Go Premium for a chance to win a PS4. Enter to Win

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 682
  • Last Modified:

Severe Exchange 2003 performance degradation

Hello - we are experiencing performance problems with our Exchange Outlook users, where it takes a very long time to retrieve data from Exchange.  We hace a single Exchange 2003 Server with about 75 users on it.  The evbent viewer application log shows many 510 errors like this one:
Source=ESE, Event=510:
Information Store (6324) First Storage Group: A request to write to the file "C:\Exchsrvr\mdbdata\priv1.edb" at offset 7658602496 (0x00000001c87d0000) for 8192 (0x00002000) bytes succeeded, but took an abnormally long time (138 seconds) to be serviced by the OS. In addition, 113 other I/O requests to this file have also taken an abnormally long time to be serviced since the last message regarding this problem was posted 50 seconds ago. This problem is likely due to faulty hardware.

We are planning to bring the server down to do some hardware diagnostics, but are wondering if anyone else has experienced this type of issue and found the cause to be something other than the hardware?

Thanks
0
Damian_Gardner
Asked:
Damian_Gardner
  • 4
  • 3
  • 2
2 Solutions
 
xmlmagicianCommented:
once due to virus and malware
do you have any faults pop up on for the hard drives?
can you let a diagnostic run over night for integrity etc?
0
 
lancecurwensvilleCommented:
In addition to potential HD failures like xmlmagician, take a look at the raid controller, it may be causing problems striping the data.  

Another area I would look at is it appears that you have your exchange db on the same partition as your OS.  Best practices for performance is that your db lies on a different partition (http://www.petri.co.il/move_exchange_stores_to_a_different_disk.htm).  This link also describes how to move this database.  You may not have any physical problems, you've just hit that "magical" mark where the db stores have hit a size that performance has been degraded to the point where it is causing problems.
0
 
lancecurwensvilleCommented:
oops, that should have state like xmlmagician stated......
0
Get your Disaster Recovery as a Service basics

Disaster Recovery as a Service is one go-to solution that revolutionizes DR planning. Implementing DRaaS could be an efficient process, easily accessible to non-DR experts. Learn about monitoring, testing, executing failovers and failbacks to ensure a "healthy" DR environment.

 
Damian_GardnerAuthor Commented:
ok - thanks for the suggestions, gentlemen.  I'll check the hardware after hours today, and then also consider the best practices.

Thanks,
Damian
0
 
Damian_GardnerAuthor Commented:
Have checked all the hardware and the RAID controller, and everything is clean.  I've turned on diagnostics for the IS and am monitoring for errors this week.  After this stage, I'll have to figure out how best to split up the EDB from the log file, when only using a 2 disk mirrored array.  I'm wondering - at the very least, would it help to split 1 disk into 2 partitions, to split up the database and log files?
0
 
lancecurwensvilleCommented:
Having two partitions on that array would be a much better situation, but...doing so after the fact is a **little** scary as if it fails in the process you've really created a nightmare.  I didn't realize from your post that you had a single partition on the existing array.  I've done this successfully in the past, and I mean no offense to the open-source crowd but I wouldn't trust any that are out there on a production exchange server.  I've used Acronis Disk Director Server edition successfully creating and resizing partitions on Server 2003 after it is in production.  http://www.acronis.com/enterprise/products/diskdirector-server/index.html  It is a little expensive, but...when you gauge your time in fixing a failed partition using open source and the lost $$$ in downtime for the company; Acronis's price balances out.  Additionally, you would have access to their support if you have questions.
0
 
Damian_GardnerAuthor Commented:
Point taken.  Thanks for the suggestion and standby for more updates.

Damian
0
 
xmlmagicianCommented:
Looking forward to them
0
 
Damian_GardnerAuthor Commented:
Sorry for the delay.  The performance trouble seems to have cleared up without changing anything, surprisingly.  I've turned on logging and am still monitoring, but so far it's been ok.  Thanks for your help and I'll try and split the points.

Damian
0

Featured Post

VIDEO: THE CONCERTO CLOUD FOR HEALTHCARE

Modern healthcare requires a modern cloud. View this brief video to understand how the Concerto Cloud for Healthcare can help your organization.

  • 4
  • 3
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now