Re: Release scheduling

2014-12-10 Thread Samir Nassar
On Wednesday, 2014-12-10 19:01:25 Kristian Fiskerstrand wrote: > See https://bugs.archlinux.org/task/42739?opened=6005&status%5B0%5D= Well lo and behold. When I rebuilt the package with GnuTLS as a dependency and killed the running dirmngr process HKPS works. I understand that the GnuPG package

Re: Release scheduling

2014-12-10 Thread Werner Koch
On Wed, 10 Dec 2014 18:10, kristian.fiskerstr...@sumptuouscapital.com said: > The SNI issue last discussed in [0] springs to mind. But I still > experience this on gpg (GnuPG) 2.1.1-beta67 I have not yet tracked this down. For easier debugging I added some more debug output. Forgot to push the

Re: Release scheduling

2014-12-10 Thread Kristian Fiskerstrand
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 On 12/10/2014 06:44 PM, Kristian Fiskerstrand wrote: > > ... > >> Individual keyserver should be no issue as long as you don't hit >> the SNI issue, are you sure gnupg is built with gnutls / hkps >> support? I know that at least earlier builds of

Re: Release scheduling

2014-12-10 Thread Samir Nassar
On Wednesday, 2014-12-10 12:10:28 Daniel Kahn Gillmor wrote: > > Can you provide more detail (or a link to a bug report) about the > > problem with hkps in 2.1.0 ? On upgrade to 2.1.0 looking up keys from a keyserver stopped working for me. I tracked down the following comment from Hugo Hinterber

Re: Release scheduling

2014-12-10 Thread Daniel Kahn Gillmor
On 12/10/2014 12:26 PM, Samir Nassar wrote: > On Wednesday, 2014-12-10 12:10:28 Daniel Kahn Gillmor wrote: >>> Can you provide more detail (or a link to a bug report) about the >>> problem with hkps in 2.1.0 ? > > On upgrade to 2.1.0 looking up keys from a keyserver stopped working for me. > I >

Re: Release scheduling

2014-12-10 Thread Kristian Fiskerstrand
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 ... > Individual keyserver should be no issue as long as you don't hit > the SNI issue, are you sure gnupg is built with gnutls / hkps > support? I know that at least earlier builds of arch did not > include gnutls support, which caused some headac

Re: Release scheduling

2014-12-10 Thread Kristian Fiskerstrand
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 On 12/10/2014 06:26 PM, Samir Nassar wrote: > On Wednesday, 2014-12-10 12:10:28 Daniel Kahn Gillmor wrote: >>> Can you provide more detail (or a link to a bug report) about >>> the problem with hkps in 2.1.0 ? > > On upgrade to 2.1.0 looking up keys

Re: Release scheduling

2014-12-10 Thread Kristian Fiskerstrand
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 On 12/10/2014 06:10 PM, Daniel Kahn Gillmor wrote: > On 12/10/2014 12:02 PM, Samir Nassar wrote: >> It is my understanding that 2.1.0 has a problem with hkps >> keyservers (such as the hkps SKS pool) and that this is only >> fixed in the betas for 2.

Re: Release scheduling

2014-12-10 Thread Daniel Kahn Gillmor
On 12/10/2014 12:02 PM, Samir Nassar wrote: > It is my understanding that 2.1.0 has a problem with hkps keyservers (such as > the hkps SKS pool) and that this is only fixed in the betas for 2.1.1. If > this > understanding is correct and 2.1.1 fixes the hkps issues, I'd vote to release > 2.1.1

Re: Release scheduling

2014-12-10 Thread Samir Nassar
On Wednesday, 2014-12-10 17:41:21 Werner Koch wrote: > Hi! > > it is now more than a month since the 2.1.0 release and I am asking > myself whether it is time to do another release. There are about 50 > commits including It is my understanding that 2.1.0 has a problem with hkps keyservers (such

Re: Release scheduling

2014-12-10 Thread Daniel Kahn Gillmor
On 12/10/2014 11:41 AM, Werner Koch wrote: > However, there are still open bugs and new bugs are also detected every > few days. I think it is better to do a 2.1.1 now instead of trying to > get all new bugs fixed - it would delay things into the next year. > > I plan to look into the learn card

Release scheduling

2014-12-10 Thread Werner Koch
Hi! it is now more than a month since the 2.1.0 release and I am asking myself whether it is time to do another release. There are about 50 commits including * gpg: Detect faulty use of --verify on detached signatures. * gpg: New import option "keep-ownertrust". * gpg: Fixed regression in -