[exim] bounces left on the queue

Top Page
Delete this message
Reply to this message
Author: Kjetil Torgrim Homme
Date:  
To: exim-users
Subject: [exim] bounces left on the queue
I'm having a problem with bounces to users who are over quota staying
indefinitely on the queue. the setup runs Exim 4.66 and Cyrus with
virtual users and domains. the usernames and domains in Cyrus do not
match the mailing lists (historical reasons), so the config has a
redirect router from e-mail address to username@institutionname. the
transport is smtp with lmtp protocol, to localhost. we've enabled
retry_use_local_part on the router(s).

consider this message, on the queue for 8 days, even though final cutoff
is 2d.

        # exim -brt gybi10@???
        Retry rule: *  *  F,2h,5m; G,16h,1h,1.5; F,2d,6h; 


running the queue manually doesn't touch it:

        # exim -v -q 1H79bL-0007NF-W0 1H79bL-0007NF-W0
        LOG: MAIN
          cwd=/var/log/exim 5 args: /local/opt/exim-server/exim -v -q 1H79bL-0007NF-W0 1H79bL-0007NF-W0
        LOG: queue_run MAIN
          Start queue run: pid=26466
        delivering 1H79bL-0007NF-W0 (queue run pid 26466)
        LOG: retry_defer MAIN
          == gybi10@sogneskoler <gybi10@???> routing defer (-51): retry time not reached
        LOG: queue_run MAIN
          End queue run: pid=26466


here's the retry hint lookup, excerpted from -d output:

        Considering: gybi10@sogneskoler
        unique = gybi10@sogneskoler
        dbfn_read: key=R:sogneskoler
        dbfn_read: key=R:gybi10@sogneskoler
        dbfn_read: key=R:gybi10@sogneskoler:<>
        no domain retry record
        post-process gybi10@sogneskoler (1)
        LOG: retry_defer MAIN
          == gybi10@sogneskoler <gybi10@???> routing defer (-51): retry time not reached


if I use -Mc, however, it gets processed:

        # exim -v -Mc 1H79bL-0007NF-W0 
        LOG: MAIN
          cwd=/var/log/exim 4 args: /local/opt/exim-server/exim -v -Mc 1H79bL-0007NF-W0
        delivering 1H79bL-0007NF-W0
        Connecting to localhost.uio.no [127.0.0.1]:24 ... connected
          SMTP<< 220 mail1.fronter.uio.no LMTP Cyrus v2.2.12 ready
          SMTP>> LHLO mail1.fronter.uio.no
          SMTP<< 250-mail1.fronter.uio.no
                 250-8BITMIME
                 250-ENHANCEDSTATUSCODES
                 250-PIPELINING
                 250-SIZE
                 250-AUTH EXTERNAL
                 250 IGNOREQUOTA
          SMTP>> MAIL FROM:<> SIZE=21230
          SMTP>> RCPT TO:<gybi10@sogneskoler>
          SMTP>> DATA
          SMTP<< 250 2.1.0 ok
          SMTP<< 452 4.2.2 Over quota
          SMTP<< 503 5.5.1 No recipients
          SMTP>> QUIT
        LOG: MAIN
          == gybi10@sogneskoler <gybi10@???> R=local_user T=lmtp_normal defer (-44): SMTP error from remote mail server after RCPT TO:<gybi10@sogneskoler>: host localhost.uio.no [127.0.0.1]: 452 4.2.2 Over quota
        LOG: MAIN
          ** gybi10@sogneskoler <gybi10@???>: retry timeout exceeded
        LOG: MAIN
          gybi10@sogneskoler <gybi10@???>: error ignored
        LOG: MAIN
          Completed


of course, a cron job running -qf every night would "fix" the problem,
but it doesn't appeal to me to use such a band-aid :-)

any hints?
--
Kjetil T.