Re: corrupt e00.log
- From: "Bill Brehm" <don't want any spam>
- Date: Mon, 4 Jul 2005 09:29:07 +0800
Dear MVPs. I hope one of you can look at this post and suggest a course of
action. It's Monday morning here and I have to go to work and everyone will
be asking when it will be fixed. Thanks. Bill
"Bill Brehm >" <<don't want any spam> wrote in message
news:O4g8fAffFHA.3936@xxxxxxxxxxxxxxxxxxxxxxx
>I went ahead and tried to follow yuor instructions. First I moved every
>file
> from MDBDATA to a holding location. I then ran the restore specifying a
> temporary location. After the restore, MDBDATA had the following files.
>
> e00.chk, e00.log, priv1.edb, priv1.stm, pub1.edb, pub1.stm, res1.log,
> res2.log and tmp.edb. The priv1.edb and priv1.stm files are slightly
> smaller
> than the ones in the holding area, as would be expected since they are
> missing the last week's data. In the temporary location, under First
> Storage
> Group, there are two files - e0000980.log and restore.env.
>
> I copied e0000980.log into the MDBDATA. I copied e0000980 through
> e0000996.log from the holding area. At first I was concerned about the
> duplicate e0000980.log files, but I compared them first and found they are
> identical.
>
> I ran C:\Program Files\Exchsrvr\BIN>eseutil /cc "C:\_temporary location
> for
> restore log files\First Storage Group" and I get the following result.
>
> Microsoft(R) Exchange Server(TM) Database Utilities
> Version 6.0
> Copyright (C) Microsoft Corporation 1991-2000. All Rights Reserved.
>
> Using restore environment
> Restore log file: C:\_temporary location for restore log files\First
> Storage Group
>
> Restore Path: C:\_temporary location for restore log
> files\First Storage Group
> Annotation: Microsoft Information Store
> Server: MAIL
> Backup Instance: First Storage Group
> Target Instance: First Storage Group
> Restore Instance System Path: C:\_temporary location for restore log
> files\First Storage Group
> Restore Instance Log Path: C:\_temporary location for restore log
> files\First Storage Group
>
> Databases: 2 database(s)
> Database Name: Mailbox Store (MAIL)
> GUID: 08EE9F11-2389-41BA-98A641BAB4312C6E
> Source Files: C:\Program
> Files\Exchsrvr\mdbdata\priv1.edb C:\Program
> Files\Exchsrvr\mdbdata\priv1.stm
> Destination Files: C:\Program
> Files\Exchsrvr\mdbdata\priv1.edb C:\Program
> Files\Exchsrvr\mdbdata\priv1.stm
>
> Database Name: Public Folder Store (MAIL)
> GUID: 64AF02A4-8422-43DC-9C8343DC99767E27
> Source Files: C:\Program
> Files\Exchsrvr\mdbdata\pub1.edb C:\Program Files\Exchsrvr\mdbdata\pub1.stm
> Destination Files: C:\Program
> Files\Exchsrvr\mdbdata\pub1.edb C:\Program Files\Exchsrvr\mdbdata\pub1.stm
>
> Log files range: E0000980.log - E0000980.log
> Last Restore Time: Thu Jan 01 08:00:00 1970
>
> Recover Status: recoverEnded
> Recover Error: 0xC8000262
> Recover Time: Fri Jul 01 11:35:52 2005
>
> Restoring ....
> Restore to server: MAIL
> Target Instance: First Storage Group
> Operation terminated with error -939523486 (Existing log file has bad
> signature.) after 0.640 seconds.
>
> Now what please?
>
> Thanks,
>
> Bill
>
>
> "Bill Brehm" <don't want any spam> wrote in message
> news:%23%23P9l5cfFHA.3904@xxxxxxxxxxxxxxxxxxxxxxx
>> Just one question, since I want to make sure I understand and can get it
>> right the first time. Btw, my log files and database files are all in the
>> same directory on the same drive, as I only have one (two RAIDed as one)
>> drive.
>>
>> So I move the files into a holding location so the original MDBDATA is
> free
>> to take the fixed files. I restore to a temporary directory, not the
>> original MDBDATA. The restore will bring in the databases and the log
> files
>> that there before the last backup. Then I copy the e*.log, except e00.log
>> from the holding location to the temporary directory. So it will look
>> like
> I
>> never ran the last backup and the the number of log files had just kept
>> increasing. Running eseutil will read through all the databases and log
>> files and rebuild the databases in the MDBDATA directory.
>>
>> Sound right?
>>
>> I will need to free up some disk space since it looks like I'll have
>> three
>
>> copies of the files at the same time.
>>
>> "Leif Pedersen [MVP]" <leif.pedersenNOSPAM@xxxxxxxxxx> wrote in message
>> news:udWPvPafFHA.1416@xxxxxxxxxxxxxxxxxxxxxxx
>> > Hi,
>> >
>> > This should help:
>> >
>> > 1. Clean out all store directories of log files and databases
>> > 2. Restore from backup without choosing the "last restore set" check
> box
>> > 3. Move the log files from after the backup and just before E00.log
> into
>> > the transaction log directory. So for example if you have
>> > E000003.log-E00000E.log that goes along with the backup those should go
>> > into
>> > the temp directory that you specify when you do the restore. Then you
>> > will
>> > move the E00000F.log-whatever the last one you have is into the
>> > transaction
>> > log directory.
>> > 4. Now run the eseutil /cc and it will play through all of the log
> files
>> > in
>> > the temp directory and the transaction log directory.
>> > 5. Once the /cc completes then your databases should be in a
>> > consistent
>> > state. At that point you should be able to mount your stores. You may
>> > have
>> > to move the transaction log files out of the log directory in order to
> do
>> > so.
>> >
>> >
>> > Leif
>> >
>> > "Bill Brehm" <don't want any spam> skrev i en meddelelse
>> > news:uP1BsJYfFHA.3656@xxxxxxxxxxxxxxxxxxxxxxx
>> >> I read every related post before posting my own. I'm using Windows
>> >> 2000
>> >> Server and Exchange 2000. Sorry, but I'm writing this from home so
>> >> some
>> >> of
>> >> the terms and quotes may be paraphrased.
>> >>
>> >> Just today the power cord was kicked out while the server was running.
>> > After
>> >> restart, the Information Stores won't mount. They give an error that
> "The
>> >> database files in this store are corrupt." I ran eseutil /ml on all
>> >> the
>> > log
>> >> files to discover that only e00.log is corrupt. I looked with a hex
>> >> editor
>> >> and compared to the most recently created e0000960.log. e00.log seems
> to
>> >> have double headers at the beginning. It's not a torn write, because
> that
>> >> should be fixed automatically.
>> >>
>> >> I tried using eseutil recover, repair and the other one, options /r,
>> >> /p
>> > and
>> >> /c. None could help. I tried renaming (to hide it) e00.log. I tried
>> >> replacing e00.log with a copy of e0000960.log. Nothing helped. An
> eseutil
>> >> /mh of priv1.edb showed no problems there. I would be happy to lose
>> > anything
>> >> stored in e00.log. I just want to be able to mount the store and run
>> >> exchange again. How can I do this?
>> >>
>> >> Can I safely run a backup with MSBackup to recover(, as backup normal
>> >> removes all log files and starts over with them)? Can I get a neutral
>> >> e00.log file somewhere? Any other clue? I have a week old backup. If I
>> > have
>> >> to restore that to recover my server quickly I would. What's the
>> >> procedure
>> >> for that if I can't get a suggestion to recover e00.log.
>> >>
>> >> Thanks in advance...
>> >>
>> >>
>> >
>> >
>>
>>
>
>
.
- Follow-Ups:
- Re: corrupt e00.log
- From: Leif Pedersen [MVP]
- Re: corrupt e00.log
- References:
- Re: corrupt e00.log
- From: Bill Brehm
- Re: corrupt e00.log
- From: Bill Brehm
- Re: corrupt e00.log
- Prev by Date: RE: Reposted
- Next by Date: RE: Reposted
- Previous by thread: Re: corrupt e00.log
- Next by thread: Re: corrupt e00.log
- Index(es):
Relevant Pages
|
