pyopenssl-list — General discussion list for users and developers of pyOpenSSL

Re: [pyOpenSSL] FTP/SSL (FTPS)

[Dan W. <dc...@re...>]

On Wed, 2007-06-06 at 11:25 -0400, Nancy Head wrote:
>  
> I'm a beginning programmer trying to figure out how to use Python to
> do FTP/SSL (FTPS) - just as a client. 
> Will pyOpenSSL help me to do this?

Probably, yes.  There aren't any good documents on it, you probably just
have to look for code samples and follow them.  Is FTPS just FTP over an
SSL connection?  If so, the SSL connection bits are pretty trivial to
set up, and you just point the FTP parts at the socket.

Dan

> Thanks for any help...
>  
> -Nancy
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________ pyopenssl-list mailing list pyo...@li... https://lists.sourceforge.net/lists/listinfo/pyopenssl-list

[pyOpenSSL] FTP/SSL (FTPS)

[Nancy H. <ecl...@gm...>]

 I'm a beginning programmer trying to figure out how to use Python to do
FTP/SSL (FTPS) - just as a client.
Will pyOpenSSL help me to do this?

Thanks for any help...

-Nancy

[pyOpenSSL] X509 Extension Bug

[Eli C. <el...@as...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,
This is to Dave or whoever ends up maintaining the authoritative fork:
I just wanted to draw your attention to a bug in the X509 Extension class
which debian has applied a patch for, but seems to be missing
from Dave's svn fork as of r709.

The debian bug report is here:
	http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=355947

The original report as well as the patch is here:
	http://sourceforge.net/tracker/index.php?func=detail&aid=1166109&group_id=31249&atid=401760

- -Eli Collins
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGPrbkbY9ADrofCSwRAhD4AJwNGMYA1ygyIbJKVKgtee2Jj8CrfgCePYVi
OaH5ejDFQJ0R3jZ0zNrS2R4=
=Otu8
-----END PGP SIGNATURE-----

Re: [pyOpenSSL] time for a fork

[Dave C. <da...@cr...>]

On Wed May  2 23:00:59 2007, gl...@di... wrote:
> 
> On 08:49 am, da...@cr... wrote:
>> One key thing would be for a number of the key consumers of the
>> package to "bless" one particular fork as the successor. In case
>> you're wondering, while I'm happy if this happens to be mine, I'd 
>> be
>> equally happy to follow consensus on this.
> 
> Well, if I might flatter myself, Twisted is _the_ key consumer of 
> the package ;-) and I (we) would not hesitate for a moment to 
> "bless" your fork if you're willing to take up this responsibility.
> 
> 
Okay. I'll sort out some slightly more formalized structure to it in 
a week or so, unless anyone else wants to volunteer to maintain the 
fork.


> There are other parties we need to alert, though, especially the 
> various operating system packagers for pyOpenSSL.  Any of you 
> listening?  I suspect there will be some discussion of whether 
> there should be a name- change or not, but I don't know if anyone 
> will insist.  It would, of course, make things easier if someone 
> could figure out how to raise Martin!

Yup, packagers are Most Wanted. Martin is/was the maintainer of the 
Debian package, too, so assuming Martin's vanished, we'll need 
someone entirely new there.


>> That's the tricky bit - everyone comes to the SF website for
>> PyOpenSSL, and finds it dead.
> 
> Twisted and Divmod also have an obscene amount of Google juice, so 
> a few links on our websites should rapidly drive your site to the 
> top of the rankings for "pyopenssl".

Right - as I say, I'll "do" a more formal site for it next week, such 
as its own Trac installation. (It's currently in a little corner of 
my site).

Dave.
-- 
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

Re: [pyOpenSSL] time for a fork

[<gl...@di...>]

On 08:49 am, da...@cr... wrote:
>I've been collecting patchs and maintaining a fork, and I'm more than
>happy to continue. I've tried to be active in seeking out patches and
>applying them. (I'll look at the bugs you quote there, and see if I
>can sort them out).

I noticed the last few responses you sent to this list, and I hoped =

you'd say something like this.
>One key thing would be for a number of the key consumers of the
>package to "bless" one particular fork as the successor. In case
>you're wondering, while I'm happy if this happens to be mine, I'd be
>equally happy to follow consensus on this.

Well, if I might flatter myself, Twisted is _the_ key consumer of the =

package ;-) and I (we) would not hesitate for a moment to "bless" your =

fork if you're willing to take up this responsibility.

There are other parties we need to alert, though, especially the various =

operating system packagers for pyOpenSSL.  Any of you listening?  I =

suspect there will be some discussion of whether there should be a name- =

change or not, but I don't know if anyone will insist.  It would, of =

course, make things easier if someone could figure out how to raise =

Martin!
>That's the tricky bit - everyone comes to the SF website for
>PyOpenSSL, and finds it dead.

Twisted and Divmod also have an obscene amount of Google juice, so a few =

links on our websites should rapidly drive your site to the top of the =

rankings for "pyopenssl".
>This is why I feel the first step would be to enumerate the forks out
>there, choose one, and direct people consistently to there.

I don't hear anyone else volunteering besides you :).

Re: [pyOpenSSL] time for a fork

[Dave C. <da...@cr...>]

On Wed May  2 05:33:39 2007, gl...@di... wrote:
> Personally I've discovered, investigated, worked around, and 
> forgotten at least half a dozen pyopenssl bugs because it seems 
> like there's no activity here.  I really should have reported them, 
> and I suspect that there are others in the same boat, not reporting 
> bugs because they assume there's no one around to integrate patches 
> and do a release.
> 
> 
I've been collecting patchs and maintaining a fork, and I'm more than 
happy to continue. I've tried to be active in seeking out patches and 
applying them. (I'll look at the bugs you quote there, and see if I 
can sort them out).


> I have seen a few messages on this list in the last few years 
> indicating that folks are starting to maintain some forks.  Given 
> that the maintainer seems to have moved on completely, if anyone is 
> listening to this list, I think it's time for someone to take over 
> maintenance, start a new website, and continue development / 
> releases of pyopenssl somewhere else.
> 
> 
I'm also happy to increase my activity in this area. I'd be happy to 
share the load on this. I've never done much with Python modules 
before, so help's always welcome.

One key thing would be for a number of the key consumers of the 
package to "bless" one particular fork as the successor. In case 
you're wondering, while I'm happy if this happens to be mine, I'd be 
equally happy to follow consensus on this.


> Hopefully someone involved in such an effort could eventually get 
> someone else to hand over the keys to the sf.net project so that 
> the website here can be updated, but the important thing is to 
> start pointing the community somewhere else for recent versions of 
> pyopenssl.

That's the tricky bit - everyone comes to the SF website for 
PyOpenSSL, and finds it dead.

This is why I feel the first step would be to enumerate the forks out 
there, choose one, and direct people consistently to there.

Dave.
-- 
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

[pyOpenSSL] time for a fork

[<gl...@di...>]

pyopenssl.sf.net has been moribund for almost 3 years now.

Tickets like this:

    http://twistedmatrix.com/trac/ticket/2293

and this:

    http://allmydata.org/trac/tahoe/ticket/11

are starting to crop up on various projects which use pyopenssl.

Personally I've discovered, investigated, worked around, and forgotten =

at least half a dozen pyopenssl bugs because it seems like there's no =

activity here.  I really should have reported them, and I suspect that =

there are others in the same boat, not reporting bugs because they =

assume there's no one around to integrate patches and do a release.

I have seen a few messages on this list in the last few years indicating =

that folks are starting to maintain some forks.  Given that the =

maintainer seems to have moved on completely, if anyone is listening to =

this list, I think it's time for someone to take over maintenance, start =

a new website, and continue development / releases of pyopenssl =

somewhere else.

Hopefully someone involved in such an effort could eventually get =

someone else to hand over the keys to the sf.net project so that the =

website here can be updated, but the important thing is to start =

pointing the community somewhere else for recent versions of pyopenssl.

[pyOpenSSL] Cert Generation Issues

[<sco...@ra...>]

Hi there.  First off, great package.  It has saved me a bunch of work.  
  However, I'm automating some CSR generation and the CSR's i'm  
sending off aren't being liked by Thawte.  I cannot really tell what  
the problem is, because when i parse the CSR with openssl it reads it  
fine.  Not sure what the problem is, or if the output needs to be  
reformatted in any way.  I've written code very similar to the  
example.  Any clues?

Thanks in advance!
Scott Simpson


Confidentiality Notice: This e-mail message (including any attached or
embedded documents) is intended for the exclusive and confidential use of the
individual or entity to which this message is addressed, and unless otherwise
expressly indicated, is confidential and privileged information of Rackspace
Managed Hosting. Any dissemination, distribution or copying of the enclosed
material is prohibited. If you receive this transmission in error, please
notify us immediately by e-mail at ab...@ra..., and delete the
original message. Your cooperation is appreciated.

Re: [pyOpenSSL] Project status

[Dave C. <da...@cr...>]

On Tue Mar  6 10:54:17 2007, Thomas Damgaard wrote:
> Hi
> 
> On the website: http://pyopenssl.sourceforge.net/ I see that the 
> most
> recent release was in 2004. Is this project discontinued?

The original maintainer seems to have moved on to other things.

I've been keeping a fork up to date with various patches at 
http://trac.dave.cridland.net/ if that helps, but it's probably worth 
noting that the 2004 release still works fine if you prefer something 
that's seen more deployment.

Dave.
-- 
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

[pyOpenSSL] Project status

[Thomas D. <tho...@gm...>]

Hi

On the website: http://pyopenssl.sourceforge.net/ I see that the most
recent release was in 2004. Is this project discontinued?

-- 
Med venlig hilsen/Kind regards
Thomas Damgaard Nielsen
http://thomasdamgaard.dk

[pyOpenSSL] Example for crypto.X509Extension

[Christopher K. <ku...@rv...>]

Hi,

I'm currently fiddling around with extending certificates with a custom X509
extension (identified by OID), but can't seem to get it working. When using an
OID for the first method argument to crypto.X509Extension, the error message is
"ValueError: Unknown extension name". Using a known extension name or OID (from
objects.h), the error message changes to "ValueError: Unknown extension".

>From looking into the source code to src/crypto/x509ext.c, it seems that the
second error message is triggered by the extension not having any methods (?).

Could somebody point me to more literature regarding X509 extensions in
pyOpenSSL and/or provide some working examples for this?

Thanks a lot,

--ck

Re: [pyOpenSSL] CRL patch

[Dave C. <da...@cr...>]

On Wed Jan  3 14:24:25 2007, Arnaud Desmons wrote:
> Right, I'm not really interested in merging my code with
> pyopenssl-extended as far as I don't need pyopenssl-extended... :-)
> 
> 
My attitude is really that some people have needed the -extended 
stuff, so it might as well go in as long as it does no harm. Better 
to try to unify these forks.


> Anyway, I am not sure that users need both fonctionnalities at the
> same time...

True, but there's nowhere else to put it, unless we start fragmenting 
the pyOpenSSL modules further. I'm not sure that gains much other 
than complexity.

Dave.
-- 
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

Re: [pyOpenSSL] CRL patch

[Arnaud D. <arn...@fr...>]

On Wed, Jan 03, 2007 at 01:55:19PM +0000, Dave Cridland wrote:
> On Thu Dec 28 21:26:03 2006, Arnaud Desmons wrote:
> >Maybe I will do a separate sign(cacert, cakey) fonction instead of
> >doing everything in dump_crl???
> 
> I get quite a few failed hunks on this, applying to my tree - I'll 
> take a look at this in more detail later, but would you like to have 
> a look?

[...]

> It might have some CRL capability already - I've not examined what 
> the pyopenssl-extended patches do, yet. So far, where there's been 
> clashes between that and your code, I've preferred yours, but you 
> might not want to reimplement everything. :-)

Right, I'm not really interested in merging my code with
pyopenssl-extended as far as I don't need pyopenssl-extended... :-)

Anyway, I am not sure that users need both fonctionnalities at the
same time...

-- 
Arnaud

Re: [pyOpenSSL] CRL patch

[Dave C. <da...@cr...>]

On Thu Dec 28 21:26:03 2006, Arnaud Desmons wrote:
> Maybe I will do a separate sign(cacert, cakey) fonction instead of
> doing everything in dump_crl=E2=80=A6

I get quite a few failed hunks on this, applying to my tree - I'll=20
take a look at this in more detail later, but would you like to have=20
a look?

My tree's in SVN at:=20
http://svn.dave.cridland.net/svn/projects/pyopenssl/dwd/

It might have some CRL capability already - I've not examined what=20
the pyopenssl-extended patches do, yet. So far, where there's been=20
clashes between that and your code, I've preferred yours, but you=20
might not want to reimplement everything. :-)

Dave.
--=20
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

[pyOpenSSL] CRL patch

[Arnaud D. <arn...@fr...>]

Hi,

I wrote a second patch for pyOpenSSL 0.6 (to be applied just after my
pkcs12 one). This allow generation of CRL like this :

crl =3D crypto.CRL()
crl.make_revoked("100928084218Z", "1")
crl.make_revoked("100928084218Z", "2")
print crypto.dump_crl(crl, cacert, capkey)

Maybe I will do a separate sign(cacert, cakey) fonction instead of
doing everything in dump_crl=E2=80=A6

This is mainly for web (not tested for others purposes)

This patch is available here :
http://arnaud.desmons.free.fr/pyOpenSSL-0.6-crl.patch

Let me know what you guys think about it.
Thanks.

--=20
Arnaud

Re: [pyOpenSSL] PKCS12 patch

[Dave C. <da...@cr...>]

On Wed Nov 29 22:04:01 2006, Arnaud Desmons wrote:
> This patch is available here :
> http://arnaud.desmons.free.fr/pyOpenSSL-0.6-pkcs12.patch

Since this (and other) patches seem to be going unnoticed, I've put 
this into a fork I've made on http://trac.dave.cridland.net/

I've got the source in Subversion, at 
http://svn.dave.cridland.net/svn/projects/pyopenssl/dwd/

I've also found another fork since, 
http://www.keyphrene.com/products/pyOpenSSL-extended/index.php?lng=en 
- this seems to be somewhat dormant too, so I've integrated this as 
well. (It has a dump_pkcs12 as well, but comparing the two, I 
preferred Arnaud's, so that went in instead.)

I'll happily accept more patches.

Dave.
-- 
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

Re: [pyOpenSSL] PKCS12 patch

[Arnaud D. <arn...@fr...>]

On Wed, Nov 29, 2006 at 10:28:20PM +0000, Dave Cridland wrote:
> On Wed Nov 29 22:04:01 2006, Arnaud Desmons wrote:
> >p12 = crypto.PKCS12()
> >p12.set_privatekey(pkey)
> >p12.set_certificate(cert)
> >open("test.p12", "w").write(crypto.dump_pkcs12(p12, 
> >"my_passphrase"))
> >
> >
> >This patch is available here :
> >http://arnaud.desmons.free.fr/pyOpenSSL-0.6-pkcs12.patch
> 
> I'll take a look - PKCS12 is the stuff used in S/MIME, isn't it?

No. This is a file format used to import your certificate/private key
in your browser for website authentication for instance.

-- 
Arnaud

Re: [pyOpenSSL] PKCS12 patch

[Dave C. <da...@cr...>]

On Wed Nov 29 22:04:01 2006, Arnaud Desmons wrote:
> p12 = crypto.PKCS12()
> p12.set_privatekey(pkey)
> p12.set_certificate(cert)
> open("test.p12", "w").write(crypto.dump_pkcs12(p12, 
> "my_passphrase"))
> 
> 
> This patch is available here :
> http://arnaud.desmons.free.fr/pyOpenSSL-0.6-pkcs12.patch

I'll take a look - PKCS12 is the stuff used in S/MIME, isn't it?

Dave.
-- 
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

[pyOpenSSL] PKCS12 patch

[Arnaud D. <arn...@fr...>]

Hi,

I wrote a patch for pyOpenSSL 0.6 to allow generation of PKCS12 files
like this :

p12 = crypto.PKCS12()
p12.set_privatekey(pkey)
p12.set_certificate(cert)
open("test.p12", "w").write(crypto.dump_pkcs12(p12, "my_passphrase"))


This patch is available here :
http://arnaud.desmons.free.fr/pyOpenSSL-0.6-pkcs12.patch

Let me know what you guys think about it.

Regards,

-- 
Arnaud

Re: [pyOpenSSL] PyOpenSSL patches and stuff

[Dave C. <da...@cr...>]

On Thu Nov 23 16:13:45 2006, Dave Cridland wrote:
> Given there's some interest, I'll tidy up the patch and mail it to 
> the list, and redistribute a complete Python package from my 
> personal dev site.
> 
> 
Done the latter, it's at 
http://trac.dave.cridland.net/cgi-bin/trac/cgi/downloads as a 
standard Python source package. It's probably excrutiatingly hacky, 
and I do apologise.

Dave.
-- 
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

Re: [pyOpenSSL] PyOpenSSL patches and stuff

[Dave C. <da...@cr...>]

On Thu Nov 23 14:29:48 2006, Junglecow wrote:
> On Wed, Nov 22, 2006 at 08:24:01PM +0000, Dave Cridland wrote:
> 
> > My little email client, Polymer, has been using PyOpenSSL for a > 
> while. I've been missing certain functionality, so I've patched it 
> in > as best I can. I'd like to guage interest in putting the patch 
> into > the main distribution.
> 
> Hello Dave,
> 
> Your patch seems very useful to me. I am currently adding PyOpenSSL 
> to
> Gajim, and I have missed some functionality as well, especially 
> getting
> the current cipher name.
> 
> 
Gajim, being XMPP, will benefit from the get_comp_name method added - 
I did that as a relatively easy way for my mail client to detirmine 
whether or not to use COMPRESS=DEFLATE support (assuming it's 
available). XMPP also has application level stream compression 
(albeit it's a little weirdly specified), so that might be handy too.

Given there's some interest, I'll tidy up the patch and mail it to 
the list, and redistribute a complete Python package from my personal 
dev site.

Is there anything else people want added while I fiddle?

Dave.
-- 
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

Re: [pyOpenSSL] PyOpenSSL patches and stuff

[Junglecow <pyo...@9o...>]

On Wed, Nov 22, 2006 at 08:24:01PM +0000, Dave Cridland wrote:

> My little email client, Polymer, has been using PyOpenSSL for a 
> while. I've been missing certain functionality, so I've patched it in 
> as best I can. I'd like to guage interest in putting the patch into 
> the main distribution.

Hello Dave,

Your patch seems very useful to me. I am currently adding PyOpenSSL to
Gajim, and I have missed some functionality as well, especially getting
the current cipher name.

By the way, if anyone is interested in that OpenSSL.SSL.Error 'first num
too large' problem: I am now treating it like a WantRead/WantWrite
exception, i.e. basically just ignoring it, which seems to work fine.
However, I don't feel good about ignoring an error which I don't
understand and for which I can find no documentation.

[pyOpenSSL] PyOpenSSL patches and stuff

[Dave C. <da...@cr...>]

Hiya folks,

My little email client, Polymer, has been using PyOpenSSL for a 
while. I've been missing certain functionality, so I've patched it in 
as best I can. I'd like to guage interest in putting the patch into 
the main distribution.

Basically:

1) Add a get_cipher_name method, returns current cipher name or None 
(if there's no cipher).

2) Add a get_comp_name method, returns current compression algorithm 
name or None.

3) Added support for Python file protocol objects. Basically, if you 
pass in something that has a read and write method, then PyOpenSSL 
will use them. Sockets don't have these methods, so you still get the 
efficient C-path direct to the socket if you pass in a socket, but it 
allows you to insert Python code after the TLS layer. I find this 
useful to capture statistics, but it could be useful for other things.

These three could, of course, be three patches, or they could be one. 
I don't much care either way. :-)

Dave.
-- 
Dave Cridland - mailto:da...@cr... - xmpp:dw...@ja...
  - acap://acap.dave.cridland.net/byowner/user/dwd/bookmarks/
  - http://dave.cridland.net/
Infotrope Polymer - ACAP, IMAP, ESMTP, and Lemonade

[pyOpenSSL] [REPOST] Traceback: first num too large

[Junglecow <pyo...@9o...>]

Hi,

Please forgive me if you receive this mail twice. Even though it was
echoed back to me with all the right headers, it hasn't shown up
on SF's mail archives in 3 days. Furthermore, I received a subscription
confirmation one day *later*, adding to my suspicion that it got lost in
limboland.

============================================================================

Hi,

I am new to pyOpenSSL and OpenSSL. I'm currently trying to make Gajim run on
pyOpenSSL. This seems to be working fine, but when joining a room (any room), I
get this OpenSSL error:

Traceback (most recent call last):
  File "/tmp/gajim-svn/src/common/xmpp/transports_nb.py", line 244, in _do_receive
    received = self._recv(RECV_BUFSIZE)
  File "/tmp/gajim-svn/src/common/xmpp/transports_nb.py", line 57, in recv
    if flags is None: return self.sock.recv(bufsize)
OpenSSL.SSL.Error: [('asn1 encoding routines', 'a2d_ASN1_OBJECT', 'first num too large'), ('asn1 encoding routines', 'a2d_ASN1_OBJECT', 'first num too large')]

What is this error, what is causing it, and what can I do about it? Where
should I look for more information? Google is being rather unhelpful, and
searching this mailing list, python-crypto and the OpenSSL lists didn't turn up
anything useful either.

I am using python-pyopenssl 0.6-2.3 with openssl 0.9.8c-3 and python 2.4.3-8
from Debian testing.

If anyone wants to get their hands on some code to try, get Gajim svn using:
	svn checkout svn://svn.gajim.org/gajim/trunk gajim
and apply my preliminary patch at:
<url:http://trac.gajim.org/attachment/ticket/2499/gajim-pyopenssl-with-bugs-1.patch?format=raw>

If you just want to look, prettified patch is as:
<url:http://trac.gajim.org/attachment/ticket/2499/gajim-pyopenssl-with-bugs-1.patch>

For more information, see ticket at <url:http://trac.gajim.org/ticket/2499>

Please advise.
-J

[pyOpenSSL] Traceback: first num too large

[Junglecow <pyo...@9o...>]

Hi,

I am new to pyOpenSSL and OpenSSL. I'm currently trying to make Gajim run on
pyOpenSSL. This seems to be working fine, but when joining a room (any room), I
get this OpenSSL error:

Traceback (most recent call last):
  File "/tmp/gajim-svn/src/common/xmpp/transports_nb.py", line 244, in _do_receive
    received = self._recv(RECV_BUFSIZE)
  File "/tmp/gajim-svn/src/common/xmpp/transports_nb.py", line 57, in recv
    if flags is None: return self.sock.recv(bufsize)
OpenSSL.SSL.Error: [('asn1 encoding routines', 'a2d_ASN1_OBJECT', 'first num too large'), ('asn1 encoding routines', 'a2d_ASN1_OBJECT', 'first num too large')]

What is this error, what is causing it, and what can I do about it? Where
should I look for more information? Google is being rather unhelpful, and
searching this mailing list, python-crypto and the OpenSSL lists didn't turn up
anything useful either.

I am using python-pyopenssl 0.6-2.3 with openssl 0.9.8c-3 and python 2.4.3-8
from Debian testing.

If anyone wants to get their hands on some code to try, get Gajim svn using:
	svn checkout svn://svn.gajim.org/gajim/trunk gajim
and apply my preliminary patch at:
<url:http://trac.gajim.org/attachment/ticket/2499/gajim-pyopenssl-with-bugs-1.patch?format=raw>

If you just want to look, prettified patch is as:
<url:http://trac.gajim.org/attachment/ticket/2499/gajim-pyopenssl-with-bugs-1.patch>

For more information, see ticket at <url:http://trac.gajim.org/ticket/2499>

Please advise.
-J