> are planning to increase the mailbox limitations for our users and as such
> we will need to spread them across the existing resources to support this.
>
> I would like to know, given all mailbox stores have the same limits
> assigned, how do you generally decide what users get assigned to what
> store/group? What practices/method do you follow - random, alphabetical,
> by
> department, etc? Is there a 'best practice' for this?
>
> Appreciate any feedback. Thanks.