On Nov 18, 2009, at 3:14 PM, Sam Varshavchik <[email protected]> wrote:

> Greg Earle writes:
> 
>> I have a really old version of Courier (0.45.6) running in an environment 
>> that is locked down for a JPL flight project, so it cannot be replaced 
>> until/unless the whole machine is replaced.
>> On extremely rare occasions, I get e-mails that are empty, as received on my 
>> end via this Courier server.  As I receive it, it only has
>> Return-Path:
>> Delivered-To:
>> Received:
>> Received:
>> Received:
>> From:
>> and that's it.  No other headers, no message body.
>> I asked someone else who received the same message but on a different 
>> (non-Courier) server to forward me the raw message, and he got it in its 
>> entirety.
>> When I examined his copy, the "To:"/"Cc:" lines are both one line long, but 
>> with a huge amount of recipients.  Not many in the "Cc:" field - only 6 - 
>> but the "To:" field is huge - I'm guessing well over 100 people, maybe even 
>> 200.
>> It's telling that in the empty message I received, the "To:" header is 
>> missing as is "CC:" and everything that follows.
>> Clearly this is giving Courier heartburn - is there any knob to twiddle to 
>> let these messages through?
> 
> There are no knobs to tweak. Courier itself does not care what's in the 
> headers of incoming mail. The only time Courier modifies or does anything 
> with headers is on messages originating locally, by submitting them to the 
> sendmail wrapper. Its only treatment on all other mail is to prepend its own 
> Received: header.
> 
> Perhaps you have some custom local mail delivery script that's barfing.

Sam,

Sorry to dredge up this old issue but I am still having this same problem, and 
my users are getting ornery about it.

Today I just discovered something new.  I got another "empty" message.

The entire contents of one of the "empty" messages (as it sits on-disk) is 798 
bytes:

> wc -c 1267750881.M963940P19165V00800078I000DC563_55.XXXX,S=798:2,S     798 
> 1267750881.M963940P19165V00800078I000DC563_55.XXXX,S=798:2,S

If the "Return-Path" and "Delivered-To" header lines are removed, the count is 
709:

> egrep -v Return-Path\|Delivered-To 
> 1267750881.M963940P19165V00800078I000DC563_55.XXXX,S=798:2,S | wc -c
     709

I just looked in /var/log/syslog and the line for this delivery is

> grep elizabeth.r syslog | grep size= | grep earle | head -1
Mar  4 16:42:51 XXXX courierlocal: [ID 702911 mail.info] 
id=000377CA.4B90538B.00001F84,from=<[email protected]>,addr=<[email protected]>,size=709,success:
 Message delivered.

You can see the "size=709", matching the size of the file without the 
Return-Path and Delivered-To.

Given that Courier hands off the message to my preline/procmail back-end for 
actual delivery, doesn't this imply that the message is, in fact, already empty 
when Courier is finished processing it, given the "size=709"?

In other words, if the message was still "good" when Courier handed it off to 
the back-end for delivery and something else mangled it afterwards, I would 
expect the "size=NNNN" number in the syslog entry to be different/larger.  
Right?

Does Courier get back any return status from the back-end pipeline ("procmail" 
in my case) saying "OK, I've delivered this message for you, and the size 
as-delivered was this number"?  (I suspect not.)

It looks to me like this "size=709" came from one of

module.esmtp/esmtpclient.c:             ctf->msgsize=stat_buf.st_size;
module.local/deliver.c: createInfo.msgsize=stat_buf.st_size;
module.local/localmail.c:                       ctf.msgsize=stat_buf.st_size;

but I can't tell which one.  Since I am using preline/procmail in 
"etc/courierd" for delivery, I'm guessing it's the first one.

        - Greg



------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
courier-users mailing list
[email protected]
Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users

Reply via email to