Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations Wanet Telecoms Ltd on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

-1018 error 2

Status
Not open for further replies.

Haleon

IS-IT--Management
Feb 2, 2004
80
US
Hello, I'm an Exchange newb, and we've got a problem with our server. Whenever we try to back up the database or anything, we get a -1018 error thrown. I've been told that this is often a hardware error, but I ran a full diagnostics on the machine last Saturday and came up with nothing. If anyone could help me out in this area, I'd be very appreciative. Thanks!
 
Haleon,

I'd search MS's Exchange site for the error code. I thought the 1018 was a database error and depending which Exchange you have there are different ways of dealing with it.

I went through about a 12 hr window with no exchange last week because of that and other errors. Good luck.

Pat
 
Repetive 1018 errors often indicate a corrupted database. The underlying problem is usually a disk subsystem problem.(ie bad hard drive) The solution is to find and fix the problem then restore the datebase from a known good copy.

The most common causes:
*hardware problem
*bad sectors on a harddrive
*improper SCSI temination
*Trouble with write back cache on disk controller
*faulty device drivers
*firmware bugs in the disks or disk controller

Hope this helps
George
MCSA (Win2k)A+, Net +
 
Thanks for the help guys,

The problem is that we don't have a good backup of our Exchange server. The guy in charge of it got fired three weeks ago, so I'm supposed to get a backup of it. I guess the only thing I can do at this point is to run integrity checker and defrag to see if I can purge the bad emails from the database? Then I could back it up, right?

And I ran a full diagnostics on our server last Saturday and nothing showed up. Any advice on how to find out what may be causing the 1018 error?
 
Thanks again for your help, George.

One more thing (since you seem to be very knowledgeable), is it safe to delete the contents of the badmail folder? Mine is a couple of gigs large.

Also, does the badmail folder count towards the size of the information store, or is it seperate?

Thanks!
 
Yes, you can delete the contents of Badmail folder. And No, it does NOT count towards the size of the information store. The default location is root:\Exchsrvr\Mailroot\<vsi #>\Badmail, where <vsi #> is the specific virtual server.

&quot;BadMail Contains all messages that caused errors when the connector tried to process them. When messages are transferred here, no further delivery attempts will be made. They are stored indefinitely until an administrator manually deletes them.&quot; From Exchange Help files.

I &quot;seem to be very knowledgeable&quot; mostly because I am preparing to take the MS Exchange 2000 exam next week and remembered where to find the answers (-:


George
MCSA (Win2k) A+, Net +
 
George, you're awesome. I'll pay for the next training book you need. ;)

Maybe you could offer me some advice.

Twice in the last week I've gotten the error that the Exchange store has reached it's maximum size of 16348 megs (something like that); however, when I manually check the size of the mailbox folders on the M: it only comes up to about 10 gig. I also deleted nearly a gig worth of &quot;deleted items&quot; contents to try and bring the size down, but it doesn't seem to have helped.

Is right clicking the mailbox folder on the M: drive and selecting its properties not an accurate way to determine the size of the private database? Our public folders is only about 5 meg. It's the private one I'm worried about.

I've done that little reghack twice that allows you to temporarily mount a 17 gig database, but that's only going to work for so long. I plan to come in this Saturday to do a little work, but I don't really know where to start. I was going to attempt to defrag the database, but I think the last time I tried that I got -1018 error thrown again.

The situation here (as I'm sure you can tell) isn't that great. The guy taking care of all of this before me got fired, and I kind of got stuck with fixing a ton of problems that may not even be fixable.
 
The solution to the database size is getting it backed up properly. It is the transaction logs that are taking up the space and they are not deleted until they have been written to the database and marked as archived (backed up). DO NOT manually delete the transaction logs.

What backup software are you using?

If you are using the Windows back-up, you could do a restore of the last good back-up, using the &quot;Last Restore set&quot; option. This would replay the transaction logs and bring the database up to date. I would make sure my boss understands that there &quot;could&quot; be a loss of some emails from the system using this method. But if you don't do something soon your database will fill up and you won't be able to fix the problem without losing everything since the last successful back up.

Good luck.

George
MCSA (Win2k) A+, Net +
 
George,

We're using ArcServe 2000. The backup is the initial problem. I don't know how long this has been going on for (I just got here three weeks ago), but we CAN'T backup the Exchange database using arcserve or the NTBackup program. That's when the -1018 error gets thrown.

Our last good backup is from too long ago to be even remotely useful now. Pretty much our only option is to fix this current database somehow. Can I run integrity checker and use the eseutil repair switch to fix the -1018 and then make a backup? Is that possible?

Oh, and you're going to hate me, but we have been deleting those transaction logs. That's what the guy before me was doing, and I've had to do it once already. There's no other way to make them go away since we can't do backups.
 
Nah, I don't hate you. It's not my database that is messed up. :)

looks like ESEUTIL /r (recovery mode) would be first thing to try. it attempts to put the database back into a consistent state but does not truncate the data.

if that doesn't work then I would probably try ESEUtil /CC which forces a hard recover that plays the transaction log files and patch files.

lastly would be ESEUTIL /p Repair mode...it validates the database and links AND can truncate data AND cause data loss.




George
MCSA (Win2k) A+, Net +
 
Great, thanks again for all your help.

I'm coming in on Midnight on Saturday to shut down the server and do some work. I'll try all of what you suggested above, or at least all that can be tried in six hours.

Luckily, my boss doesn't particularly care about data loss (as long as it's not massive) because he just wants the problem fixed. If a couple of people lose a couple of emails, so be it. Better than everybody losing everything.

Thank you so much for your help. You've been absolutely invaluable.
 
I almost forgot. Here is a link to a site I have used a number of times. The &quot;Articles and Tutorial&quot; section has over 60 different articles on Exchange 2000.


George
MCSA (Win2k) A+, Net +
 
Thanks for the link, George.

Quick question. Is it the COMBINED size of the priv1.edb and priv1.stm that makes up the information store? Can both of those be 16 gigs apiece, or can they only be 16 gigs combined?
 
Well, I fixed it :)

I think anyway. I ran isinteg and the repair option of eseutil. That allowed me to defrag the database, and now I can back it up normally. It's been running well since Saturday, so now I just cross my fingers. :)

At least we now have a very current backup. I (and my boss) am very happy about that. :)

Thanks so much for everyones help. You were absolutely invaluable. Especially George.
 
That's Great! I've been waiting to hear if you got it fixed.

George
MCSA (Win2k) A+, Net +
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top