Customer had an exchange 2003 box running on a virtual VMWare ESXi infrastructure. After a reboot the store would not mount. Another technician had been checking it out so I came into the game a little late.

After trying various methods to fix the issue, the technician was able to mount the store. Unfortunately there were no mailboxes listed on the store – bad news bears.

The problem was escalated to me and I ran a checkdisk on their drive
chkdsk /r E:
This checkdisk found a few sector corruptions on the primary EDB database file and fixed them. It also found a few miscellaneous table errors and filename issues – which it fixed – but nothing really major besides the EDB file.

Restarted the server and all services came back up. Unfortunately, while mail started to actually flow, I was receiving other eventlog errors:

ESE Database Page Cache – event ID 474. Information Store First Storage Group: The database page read from (file) at offset (offset) failed verification due to a page checksum mismatch. The read operation will fail with error -1018 (0xfffffc06).

Awesome.
Then I got this gem:

MSExchangeIS Mailbox – event ID 1025. An error occurred on database “First Storage Group\Mailbox Store (servername)”. Function name or description of problem: SLINK::EcUpdate Error: Read verification error

Better, no?
Basically it looks like some data corruption had occurred. Now I needed to fix this and get it back up and running ASAP.
Open a command prompt
cd \program files\exchsrvr\bin
eseutil /g "path\to\mdbdata\priv1.edb"
This will check database integrity

/D will defrag mode
/P will repair mode
/C will restore mode
/R will recover mode
/M will file dump mode
/K will checksum check mode
/Y will copy file mode

***EDIT***
Well, looks like I have to do the /R to repair as it came back saying the database was corrupted. /R should only be used if you get a similar message as it attempts to reorganize the database instead of rebuilding from the log files.