Mailing List Archive

In <0AD4DE2DC31C914A81FE604450BE9A8EC483@einstein.johan.homeip.net> <spf-team@strandhuset.net> writes:

> Is there any interest in this group for such a component? If there is I
> am willing to publish it along with the source code.

Yes, I think there would be a lot of interest in such a component.

-wayne

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

>
> Yes, I think there would be a lot of interest in such a component.
>
> -wayne

Great!

This is what is does so far:
- It is a smtp transport event sink implemented as a DLL.
- It is written in C++
- It will parse A, MX, PTR, redirect and ALL mechanisms
- It adds X-headers to the mail describing the rule and the result.

I need to do:
- Macro expansion
- Exist mechanism
- Explain mechanism
- A proper X-SPF-RECEIVED header
- Better error handling, especially for DNS timeouts.
- Listen to anything you have to tell me about errors I have made. (I'm
sure there are several)

Is there a established way this group uses to publish their development
efforts?

/Johan

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

In <0AD4DE2DC31C914A81FE604450BE9A8EC484@einstein.johan.homeip.net> <spf-team@strandhuset.net> writes:

> This is what is does so far:
> - It is a smtp transport event sink implemented as a DLL.
> - It is written in C++

I would suggest investigating the libspf and libspf-alt
implemenations. While they are written in C instead of C++, they will
give you a big head start.

See:
http://libspf.org
http://www.midwestcs.com/spf/libspf-alt/


> - Listen to anything you have to tell me about errors I have made. (I'm
> sure there are several)

In order to try and keep all implemenations as compatible as possible,
there is a test suite available. You need to create a compatible
"spfquery" program, which shouldn't be too hard. Then, run the tests
available at:

http://www.midwestcs.com/spf/tests

I should say that the tests are under very active developement and a
new release should be available Any Day Now.


> Is there a established way this group uses to publish their development
> efforts?

Mostly here and on the #libspf channel on irc.pobox.com


-wayne

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

Thank you very much as I am very interested in this. I have been
following the discussion of spf but have been unable to contribute as my
servers are all Exchange 2000 and "rolling my own" from scratch is
beyond my abilities. I, too, have a test domain and am very much looking
for the day that this kind of protection is available to Exchange
servers.

-----Original Message-----
From: owner-spf-devel@v2.listbox.com
[mailto:owner-spf-devel@v2.listbox.com] On Behalf Of
spf-team@strandhuset.net
Sent: Friday, March 12, 2004 1:04 AM
To: spf-devel@v2.listbox.com
Subject: [spf-devel] MS smtpsvc

Hi!

I have been watching the the implressive growth of SPF the last months.
I have a couple of domains that I administer. For both of them I have
published spf-records, but since they are MS Exchange servers I have
been unable to find a server component to use. Therefore I decided to
build it myself. It is still a very early version, but it seems to work
OK in my test domain.

Is there any interest in this group for such a component? If there is I
am willing to publish it along with the source code.

Best regars,

Johan Strand

-------
To unsubscribe, change your address, or temporarily deactivate your
subscription, please go to
http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

> -----Ursprungligt meddelande-----
> Från: Matthew Speed [mailto:mspeed@mspeed.net]
> Skickat: den 12 mars 2004 19:01
> Till: spf-devel@v2.listbox.com
> Ämne: RE: [spf-devel] MS smtpsvc
>
>
> Thank you very much as I am very interested in this. I have
> been following the discussion of spf but have been unable to
> contribute as my servers are all Exchange 2000 and "rolling
> my own" from scratch is beyond my abilities. I, too, have a
> test domain and am very much looking for the day that this
> kind of protection is available to Exchange servers.
>

I have put up a binary version at http://einstein.strandhuset.net/spf. It will parse most of the mechanisms except macros and include. If you want to test it please note that it is still a very early version. It is NOT to be used in a production environment! I have tested with the official test scripts and it parses those without macro or include correctly. I have also been running it on my test server for several hours without trouble. There might be an issue with coexistence with ORFEE from VamSoft, but I need to investigate that further.

Please send feedack to mailto:spf-team@strandhuset.net.

Source code will be up as soon as I have cleaned it up a bit and added coments where appropriate.

/Johan

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

Hi all!

There is now a SourceForge projet for this component. Please go to
http://sourceforge.net/projects/spf-event-sink/ for further information
and downloads.

Best regards,

Johan

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

> -----Ursprungligt meddelande-----
> Från: wayne [mailto:wayne@midwestcs.com]
> Skickat: den 12 mars 2004 15:36
> Till: SPF Development
> Ämne: Re: SV: [spf-devel] MS smtpsvc
>
> Then, run the tests available at:
>
> http://www.midwestcs.com/spf/tests

> I should say that the tests are under very active developement and
> a new release should be available Any Day Now.

Thanks!
The tests are very useful. If I may, I would like to suggest you add the test numbers to the new format as a you had in the old formt. That made it a lot easier to locate the test that failed.

There is one test that I need to ask about:
99.spf1-test.mailzone.com returns "v=spf1 -all exp=99txt.spf1-test.mailzone.com moo"
The test script suggest that this should return unknown because fo the 'moo'. If I understand the RFC correctly I am required to parse all the entries to the left, even if there is an unknown mechanism. That would make it return fail.

/Johan

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

On Mon, Mar 15, 2004 at 07:17:01AM +0100, spf-team@strandhuset.net wrote:
| 99.spf1-test.mailzone.com returns "v=spf1 -all exp=99txt.spf1-test.mailzone.com moo"
| The test script suggest that this should return unknown because fo the 'moo'. If I understand the RFC correctly I am required to parse all the entries to the left, even if there is an unknown mechanism. That would make it return fail.
|

You're right, that should be a fail.

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

In <0AD4DE2DC31C914A81FE604450BE9A8EC489@einstein.johan.homeip.net> <spf-team@strandhuset.net> writes:

>> Then, run the tests available at:
>>
>> http://www.midwestcs.com/spf/tests
>
>> I should say that the tests are under very active developement and
>> a new release should be available Any Day Now.
>
> The tests are very useful. If I may, I would like to suggest you add
> the test numbers to the new format as a you had in the old
> formt. That made it a lot easier to locate the test that failed.

Removing the numbers was a recommendation by the "perl test guru".
The numbers are also incredibly easy to get out of sync, and if you
renumber the file, it makes it really hard to track changes.

What I do is imply search on the options passed to the spfquery
command.

> There is one test that I need to ask about:
> 99.spf1-test.mailzone.com returns "v=spf1 -all exp=99txt.spf1-test.mailzone.com moo"
>
> The test script suggest that this should return unknown because fo
> the 'moo'. If I understand the RFC correctly I am required to parse
> all the entries to the left, even if there is an unknown
> mechanism. That would make it return fail.

To the best of my knowledge, none of the test scripts mentioned in the
above url ( http://www.midwestcs.com/spf/tests ) say that this should
pass.

There are test files in the libspf-alt that have been modified so that
tests that are known fail do not do not trigger debug output. The new
version uses the "TODO" function instead.


I'm about ready to give up working on the new test scripts. There is
still a *HUGE* amount of work to do on them, but it is sucking all my
time.


-wayne

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

> -----Ursprungligt meddelande-----
> Från: wayne [mailto:wayne@midwestcs.com]
> Skickat: den 15 mars 2004 17:08
>
> Removing the numbers was a recommendation by the "perl test
> guru". The numbers are also incredibly easy to get out of
> sync, and if you renumber the file, it makes it really hard
> to track changes.

OK, point taken.

> To the best of my knowledge, none of the test scripts
> mentioned in the above url (
> http://www.midwestcs.com/spf/tests ) say that > this should
> pass.
OK, My question was answered earlier anyway. Maybe I have an old version. I found a similar test list in the reference implementation. I vill take a look at that also.

> I'm about ready to give up working on the new test scripts.
> There is still a *HUGE* amount of work to do on them, but it
> is sucking all my time.

I really understand that. The tests are a huge help, so all the work is really appreciated!

Best regards,

Johan

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

In <0AD4DE2DC31C914A81FE604450BE9A8EC48C@einstein.johan.homeip.net> <spf-team@strandhuset.net> writes:

>> To the best of my knowledge, none of the test scripts
>> mentioned in the above url (
>> http://www.midwestcs.com/spf/tests ) say that > this should
>> pass.

*ACK* I meant "fail" here instead of pass. *sigh*. None of the SPF
implementations that I know of say that the particular test pass,
rather most say it should "fail", while libspf-alt says "unknown.

Sorry for the confusion!


> OK, My question was answered earlier anyway. Maybe I have an old
> version. I found a similar test list in the reference
> implementation. I vill take a look at that also.

Yes, the Perl implementation should have this test.


-wayne

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com

In <0AD4DE2DC31C914A81FE604450BE9A8EC48C@einstein.johan.homeip.net> <spf-team@strandhuset.net> writes:

>> I'm about ready to give up working on the new test scripts.
>> There is still a *HUGE* amount of work to do on them, but it
>> is sucking all my time.
>
> I really understand that. The tests are a huge help, so all the work
> is really appreciated!

I guess I should point out that I'm not the only one doing work on
these test. For example, I think that 100% of the test stuff in the
perl SPF system was done by others. My work only shows up in the
libspf and libspf-alt distributions.


-wayne

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-devel@v2.listbox.com