Next message: Hans du Plooy: "Certificate recommendation?"
Hello
While the problem I have with my server is not life and death (or even of
great operational importance), I would appreciate some direction on the
possible cause. In the UK, we have recently moved to BST (British Summer
Time) and my SuSE 10.1 machine, on which Postfix is running, is clearly
running on local time - and indicates the correct local time.
I have 2 instances of Postfix 2.2.9 running, one inbound with loads of
checks and filters and one outbound with very few; the latter acts as the
smart gateway for an Exchange 2000 server on a W2K server on an internal
domain.
The problem is in the log, an extract of which is here:
Apr 8 16:04:02 server247 Mail-Out/smtpd[10787]: connect from
unknown[10.1.1.248]
Apr 8 15:04:02 server247 Mail-Out/smtpd[10787]: B2CB53280C4:
client=unknown[10.1.1.248]
Apr 8 16:04:02 server247 Mail-Out/cleanup[10788]: B2CB53280C4:
message-id=<1176044642 dot 10780 at server247>
Apr 8 16:04:02 server247 Mail-Out/qmgr[3593]: B2CB53280C4:
from=<>, size=573, nrcpt=1 (queue active)
Apr 8 15:04:02 server247 Mail-Out/smtpd[10787]: disconnect from
unknown[10.1.1.248]
The address 10.1.1.248 is IP for outgoing messages, while 10.1.1.247 is for
incoming (from the ADSL router) messages. Another example is:
Apr 8 15:04:03 server247 Mail-Out/smtpd[10787]: connect from
unknown[10.1.1.245]
Apr 8 15:04:03 server247 Mail-Out/smtpd[10787]: CDCFF3280C4:
client=unknown[10.1.1.245]
Apr 8 16:04:03 server247 Mail-Out/cleanup[10788]: CDCFF3280C4:
message-id=<.
local>
Apr 8 16:04:03 server247 Mail-Out/qmgr[3593]: CDCFF3280C4:
from=<>, size=5191, nrcpt=1 (queue active)
In both cases, the smtpd process seems to be using UTC, while every other
process is using local time. In the second instance, the connection was from
the Exchange server, but in both cases, the message was directed to the NIC
with IP 10.1.1.248.
Any suggestions why the time is not consistent?
In expectation, thank you!
Jan Cobb (Miss)
