buildroot INSTALL_PREFIX and hardcoded paths

2018-06-14 Thread Konstantin Ryabitsev

Hello:

I'm trying to package a static build of gnupg22 so I don't have to copy
things manually to each CentOS-7 system where I need ECC crypto support.
I'm using the following to build gnupg-2.2.8 inside the RPM:

make -f build-aux/speedo.mk STATIC=1 CUSTOM_SWDB=1 \
   INSTALL_PREFIX=%{buildroot}/opt/gnupg22 this-native

RPM needs to "install" things into a buildroot, so in the above case, 
%{buildroot} is /buildroot/build/BUILDROOT/gnupg22-static-2.2.8-2.x86_64.

Of course, this means that gpg and other binaries expect to call
gpg-agent, dirmngr, and other auxiliary programs in
/buildroot/build/BUILDROOT/gnupg22-static-2.2.8-2.x86_64/opt/gnupg22/bin
instead of just /opt/gnupg22.

Is there a simple solution to this? I know it's not the intended purpose
of speedo.mk, but it's convenient to use it in my case, if only I could
teach it to distinguish between where things are during build and where
they will be after installation.

Currently, I work around this by setting agent-program and
dirmngr-program in gpg.conf.

Thanks for any suggestions!

Best,
Konstantin


signature.asc
Description: PGP signature
___
Gnupg-users mailing list
Gnupg-users@gnupg.org
http://lists.gnupg.org/mailman/listinfo/gnupg-users


Re: Silencing MDC Warning with gnupg 2.2.8.

2018-06-14 Thread Werner Koch
On Thu, 14 Jun 2018 13:56, ra...@inputplus.co.uk said:

> I see that --ignore-mdc-error downgrades the error to a warning allowing

Right, this is the suggest method to decrypt old mails.

> --no-mdc-warning is now a no-op and so doesn't work in concert with

Right, this is on purpose.  The warning is important to educate users.

> remove the MDC warning from stderr, allowing the rest of stderr's
> content to pass?  Downstream of this command is unhappy otherwise.

Why do you need stderr at all?  These are diagnositics for human
consumption.  And, as I said, we want that the warning show ups if the
override option is enabled.  

> gpg(1) still documents --force-mdc and --disable-mdc, saying they're now
> no-ops, but --no-mdc-warning is undocumented despite still being
> accepted and also a no-op.  This hampers investigating why

There is no reason to document a dummy option which never affected the
behaviour of gpg.  This is in contrast to the other mentioned NOPs.


Shalom-Salam,

   Werner

-- 
#  Please read:  Daniel Ellsberg - The Doomsday Machine  #
Die Gedanken sind frei.  Ausnahmen regelt ein Bundesgesetz.


pgpIY0MqKxlPm.pgp
Description: PGP signature
___
Gnupg-users mailing list
Gnupg-users@gnupg.org
http://lists.gnupg.org/mailman/listinfo/gnupg-users


Silencing MDC Warning with gnupg 2.2.8.

2018-06-14 Thread Ralph Corderoy
Hi,

With the arrival of gnupg 2.2.8-1 on Arch Linux this command applied to
members of an archive of files on read-only media fails because they old
enough not to have Modification Detection Codes.

gpg -q --batch --no-mdc-warning -d --passphrase-fd 0 foo.gpg

I see that --ignore-mdc-error downgrades the error to a warning allowing
the decrypted content to appear on stdout, but unfortunately for me
--no-mdc-warning is now a no-op and so doesn't work in concert with
--ignore-mdc-error to silence the warning.

It seems from skimming the source that my only option is to expect and
remove the MDC warning from stderr, allowing the rest of stderr's
content to pass?  Downstream of this command is unhappy otherwise.

gpg(1) still documents --force-mdc and --disable-mdc, saying they're now
no-ops, but --no-mdc-warning is undocumented despite still being
accepted and also a no-op.  This hampers investigating why
--no-mdc-warning isn't working.

BTW, 2.2.8's `NEWS' has `--no-mdc-warn', but the option ends in
`warning' and so my searches didn't find the news item.

-- 
Cheers, Ralph.
https://plus.google.com/+RalphCorderoy

___
Gnupg-users mailing list
Gnupg-users@gnupg.org
http://lists.gnupg.org/mailman/listinfo/gnupg-users


Re: Is something not right with my ~/.gnupg/pubring.kbx

2018-06-14 Thread Ajax
On Fri, Jun 8, 2018 at 8:21 PM, Ajax  wrote:

> Most gpg commands give me something like this:
>
> $ gpg -Kv
> gpg: using classic trust model
> gpg: keydb_search failed: Invalid value
> gpg: Oops: keyid_from_fingerprint: no pubkey
> ~/.gnupg/pubring.kbx
>
> Followed by what appears to me to be normal output.
>
> I also see:
>
> $ kbxutil --stats ~/.gnupg/pubring.kbx
> Total number of blobs:  600
>header:1
> empty:0
>   openpgp:  599
>  x509:0
>   non flagged:  599
>secret flagged:0
> ephemeral flagged:0
>
> What should I do to clear the Invalid value and no pubkey?
>
> I only saw this problem after upgrading from gpg 1 to gpg2 and using the
> given script to migrate from pubring.gpg to pubring.kbx.
>
> These outputs seem to throw a monkey wrench into some scripts such as, for
> example, with pass generate -i
>
> This is gpg 2.1.18 atop Debian stretch and Linux  4.9.0-6-amd64,
>
> Thank you.
>

Are there no sugestions on how I should try to eliminate the indication of
no pub key althoug gpg -k shows my public keys?

Thanks.
___
Gnupg-users mailing list
Gnupg-users@gnupg.org
http://lists.gnupg.org/mailman/listinfo/gnupg-users


Re: Problem refreshing keys

2018-06-14 Thread Jerry
On Wed, 13 Jun 2018 23:22:19 -0400, Phil Pennock stated:

>On 2018-06-13 at 09:52 -0400, Jerry wrote:
>> On Wed, 13 Jun 2018 15:25:04 +0200, Werner Koch stated:  
>> >The common problem on Windows: You can't use ' to quote; we Unix folks
>> >always forget about that.  Use  
>
>Bah, I just didn't know.  :D  I suspected though, which is why I
>mentioned typing interactively as a fallback.
>
>> gpg-connect-agent --dirmngr "KEYSERVER --hosttable" /bye
>> S # hosttable (idx, ipv6, ipv4, dead, name, time):
>> S #   0   hkps.pool.sks-keyservers.net (216.66.15.2)
>> OK
>> 
>> Is that what it should be reporting?  
>
>What version is it?  Is there a newer version available?
>
>  gpg-connect-agent --dirmngr "GETINFO version" /bye
>
>There have been a bunch of fixes for various DNS issues with dirmngr, I
>would expect to see something showing that it's a pool.
>
>You're talking to zimmermann.mayfirst.org, which works fine; I just
>overrode DNS for the pool and made sure that
>hkps.pool.sks-keyservers.net only reached that IP (/etc/hosts override)
>and I was able to retrieve a key fine, after which:
>
>> KEYSERVER --hosttable  
>S # hosttable (idx, ipv6, ipv4, dead, name, time):
>S #   0   hkps.pool.sks-keyservers.net
>S #   .   hkps.pool.sks-keyservers.net
>S #   .   --> 1*
>S #   1   4   216.66.15.2 (hkps.pool.sks-keyservers.net)
>OK
>
>I suspect that you have an old dirmngr and the problems are fixed with a
>newer release of gpg4win.
>
>-Phil

gpg-connect-agent --dirmngr "GETINFO version" /bye
gpg-connect-agent: no running Dirmngr - starting 'C:\Program Files 
(x86)\Gpg4win\..\GnuPG\bin\dirmngr.exe'
gpg-connect-agent: waiting for the dirmngr to come up ... (5s)
gpg-connect-agent: waiting for the dirmngr to come up ... (4s)
gpg-connect-agent: connection to the dirmngr established
D 2.2.7
OK

I have Gpg4win Version 3.1.1 (2018-05-03) installed. That is supposed to be
the latest version.

-- 
Jerry

___
Gnupg-users mailing list
Gnupg-users@gnupg.org
http://lists.gnupg.org/mailman/listinfo/gnupg-users