monit-general
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Please help: monit alerts make qmail complain!


From: Jan-Henrik Haukeland
Subject: Re: Please help: monit alerts make qmail complain!
Date: 21 Jan 2003 20:37:50 +0100
User-agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.4 (Civil Service)

Jeremy Henty <address@hidden> writes:

> I am setting up monit for the first time to monitor AOLserver.  ATM,
> I am running it in passive mode just to see what it does and does
> not notice.  As a test I set the host to a bogus IP, and monit logs
> these lines:
> 
> [GMT Jan 21 16:40:31] 'nsd' timed out when testing INET[192.168.1.45:80] 
> [HTTP]
> [GMT Jan 21 16:40:31] Sendmail error: 451 See 
> http://pobox.com/~djb/docs/smtplf.html.
> 

On a general note, someone ought to cover D. J. Bernstein with tar and
feathers for reading the RFC 822 like the devil reads the bible. Uhm,
that aside, I suspect the problem is partly due to the passive mode.
It's probably not tested much and when I look at the code I see now
that there is an error:

 - smtp_alert_failed is used in validate:do_restart when in passive
    mode.

 - In alert.c:smtp_alert the event is thus DO_FAILED but this is not
   in the notify[] array further down in the same function and thus
   trigger an alert_on_timestamp instead of a failed event.

(Martin, do you have time to fix this? If not, I can look into this a
bit later). 

Anyway, I'm still not sure how the above problem can manifest itself
in sendmail.c (maybe it's not even releate) but I guess qmail is right
and that somewhere in the mail a bare-line-feed pop's up and is sent.

Since I do not have access to qmail (thank God) maybe you can debug it
(Jeremy) so we can pinpoint the do_send line where the problem occurs.
Debug like so:

 In the directory where monit was built do:

 gdb ./monit
 break sendmail
 break do_status
 run -I

 This will run monit in non-daemon mode and stop in the sendmail
 function. When monit stops in this function, step one line at the
 time until you get the reported error in the do_status function.
 Note which send-line caused the problem and send info back to this
 list. Thanks!

-- 
Jan-Henrik Haukeland




reply via email to

[Prev in Thread] Current Thread [Next in Thread]