We are restoring just the public folders database using Backup Exec 9.1.
The main EDB and STM files restored okay. However, the byte count has stalled and now there are tons of .IFS files in the database folder. 40-60 to be precise.
They appear to be rotating every few minutes, with incrementing names such as:
14D-871CFBD0-1541491D.IFS
I am worried about doing anything, as this has taken over 7 hours to restore 70GB worth of public folders - plus another 4 hours doing this. Should I cancel the restore, should I try mounting the database manually, do nothing?
Note - although the database was mounted okay before we had the server crash (due to a failed disk) I am concerned that maybe the size of the database has fallen over the 75GB limit, which might be causing this?
- pub1.edb = 53.6GB
- pub1.stm = 24.79GB
Any suggestions on next steps?
EDIT - Restore has now been running for over 23 hours. Further note, there are 5 transaction log files in the temp folder, with restore.env at 1KB.
While I've never had to restore the entire PF database, and certainly not with a version of BE as old as 9.1, my advice would be to let the job run. If it fails, it will fail and you'll be able to use the error codes in the job history to troubleshoot. Exchange restores (especially one of 75GB) generally take quite a bit longer than file restores as there's much more going on "under the hood".
Does 9.1 have SGMon? the logging tool.. it will be in the program directory, you can actually see it working if the status stops updating.
It will show you all the "under the hood" you can take!
Urgh. Yet another whole day lost, speaking to Microsoft Exchange support.
We had to kill off all the BackupExec services to release locks on the files and then perform a hard recovery on the restored files using eseutil.
Painful, but all up and running again.