Storage Groups With LCR Disappeared?

[LibertyGuy]

Hello,

We implemented exchange 2007 several months ago at our company. Originally, we had 4 storage groups set up, all of which had LCR enabled that were copying to a secondary hard drive.

This morning, I had an error saying that our E drive was full (the drive where the LCR copies to). I realized that I did not have circular logging enabled, and so we had a ridiculous number of logs, which had obviously over time taken a large amount of space on the disk.

I went through and cleared out a lot of the old logs to free up some space, and to allow LCR to continue. I then checked the status of the storage groups under Database Management, and saw that all of their status was set to "Failed". Figuring now that I had the space, I decided to click 'refresh' and expected to see everything revert back to 'Healthy' status. However, all of my storage groups under Database Management simply VANISHED!!??

What has happened here?

Now, when I attempt to create a new storage group, it gives me the error: "The parent object for <Storage Group Name> could not be found. Please check that InformationStore exists."

I am unable to create these storage groups. I do not have a "Recovery Storage Group" simply because it is only just today that I have even heard about it. I am an entry-level IT Assistant, just getting my foot in the door. I built this domain from the ground up, and this Exchange 2007 has been anything but a joy to work with. I do not understand how or what has caused this, but a solution must be found, because I simply cannot allow backups/replication to be nonexistant.

Please, anyone who has any insight on this would be very much appreciated.

 

[58sniper]

I realized that I did not have circular logging enabled

You never should have that enabled.

I went through and cleared out a lot of the old logs to free up some space

Probably not the best idea, as you've now eliminated the chance of a point-in-time recovery

What has happened here?

What do the event logs say?

because I simply cannot allow backups/replication to be nonexistant.

Your backups are nonexistant if you're accumulating excess logs everyday. An Exchange aware backup solution will flush the logs automatically. If that's not happening, then your backups aren't working correctly.

When LCR fails over, it's a one time issue and usually "unconfigures" itself. So, if you failover to the LCR copy, you generally have to set it backup in order to fail back.

Pat Richard MVP

 

[LibertyGuy]

I was under the impression that this LCR feature was a 'set it and forget it' type of solution. I do not see any options that would automatically flush the logs. I thought that circular logging overwrote old log files instead of keeping them around to take up space? Is this not what I would want?

when I run the restore-storagegroupcopy cmdlet in EMS, it simply says that my storage group identities do not exist.

I haven't fully deleted the logs that I've cleared. They are in my recycle bin. If I restore these, what options would then be open to me?

I'd like to avoid a point-in-time recovery, as this would hose any updates that people have done since yesterday to the company calendar, among other things. However, I know that I must do what I must do. Any ideas?

Thanks.

 

[LibertyGuy]

Apologies for the double post, but I forgot to mention that the Event Log shows nothing in reference to this issue.

 

[58sniper]

I was under the impression that this LCR feature was a 'set it and forget it' type of solution. I do not see any options that would automatically flush the logs. I thought that circular logging overwrote old log files instead of keeping them around to take up space? Is this not what I would want?

LCR has nothing to do with logs. You must use an Exchange aware backup solution to flush the logs. Enabling circular logging will truncate the logs, but it completely destroys any chance of a point in time recovery - something that you generally want.

I'd like to avoid a point-in-time recovery, as this would hose any updates that people have done since yesterday to the company calendar, among other things.

Say what? You wouldn't have the ability to EVER do a point in time recovery, such as in the event of a database corruption.

You need to implement a proper backup solution to deal with the logs. Simple as that.

Pat Richard MVP

 

[LibertyGuy]

So basically, my storage groups disappeared for no apparent reason, and now I have no way of getting them back?

I don't quite understand why it won't let me recreate these storage groups and re-enable LCR.

LCR automatically generates all these log files. I don't get what a 'proper backup solution to deal with the logs' would involve.

All I wanted was a backup of our mailbox databases, which is no longer occuring due to the excessive amount of logs causing our drive to overflow, and suddenly making all my storage groups disappear.

I'd really like some help. Maybe I coulda/woulda/shoulda on some things, but none of those seem to relate to why this problem has occured. As you've probably figured out, I am more of a hardware and physical network kind of guy. I haven't much experience with Exchange Server.

What can I do? The event logs don't reference any part of this problem other than the entries warning me of my E drive becoming full.

 

[Davetoo]

Not following exactly which storage groups vanished. Did you lose your main stores and thus all of your users email within those stores? or did you lost the LCR copy of the storage groups they were replicating?

I'm Certifiable, not cert-ified.
It just means my answers are from experience, not a book.

There are no more PDC's! There are DC's with FSMO roles!

 

[LibertyGuy]

In Exchange Management Console, under Server Configuration -> Mailbox and under the 'Database Management' tab, I had 4 storage groups listed there, and they all had LCR enabled. The replication was backing up to a second HDD (E:\)

I got warnings that LCR was failing because of my E: drive becoming full.

I cleared some room on E, refreshed that page in EMC, and it changed from showing my storage groups and LCR copy status, to simply saying "There are no items to show in this view"

If I try to Create New Storage Group, it gives me the error: "The parent object for <Storage Group Name> could not be found. Please check that InformationStore exists"

I'm dumbfounded.

 

[Davetoo]

Start with the obvious first...you don't have a filter setup on the view do you? Look up at the top of that window...

I'm Certifiable, not cert-ified.
It just means my answers are from experience, not a book.

There are no more PDC's! There are DC's with FSMO roles!

 

[58sniper]

Time for a call to PSS.

Maybe I coulda/woulda/shoulda on some things, but none of those seem to relate to why this problem has occured

Your lack of an Exchange aware backup solution is what caused the excessive number of logs, which you deleted, which is likely a contributory factor to your problem.

LCR isn't a long term "backup".

Pat Richard MVP

 

[LibertyGuy]

K 58sniper, you're not helping. Instead of trying to belittle me, perhaps offer a suggestion as to what I can do NOW, because obviously I can't rewind and go fix the past.

PS: I was only doing what my manager asked me to set up: LCR. As stated, I'm new in the field and am a first timer working with Exchange. The boss checked on my work and it obviously was good enough for him.

DaveToo: Thanks for making suggestions. I have no filters applied to my view.

 

[Davetoo]

Pat was simply telling you that, indeed, it was your lack of a proper backup which caused your logs to inflate to the point of causing the failure, that's all.

Ok...next, can your Exchange users connect and use the mail server and their email on it, or is it all gone?

I'm Certifiable, not cert-ified.
It just means my answers are from experience, not a book.

There are no more PDC's! There are DC's with FSMO roles!

 

[LibertyGuy]

I didn't mean to sound hostile. I thank you both again for your insights.

Email is working just fine. Also, the .edb files are still there, and actually they seem to be still getting modified and staying current (based on the last modified date of the files)

Strange.

 

[Davetoo]

Have you rebooted the server? If not, do so.

I'm Certifiable, not cert-ified.
It just means my answers are from experience, not a book.

There are no more PDC's! There are DC's with FSMO roles!

 

[LibertyGuy]

Planning on it at the end of the day (4:30 central).

Don't want to get the 500+ calls about email not working while people are in the office.

Hehe, I'm sure you know what I mean.

 

[Davetoo]

Lost my default gateway router this morning at 10:00am...yeah, I know what you mean. I don't even answer the phone anymore when that happens.


I'm Certifiable, not cert-ified.
It just means my answers are from experience, not a book.

There are no more PDC's! There are DC's with FSMO roles!

 

[58sniper]

Pat was simply telling you that, indeed, it was your lack of a proper backup which caused your logs to inflate to the point of causing the failure, that's all.

Correct, using my normally abrasive manor.

As Dave mentioned, restarting the server might help. Just restarting the services might help as well.

When you open a PowerShell window and do a get-mailboxdatabase, is there anything listed?

Pat Richard MVP

 

[LibertyGuy]

get-mailboxdatabase causes a brief pause, and then just goes back to C:\Documents and Settings\Administrator>_

Uh oh.

 

[58sniper]

Well, the fact that the PowerShell and the EMC don't show anything at least means they are both consistent.

I don't have a great feeling about it, but bouncing the services/server might be all you need. Unfortunately, it might be the final straw and could get even worse.

Pat Richard MVP

 

[LibertyGuy]

I rebooted and it didn't seem to change a thing. I guess that's better than completely breaking down.