I have the following situation with the databases exim builds.
(ip's and some hostnames changed to protect the unaware, see [[ ]] )
I have entries in my exim_mainlog as follows:
1999-12-30 14:13:09 123B8g-0001Si-00 Failed to open retry database
1999-12-30 14:15:07 123nqc-0001L4-00 Failed to open
wait-remote_smtp database: Invalid argument
Suspecting these as indicators of bad things, I try the following
# /usr/sbin/exim_dumpdb /var/spool/exim retry
Failed to open database file retry (spool directory = /var/spool/exim)
# /usr/sbin/exim_dumpdb /var/spool/exim reject
30-Dec-1999 04:42:25 F 0.0 postmaster@[[some_ip : some other_ip]]
28-Dec-1999 18:01:54 F 0.0 spammer@???:[[some_host_name]]
# ls -l /var/spool/exim/db
total 32
-rw-r----- 1 mail mail 12288 Dec 30 04:42 reject
-rw-r----- 1 mail mail 0 Dec 28 17:37 reject.lockfile
-rw-r----- 1 mail mail 16384 Dec 1 18:36 retry
-rw-r----- 1 mail mail 0 Dec 30 14:02 retry.lockfile
-rw-r----- 1 mail mail 16384 Dec 1 18:36 wait-remote_smtp
-rw-r----- 1 mail mail 0 Dec 30 14:05 wait-remote_smtp.lockfile
# ls -l /usr/sbin/exim /usr/sbin/exim_dumpdb
-rwsr-xr-x 1 root root 468404 Dec 23 14:40 /usr/sbin/exim
-rwxr-xr-x 1 root root 22728 Dec 23 14:40 /usr/sbin/exim_dumpdb
# /usr/sbin/exim -bV
Exim version 3.12 #2 built 23-Dec-1999 01:49:00
Copyright (c) University of Cambridge 1999
I can't figure out why this is happening, and I feel it should not happen.
How do I arrange that exim can open these databases? Is there a safe way
to rebuild the retry and wait-remote_smtp databases while the system is
providing mail service? I am still poring through the docs to find an
answer. I can't offer a reason why the reject database is treated
differently by exim.
Gerhard Paseman, 1999.12.30
Prado Internet Access, Inc.
paseman@???
