No email - amavis suicidal!

Discussion in 'Installation/Configuration' started by lusso64, Mar 2, 2011.

  1. lusso64

    lusso64 New Member

    I have had ispConfig in various versions running for several years with no trouble other than that which I have caused.

    I have just come up with a new problem though... I am not receiving any email.

    Background: everything was working until the entire server froze. I am not 100% sure what happened, but it may have been DOS attack as I host a website that attracts this sort of thing. Anyways, I had to have the server power cycled. Since then, no email.

    From the mail error logs...

    Mar 1 19:31:56 ubuntu postfix/smtpd[1482]: warning: address not listed for hostname localhost
    Mar 1 19:49:05 ubuntu amavis[3063]: (!!)TROUBLE in pre_loop_hook: db_init: BDB no dbC: __fop_file_setup: Retry limit (100) exceeded, File exists. at (eval 94) line 302.
    Mar 1 19:49:05 ubuntu amavis[3063]: (!)_DIE: Suicide () TROUBLE in pre_loop_hook: db_init: BDB no dbC: __fop_file_setup: Retry limit (100) exceeded, File exists. at (eval 94) line 302.
    Any advice on where to look would be greatly appreciated.

    Thanks, Dave
  2. falko

    falko Super Moderator ISPConfig Developer

    Did you check if your hard drive(s) are alright?
  3. lusso64

    lusso64 New Member

    There's no reported errors and it's a RAID 5. I will check the controller this evening to be sure.

    All other aspects of the machine are working well - http, ftp, etc.

    Does amavis utilize a Berkely DB on ISPCONFIG? It seems the error message is indicating that it's looking for one.

    Thanks, Dave
  4. till

    till Super Moderator Staff Member ISPConfig Developer

    Amavis uses mysql in a ispconfig setup for its configuration. But it might be that spamassassin uses a berkley db for its spam training data internally.
  5. lusso64

    lusso64 New Member

    Well, I rebooted the server and checked the array controller for errors and bad sectors. Nothing unusual there.

    After this, the server went through it's normal boot process and once it was up and running - presto! About 4000 emails come through!

    It's as if the error had never happened. From the other reading I have done over the past day or 2, it seems this error could be caused when the BDB has a lock on the file that isn't released when amavis is restarted. This could be explained, I guess, by the crash that I had. I still need to find the cause of this, but for the time being, I'm happy to be up and running again.

    Thanks, Dave
  6. Davide

    Davide Member

    I know it's an OLD thread, but it happened to me today.
    Found a solution here:
    You can solve it by stoppping amavis:

    /etc/init.d/amavis stop
    /etc/init.d/amavis-mc stop
    /etc/init.d/amavisd-snmp-subagent stop

    deleting all berkeleydb files:

    rm /var/lib/amavis/db/*

    And restarting amavis:

    /etc/init.d/amavis start
    /etc/init.d/amavis-mc start
    /etc/init.d/amavisd-snmp-subagent start
    till likes this.

Share This Page