In this collection of frequently asked questions, our SearchExchange.com experts tackle several common Exchange Server information store issues. From fixing configuration and performance problems to copying and recovering information store data, you're sure to learn something new about managing the Exchange Server information store from this assortment of expert advice.
|Frequently Asked Questions:
EXCHANGE SERVER INFORMATION STORE
- Fixing a hanging information store
- Can I create multiple information stores?
- Recovering a store after a 'dirty shutdown'
- Restoring data to another Exchange server information store
- STORE.EXE is degrading performance
- Store terminates unexpectedly after attaching documents via OWA
- Running ESEUTIL to compact an IS without installing Exchange Server
- What happens if my active cluster's information store goes down?
- Copying an existing Exchange configuration to a new server
- System attendant service problems
If the information store is hung, and the server is not clustered, what is the best way to resolve the issue so that the client can get e-mail?
Being reactive, you obviously need to get the service back up and running. If a simple restart of the Information Store Service, (don't forget any dependent services), will get your users able to connect, then that should suffice until you get some time to perform server maintenance and investigate further.
I urge you to be proactive and figure out why the service is hanging. This should not be happening regularly. Troll the application event logs and note any errors (or warnings) from the Microsoft Exchange information store approaching the time the service hangs. Then either write back to us or search TechNet in order to get some clues on what's going wrong. Then you can resolve the root cause.
I am running Exchange 5.5 Standard Edition and I am currently at my maximum size of 16 GB for the priv.edb. Is it possible to create multiple information stores to bypass the 16 GB limit set on the Standard Edition?
No, unfortunately you can't do this as Exchange 5.5 only supports a single private store. Even Exchange 2000 and Exchange 2003 Standard Editions only support a single private store. Upgrade to Enterprise Edition.
You should know that Exchange 2003 SP2 will increase the size limit of your stores from 16 GB to 75 GB.
An Exchange Server store was dismounted when a client rebooted the server manually. Now the store will not remount. I get an error that the tmp.mdb is not found. However, a temp.mdb does exist and tmp.mdb cannot be created -- otherwise the store appears intact. What could cause this? Is this error repairable?
Tmp.mdb is a temporary file that is deleted when all the stores are dismounted or the information store service is stopped -- in other words, a clean shutdown. What you experienced was a "dirty shutdown" that most likely left the store(s) in an inconsistent state.
To fix this, you will need to perform a recovery on the store using ESEUTIL with the /r option. This should bring it back to a consistent state.
Any time you run ESEUTIL on a store, it's a good idea to copy the store and transaction logs to an alternate location or take a backup of the file system. Then navigate to the storage group that the affected store belongs to, and from Properties -> General tab, note the "Log file prefix" -- E00 by default for the first storage group.
To start the recovery process, go to a command prompt and type: EXCHSRVRBINESEUTIL.EXE /r /l
I have a query about how to restore/recover a Microsoft Exchange 2000 server public and private folder to another Exchange server information store. My server SCSI drive is damaged and I can't re-install or repair. I used ESEUTIL.EXE to check and repair the said databases, but it's not working. Is there any third-party tool to recover all old mails and addresses to the new Exchange server?
Sounds like you're in a pickle. Well the standard response would be to restore from backup, but I'm assuming from your question that you weren't making backups. If that's a correct assumption, then first of all, put in that requisition to acquire some good Exchange-aware backup software. Then, have a look at an Exchange recovery solution.
Since your ESEUTIL exercise failed, there's a chance that these tools may not be able to read your database either. If that's the case, and if you really need the mail, then you're left with having to take the database into a forensic recovery company to see if they can get the data out. That's typically pretty pricey, but it may be your only option.
The process STORE.EXE on my Exchange 2003 server is degrading the performance of the machine. I'm running GroupShield with SpamKiller, but I still can see a bunch of e-mails being scanned by the engine. We don't have that kind of traffic in our enterprise, nevertheless the traffic is horrible. I've tried everything! Any help will be really appreciated.
Ah … the infamous STORE.EXE process. While this process may look malicious, it is actually an Exchange service. The STORE.EXE process is the service that manages all of the information stores (databases) on an Exchange server.
It is not uncommon to see the STORE.EXE process consuming a large amount of memory; on busy machines, it will consume much of your processor resources as well. There are ways to optimize system resources for the STORE.EXE process.
If you are running an Exchange mailbox server that has 1 GB or more physical memory (RAM), you should enable /3GB switch in the boot.ini file. This is supported on all version of Windows Server 2003.
While your end users are not generating the high volume of traffic that you are witnessing on your server, spammers are. It is estimated that up to 80% of all mail received by an organization is spam. Therefore, it is likely that your antivirus/antispam server will be busy handling this. In your case, that is Exchange Server.
The reality is that GroupShield and SpamKiller are actually contributing to your STORE.EXE problem every time they run, by accessing mailboxes and scanning mail items, because they both utilize the STORE.EXE process to do so.
I think you would be a good candidate for either a network appliance or hosted service that can perform your antivirus/spam filtering for you, so that your Exchange server doesn't have to.
I have an Exchange 2000 server (SP3 with post rollup and patches), with Outlook Web Access (OWA) on Windows 2000 SP4. When a user tries to attach a document via OWA, the store terminates unexpectedly. This just started last week for no apparent reason. What am I missing?
Have you installed any software recently that may be conflicting with Exchange Server? Antivirus program updates can cause the server to become unstable. Check the event logs for application and system errors and research the Event ID numbers. Monitor your server's memory and processes running to eliminate hardware issues.
I read the SearchExchange.com article Is an offline defragmentation really worth it? The article refers to using ESEUTIL /D to compact an information store and reclaim disk space. However, I don't have enough room on my server to perform the defragmentation. I can't seem to find the right syntax to push the temporary files to an alternate file share during defragmentation. Is there any additional risk to the information store?
ESEUTIL can be run on a computer without installing Microsoft Exchange. You will need to copy the database files to another computer to do this. In addition, you will need to copy the files Eseutil.exe, Ese.dll, Jcb.dll, Exosal.dll, and Exchmem.dll from the Exchsrvrbin directory to the path from which you want to run it. You will also need to make sure the language locale of the computer you are running ESEUTIL on matches the one on your Exchange server.
You can then copy the defragmented database back to its original location. Please note -- you will need to unmount the database you are defragmenting to be able to copy it to another computer and then copy it back to the original location on your Exchange server.
Alternatively, you could specify a different name and path using ESEUTIL's /T switch. If you mapped a shared drive to run ESEUTIL, and the drive letter you assigned to the mapped drive is Z:, you'd use the following command:
Eseutil /d /ispriv /tz:NewDatabase.edb
Microsoft recommends you have free space equal to (at least) 110% of the size of the database being defragmented.
I am setting up a clustered Exchange 2003 environment. I will have two active servers and two passive. All three will be connected to a SAN. If an information stores goes down on an active server, or there's another problem, will the passive server have the same problem? The active and passive are set to update each other all the time, and if the active server doesn't respond with the passive after one minute, the passive server then becomes active.
If you lose an active node for whatever reason (i.e., services hung, other failure) then the cluster will fail over to the passive node. The passive node should not have the same issues. Since the databases are on a shared location on the SAN, if you lose a database due to corruption, you will need to restore from backup.
I am going to move an existing Exchange 5.5 SP4 server from its current server to a new server. Both servers are Windows 2000 with the latest service packs and hotfixes. The new server is in an isolated network, so I can have the exact same name and configuration as the current one.
Is it possible to use NTBackup in Windows 2000 to back up the directory store and information store, and then restore it to the new server? I have done the backup and restore, but when I use the Exchange administrator tool to open the address book, it doesn't show any mailboxes. It looks as if I haven't done any restore.
You will want to make sure you run ISINTEG -Patch in order to restamp the information store with the proper GUID prefix. It's not just the servername that's important. While the servers may share a name, their GUIDs are different.
We have an Exchange 2000 Server. The system attendant service is automatically stopped by itself. Due to this, the information store service is also stopping by itself. I am manually re-starting the services and after some time it is again stopping itself. What is causing this?
I have seen quite a number of cases where a system attendant service that is stopping silently like yours is related to not keeping an Exchange 2000 Server current with service packs and hotfixes.
Do you have the latest service packs and hotfixes installed on your Exchange 2000 servers? You should have the Exchange 2000 Server post-service pack 3 rollup fix installed, as a minimum. This fix is described in Microsoft Knowledge Base Article 813840 "XGEN: March 2003 Exchange 2000 Server Post-Service Pack 3 Rollup" and the fix can be downloaded on Microsoft's site.. Make sure you are running the March 24, 2003 version of this hotfix and not the earlier version (the hotfix was re-released).
Applying this rollup hotfix should resolve your problem. If it doesn't, let us know and we'll follow-up some more.