Hi Justin, I've experienced exactly what you describe (certain cr@p array controllers not doing their job), and restoring the LAST GOOD backup and replaying the logs is exactly the way to go IF your logs are on a separate spindle set. If not, there is a good chance that one or more of the log files are corrupt as well, in which case you will only be able to recover to the point of the last good backup (meaning you will lose everything since the last good backup). If your logs are on a separate spindle set, and you aren't using circular logging, then this is a no-brainer. Restore! It's much faster and easier then attempting to repair the database.
ESEUtil can cause loss of data, so I would only use this as a last resort. BTW, even if the logs are on the same array, I'd try the restore anyway. If you do what MS and everyone else have suggested (which is to copy all DBs and logs first), you can always get back to where you are today and then attempt ESEUtil. HTH, Linton -----Original Message----- From: Salandra, Justin A. [mailto:jasalandra@;chcsnet.org] Sent: Tuesday, November 12, 2002 12:49 PM To: '[EMAIL PROTECTED]' Subject: RE: [ActiveDir] OT: Exchange -1018 Error Thanks for the info, I have asked MS by e-mail if I could do a eseutil /mh priv.mdb > output.txt, if they say yes I will do it tonight. Also I do not have circular logging enabled. -----Original Message----- From: Burns, Clyde [mailto:Clyde.Burns@;nortonhealthcare.org] Sent: Tuesday, November 12, 2002 10:19 AM To: [EMAIL PROTECTED] Subject: RE: [ActiveDir] OT: Exchange -1018 Error Id go with what MS is recommending. Ive gone both ways (with regards to Exchange 5.5) in restoring a known good database and replaying the transaction logs and having to recover a corrupted database. I would just make sure to copy off all the contents of your exchsrvr\mdbdata contents somewhere else before trying the restore. Worst case then is you can get Exchange back to its current state with a copy of your priv, pub, and log files. Some things to check first though. Go into the advanced tab under the server properties in the 5.5 admin program. (Org -> site -> configuration -> server) See if you have circular logging turned on or off. Ifs its off your ok with restoring the database from tape and replaying the logs. If its on you may lose data by restoring an older priv and pub.edb (though it don't sound like it if you have 500+ separate log files) One other thing to check should you decide to go the 'repair what you have' route. Shut down all your exchange services on the server. Find the path for the following 3 files. Dir.edb pub.edb and priv.edb. Open a command prompt on the server and go the directory where each of those files are at and use the following command eseutil /mh priv.mdb > output.txt Just replace priv.mdb with pub and dir. Open the output.txt file and look about 17 lines down for the "State:" line. If all 3 say Consistent you don't have to worry about the transaction logs. They have been committed to the database. When Exchange 5.5 information store service is shut down it commits the contents of the logs to the database. If the logs are not committed to the database for whatever reason the header dump will show the database to be in an Inconsistent state. Short answer is you could either move all of the edb*.log files out of your mdbdata directory or do away with them manually (definately talk to MS PSS before doing this yourself if your not confortable.) Clyde Burns -----Original Message----- From: Salandra, Justin A. [mailto:jasalandra@;chcsnet.org] Sent: Tuesday, November 12, 2002 8:59 AM To: ActiveDir (E-mail) Subject: [ActiveDir] OT: Exchange -1018 Error Hello everyone, I know this is off topic but you guys and girls always give good advice. I have been getting event Ids 116, 118 and 200 a lot in my application log. The chain of events went like this I installed SP3 for W2K The next day the Exchange Database went through some type of conversion and logged Event ID 174 hundreds of times (Table/Column/Index Definintions) On that same day one of my Hard Drives in my RAID 5 Config failed, I replaced it 3 days later Since then every time the online maintenance on the exchange database ran it logged a Database Page Cache error (Event ID 116 and 118 and 200) And now ArcServe cannot backup the database fully and commit the logs to the database so I now have 500 log files in my mdbdata directory. MS Support wants me to restore from the day the HD failed after I rename the mdbdata directory and create a new one. Then they want me to copy in all the log files and wait for the log files to replay into the database. The only other option is to do a eseutil or build a new server and migrate everyone over. Have any of you done what MS is recommending? Do any of you feel that this is the way to go? Does anyone know what would of caused these errors, was it SP3 or was it the HD Failure? Any help is appreciated Justin A. Salandra, MCSE Senior Network Engineer Catholic Healthcare System 914.681.8117 office 646.483.3325 cell [EMAIL PROTECTED] <mailto:jasalandra@;chcsnet.org> List info : http://www.activedir.org/mail_list.htm List FAQ : http://www.activedir.org/list_faq.htm List archive: http://www.mail-archive.com/activedir%40mail.activedir.org/ List info : http://www.activedir.org/mail_list.htm List FAQ : http://www.activedir.org/list_faq.htm List archive: http://www.mail-archive.com/activedir%40mail.activedir.org/ List info : http://www.activedir.org/mail_list.htm List FAQ : http://www.activedir.org/list_faq.htm List archive: http://www.mail-archive.com/activedir%40mail.activedir.org/ List info : http://www.activedir.org/mail_list.htm List FAQ : http://www.activedir.org/list_faq.htm List archive: http://www.mail-archive.com/activedir%40mail.activedir.org/
