3 messages in net.sourceforge.lists.courier-users[courier-users] Re: Maximum of 20 nes...
FromSent OnAttachments
jj...@megalomanic.netAug 20, 2007 12:19 pm 
Sam VarshavchikAug 20, 2007 4:25 pm 
jj...@megalomanic.netAug 21, 2007 8:33 am 
Actions with this message:
Paste this link in email or IM:
Paste this link in email or IM:
Atom feed for this thread
Paste this URL into your reader:
Subject:[courier-users] Re: Maximum of 20 nested SPF queries exceeded. /Discrepancy in spf.cActions...
From:jj...@megalomanic.net (jj@megalomanic.net)
Date:Aug 21, 2007 8:33:27 am
List:net.sourceforge.lists.courier-users

Okay.

I changed the source, recompiled and ran some tests using telnet on another server. It's my opinion the return value for "Maximum of 20 nested SPF queries exceeded." in ./rfc1035/spf.c should be changed from SPF_FAIL to SPF_UNKNOWN as this would match the /* comments */ above that code.

Before change (using telnet) Courier gives up completely:

helo malexmedia.net mail from:<cour@malexmedia.net>

517 SPF fail cour@malexmedia.net: Maximum of 20 nested SPF queries exceeded.

After change (in mail headers) Courier still gives up, but does not fail the SPF:

Received-SPF: unknown (Maximum of 20 nested SPF queries exceeded.) SPF=MAILFROM; sender=cour@malexmedia.net; remoteip=::ffff:127.0.0.1; remotehost=localhost; helo=hi; receiver=server;

I should also note that I tested this using courier-0.53.3.

Jim

On Mon, 20 Aug 2007 19:25:38 -0400, Sam Varshavchik <mrs@courier-mta.com> wrote:

jj@megalomanic.net writes:

I downloaded the Debian sources to confirm I had the same source. What catches my attention is that the comments of this particular block of code indicate that SPF_UNKNOWN should be returned however the code is actually returning SPF_FAIL which agrees with what I am seeing in the logs. I'm not sure that this is a bug or if it is just commented incorrectly. Which should it be?

You might be right, and the status result for this error condition should probably be changed. Feel free to change it, and see what happens.