The database is corrupted.
If is possible to recreate the secondary indexes... all but one work without problem at all. These are:
account_usernames.bdb
account_email_addresses.bdb
account_addresses_accounts.bdb
address_whitelist_account_ids.bdb
address_blacklist_account_ids.bdb
This one locks up the mpd, but seems to work ok:
quarantined_emails_retraining.bdb
To recreate these secondary indexes:
1. Stop you rmail server, and the mailwasher processes. This is sometimes a challenge under windows, as exchange doesn't really want to let go.
2. Make a copy of your database. This comprises all of the files in /usr/local/mwserver/data, /var/mwserver or \Program Files\MailWasher Server\data, depending on your operating system.
3. Remove the first 5 files above.
4. Start the mpd process. This will recreate the files.
5. Stop the mpd process.
6. Restart Mailwasher and your mail server.
If the problem persists:
Backup, and repeat 1 and 2 above.
Remove quarantined_emails_retraining.bdb
Start the mpd process.
Wait a couple of minutes
If the mpd process has started cleanly then stop it, otherwise kill it off.
Restart MailWasher and your mail server.
Don't forget... TAKE LOADS OF BACKUPS!!!
This is telling you that there is a connectivity issue, and MailWasher Server cannot talk to the FirstAlert! ( aka Content Filtration System or CFS ). As well as it being exactly what it says, and there are problems with routing somewhere, this could be caused by DNS or firewalling issues. We use port 4051 to talk between MWS and the CFS gateways. A quick test is as follows:
The command
telnet native.first-alert.net 4051
should result in the following response ( IP address may vary depending on which gateway you connect to )
Trying 209.213.221.138...
Connected to native.first-alert.net.
Escape character is '^]'.
200 CFS service ready
In one step, this has proved that DNS is working, and there's no firewalling in the way. Depress Crtl-] do return to the telnet> prompt, then quit to exit.