Exchange 2003 to 2010 Coexistence (Full Version)

All Forums >> [Microsoft Exchange 2003] >> Migration


ldudmish -> Exchange 2003 to 2010 Coexistence (5.Sep.2012 4:23:08 PM)

Hi All,

We are in the process of setting up and configuring exchange 2010 for our company. We currently have exchange 2003 however as we run a mixed Mac and PC environment there is now a requirement for us to move to exchange 2010 as the new mac office 2011 is not compatible with exchange 2003 and currently we are having to use davmail as an inbetween which uses rpc over the owa facility but does not give us a true sync with the mailboxes.

I am picking up on this task and I am after any advice anyone could give on what to watch out for during the coexistence ideally initially we are looking to migrate just the mac users at first to the new box with others over a period of time.

We currently have a bit of an unfinished installation on exchange 2003 in such that at some point a new server was added and all active mailboxes moved over this was made the master and has been acting as such for around 2 years however the previous server was never decomissioned fully and has remaind online and is acting as a member server.

I have checked the smtp logs and it is still doing some relaying along with hosting older mailboxes on the device. I have started moving the mailboxes off either by removal of the account as no longer used or moving to the current exchange 2003 master server I hope to have this completed by the end of the week at which point no mailboxes will reside on the server.

After this would be to ensure all public folders have been moved over and removal of the relaying checking all systems still work ok before decomissioning the server.

I ideally would like to do this so we only have the 1 server in which to coexist with exchange 2010 rather than the two, and would make things a little easier.

Our main server also is suffering with high load issues on the hard drives mainly we see alot of high disk queue length and page file access it seems to spike alot during the creation of IFS files on the database drive in particular the second storage group these are small files however when they are created the drives go nuts then quieten down.

I am not sure if this is something which is mac related as alot of the macs have also been using mac mail over imap however we disabled the imap service a couple of weeks back and things were ok but the last few days disk activity has been high again even with this turned off.

Setup of the server is roughly 200-300 users not all on at the same time as we are international so max would be 175 during the day, it resides on a physical box a dell r710 with 4gb ram hard drives partioned into C:\ system 135gb (60gb Free) E:\ databases 1tb (500gb free) G:\ logs 300gb (265gb free) we have 6 drives in total each array is a raid 1 mirror however having trouble installing the dell openmanage utility to check the actual configuration for sure but looking at the disks on the box tells me they are a raid 1 by the activity.

We have three storage groups and a total of 9 databases (3 in each) between 20gb and 108gb each in size, along with a very small public folder which is not used at all really.

Users are experiencing slow use and drop outs on owa which I am guessing is failing to service the request as the disks are too busy, it does not seem to affect outlook users who have yet to complain about any problems. Alot of the users on the system are using owa as a mail client instead of outlook or other client as they have no real need other than to send/recieve email and is an historic decision taken to save some much needed money to pump into infrastructure.

This issue however is forcing our hand to move to exchange 2010 as the high load together with improved performance in 2010 will help us in balancing the load and also improved compatibility with our macs.

Rough guide is 50 - 60 people on outlook pc, 20 people on outlook mac 2011 and 10 on entourage 2008 the rest use owa.

To further add things into the equation we also run blackberry server with about 30 - 40 blackberry's on which is also suffering with delayed email and failures due to the high disk activity my guess again the server unable to answer the requests normally a reboot sets this all straight of both boxes but the business operates 24/7 and has no room for downtime something which we are keen to fix is the high disk activity as this will help buy us time to do the coexistence properly rather than rush and slap it all together.

Any advice or help is much appreciated



Page: [1]