|
|
|
date: Fri, 8 Feb 2008 13:22:32 -0500,
group: microsoft.public.exchange.tools
back
Re: ESEUtils
A better question would be why? What goal are you trying to accomplish?
Shrink the physical size of the store?
If you have another exchange server in the organization or are running
enterprise, use movemailbox to a new store instead.
If you are running standard or SBS and this is the only exchange server,
then to realize space recovery you:
1. Must ensure that the store will not simply grow again. Exchange will
use whitespace in the database before increasing the size of the database.
If the store will simply grow again, reducing the physical size of the store
is a pointless exercise in counterproductivity. To ensure that the store
will not simply grow again, you need to enforce limits. Limits aren't an
arbitrary thing; work with you business units to define business objectives
and determine how much mail you need to retain to meet those business
objectives. If the mail is worth keeping, then keep it on exchange. If it
has no business value, then having it around is a liability; get rid of it.
2. After you have removed a large amount of mail (so that there is enough
whitespace that will never be reused to make the user disruption
worthwhile), you have to wait for deleted items retention to expire, then
wait for the next online maintenance cycle to complete, then wait for online
defrag to complete. At this point, you need to check the event ID 1221 to
make sure you're going to gain something before taking an outage.
3. Conservatively estimate the outage time and notify users ahead of time.
(I think this is the step you were trying to do in your original post)
4. Make sure you have a full backup and a backout plan prior to the
beginning of you announced outage window.
5. Practice the steps on a test server (in VM works, you want to validate
the steps and resolve any issues with the steps before the outage window
begins.
6. At the beginning of your outage window, execute your tested steps in
production.
"MC" wrote in message
news:OhtcT%23naIHA.484@TK2MSFTNGP06.phx.gbl...
>I want to defrag exchange 2003(Sp2) database with ESEUtils.exe,
> does anyone know how long it may take to defrag and compact abou 50Gig
> data size?
> I have 2 x Quad core Cpu, 4GRam, and enough storage on server.
date: Sat, 9 Feb 2008 12:25:05 -0800
author: John Fullbright fjohn@donotspamnetappdotcom
Re: ESEUtils
Intent was to reduce my exchange enterprise databases
event id 1221 shows that, my 3 databases has about 3G free space after
online defrag ended.
as you suggest, I guess no need to defrag
Thanks
MC
"John Fullbright" <fjohn@donotspamnetappdotcom> wrote in message
news:uNvEdn1aIHA.5164@TK2MSFTNGP03.phx.gbl...
>A better question would be why? What goal are you trying to accomplish?
>Shrink the physical size of the store?
>
> If you have another exchange server in the organization or are running
> enterprise, use movemailbox to a new store instead.
>
> If you are running standard or SBS and this is the only exchange server,
> then to realize space recovery you:
>
> 1. Must ensure that the store will not simply grow again. Exchange will
> use whitespace in the database before increasing the size of the database.
> If the store will simply grow again, reducing the physical size of the
> store is a pointless exercise in counterproductivity. To ensure that the
> store will not simply grow again, you need to enforce limits. Limits
> aren't an arbitrary thing; work with you business units to define business
> objectives and determine how much mail you need to retain to meet those
> business objectives. If the mail is worth keeping, then keep it on
> exchange. If it has no business value, then having it around is a
> liability; get rid of it.
>
> 2. After you have removed a large amount of mail (so that there is enough
> whitespace that will never be reused to make the user disruption
> worthwhile), you have to wait for deleted items retention to expire, then
> wait for the next online maintenance cycle to complete, then wait for
> online defrag to complete. At this point, you need to check the event ID
> 1221 to make sure you're going to gain something before taking an outage.
>
> 3. Conservatively estimate the outage time and notify users ahead of
> time. (I think this is the step you were trying to do in your original
> post)
>
> 4. Make sure you have a full backup and a backout plan prior to the
> beginning of you announced outage window.
>
> 5. Practice the steps on a test server (in VM works, you want to validate
> the steps and resolve any issues with the steps before the outage window
> begins.
>
> 6. At the beginning of your outage window, execute your tested steps in
> production.
>
>
>
>
>
>
> "MC" wrote in message
> news:OhtcT%23naIHA.484@TK2MSFTNGP06.phx.gbl...
>>I want to defrag exchange 2003(Sp2) database with ESEUtils.exe,
>> does anyone know how long it may take to defrag and compact abou 50Gig
>> data size?
>> I have 2 x Quad core Cpu, 4GRam, and enough storage on server.
>
>
date: Thu, 14 Feb 2008 10:25:36 -0500
author: MC
|
|
