• RSS
  • Twitter
  • FaceBook

Exchange Server Forums

Forums | Register | Login | My Profile | Inbox | RSS RSS icon | My Subscription | My Forums | Address Book | Member List | Search | FAQ | Ticket List | Log Out

Information store dropping

Users viewing this topic: none

Logged in as: Guest
  Printable Version
All Forums >> [Microsoft Exchange 2003] >> Information Stores >> Information store dropping Page: [1]
Login
Message << Older Topic   Newer Topic >>
Information store dropping - 2.Feb.2009 12:07:08 PM   
gerryr

 

Posts: 14
Joined: 31.Jan.2008
Status: offline
Hi

We have some of our information stores stored on a NAS box, the exchange server talks to this nas box over a dedicated 1GB connection.

Lately the info stores on this NAS box are becoming unmounted at various times.  They are fairly large (200GB+) but we have an identical setup in our other office with even larger stores on the NAS box and are not getting this disconnecting problem.

The drop seems to correspond with the online database defrag, ie the online defrag starts at 1am and the DB drops connection between 1 - 1.30.

Prior to the disconnect we get a few event warnings (ID 510, source ESE) indicating degraded performance and they are shortly followed by event errors (ID 482, source ESE)

Just wondering if anyone's come accross this before or if anyone has any ideas how to solve this?  (we don't have enough space direct attached to the Exch server to store the stores on it)

Thanks
Gerry
Post #: 1
RE: Information store dropping - 2.Feb.2009 1:19:14 PM   
jveldh

 

Posts: 2335
Joined: 12.Apr.2008
From: The Netherlands
Status: offline
Hi,

Please run Exbpa to check for issues, also monitor your system with Perfmon and keep an eye on the disk queue length. If this get's very high it indicates an issue with the underlying disks. If you have your disks on a SAN then maybe adding disks/units will fix the issue.



_____________________________

Best regards,

Johan Veldhuis

Visit my Exchange blog

(in reply to gerryr)
Post #: 2
RE: Information store dropping - 3.Feb.2009 1:14:47 AM   
gladone

 

Posts: 90
Joined: 2.Feb.2009
Status: offline
Can you paste the entire description of the following events ID 510, source ESE ID 482, source ESE)

(in reply to jveldh)
Post #: 3
RE: Information store dropping - 3.Feb.2009 5:26:42 AM   
gerryr

 

Posts: 14
Joined: 31.Jan.2008
Status: offline
Hi All
Thanks for the replies, I ran the bpa tool and it turned up a few warnings about not having the 3GB switch on and drivers being different than the origional.  Don't think they're related to the drop though as the same warnings are on the other exchange box.
I've been running the perf logs for the past few days and couldn't notice much changing at the time the DB dropped off.  On one disk the max queue length reaches 24 at one stage but is usually below 5.  One thing though, the drive we store the DBs that are dropping on don't show up on the perf logs even though I have all instances selected.  We used a HP feature pack for exchange to connect the NAS drive to exchange, I beleive it uses some form of iSCSI so maybe that's why the drive doesn't show up?  Any way to get it to show up in the perf logs?
Apologies for the long post!
Thanks
gerryR

Below are  the events I mentioned:
Event Type: Warning
Event Source: ESE
Event Category: Performance
Event ID: 510
Date:  03/02/2009
Time:  01:14:39
User:  N/A
Computer: W2KSLMK4
Description:
Information Store (4040) Third Storage Group: A request to write to the file "S:\{w2kslmk5.0}\Cork2.edb" at offset 25806405632 (0x00000006022e8000) for 4096 (0x00001000) bytes succeeded, but took an abnormally long time (73 seconds) to be serviced by the OS. In addition, 6 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 66 seconds ago. This problem is likely due to faulty hardware. Please contact your hardware vendor for further assistance diagnosing the problem.
For more information, click http://www.microsoft.com/contentredirect.asp.

Event Type: Error
Event Source: ESE
Event Category: General
Event ID: 481
Date:  03/02/2009
Time:  01:16:22
User:  N/A
Computer: W2KSLMK4
Description:
Information Store (4040) Third Storage Group: An attempt to read from the file "S:\{w2kslmk5.2}\Moscow2.edb" at offset 32199499776 (0x000000077f3d6000) for 4096 (0x00001000) bytes failed after 0 seconds with system error 64 (0x00000040): "The specified network name is no longer available. ".  The read operation will fail with error -1022 (0xfffffc02).  If this error persists then the file may be damaged and may need to be restored from a previous backup.
For more information, click http://www.microsoft.com/contentredirect.asp.

(in reply to gladone)
Post #: 4
RE: Information store dropping - 3.Feb.2009 5:50:53 AM   
gladone

 

Posts: 90
Joined: 2.Feb.2009
Status: offline
It looks like database corruption .. I would go with description in event ID. Take a full back of the database. restore it to the RSG. Delete and recreate the mailbox store in production. Merge the mails.

(in reply to gerryr)
Post #: 5
RE: Information store dropping - 3.Feb.2009 6:44:20 AM   
gerryr

 

Posts: 14
Joined: 31.Jan.2008
Status: offline
Hi

I don't think it's a DB corruption problem (yet!) as we have 3 or 4 DBs running on this NAS box an when one drops they all drop.

Thanks
Gerry

(in reply to gladone)
Post #: 6
RE: Information store dropping - 3.Feb.2009 12:57:46 PM   
jveldh

 

Posts: 2335
Joined: 12.Apr.2008
From: The Netherlands
Status: offline
Hi,

Which NAS solution is attached to your system and does it have the option to monitor the system ?



_____________________________

Best regards,

Johan Veldhuis

Visit my Exchange blog

(in reply to gerryr)
Post #: 7
RE: Information store dropping - 3.Feb.2009 2:56:52 PM   
a.grogan

 

Posts: 1917
Joined: 12.Apr.2005
From: London
Status: offline
Hiya chap,

Is the NAS solution being used for anything else such as File Share / SQL / Oracle or dedicated purely to Exchange. Have you had a look at perfmon and monitored disk I/O across the volumes allocated to Exchange. It looks like either performance on the SAN has degraded to the point where the LUNS phase in and out - or you might have a problem with your Fibre Connection - speaking of which are there any "kinks" in the Fibre cables?

Cheers

A

_____________________________

Andy Grogan
MSExchange.org Forums Moderator
For my general ramblings about Exchange please visit my website:
W: http://www.telnetport25.com/
B: http://telnetport25.wordpress.com/
M: manifoldmaster@gmail.com

(in reply to gerryr)
Post #: 8
RE: Information store dropping - 4.Feb.2009 2:35:57 PM   
v9brcha

 

Posts: 38
Joined: 6.Jan.2009
Status: offline
Hi,

I would like to know if you have any other jobs scheduled during the online defrag for example a backup.
Try ruling out the basics so that we can exactly find out the issue.

Thanks

(in reply to gerryr)
Post #: 9
RE: Information store dropping - 4.Feb.2009 9:15:32 PM   
MTK

 

Posts: 84
Joined: 6.Oct.2005
Status: offline
Hi Guys,

I am having the same problem. The mailbox store (servername) is unmounted every another day morning. At first I thought that this is because of MDBDATA folder size and sometime I don't backup regularly. Now I backup regularly and already configure for Circular Logging in order to reduce the folder size. But the problem is still happening. I saw Event ID : 116 and 9175 when I check the event viewer.

Any idea?

MTK

(in reply to gerryr)
Post #: 10

Page:   [1] << Older Topic    Newer Topic >>
All Forums >> [Microsoft Exchange 2003] >> Information Stores >> Information store dropping Page: [1]
Jump to:

New Messages No New Messages
Hot Topic w/ New Messages Hot Topic w/o New Messages
Locked w/ New Messages Locked w/o New Messages
 Post New Thread
 Reply to Message
 Post New Poll
 Submit Vote
 Delete My Own Post
 Delete My Own Thread
 Rate Posts


Follow TechGenix on Twitter