Re: [ptxdist] Antwort: Re: Antwort: Re: [PATCH] openvpn: version bump 2.3.11 -> 2.3.13

[Michael Olbrich]

Hi,

On Fri, Dec 02, 2016 at 01:28:51PM +0100, Tobias Schmidl wrote:
> -"ptxdist"  schrieb: -
> > FWIW: There are some helpful hints on e-mail clients in the Linux
> > kernel documentation:
> 
> > https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/Documentation/email-clients.txt
> 
> > The patch we are talking about was sent with Lotus according to mail
> > headers. Lotus was the default in my company as well for years and it
> > really sucks at e-mail. So what I did was pushing my stuff to my
> > personal public git clone [1], pull it at home and send it with my
> > personal mail account. ¯\_()_/¯

> Tell me about it... But then I tried three different MTAs (Evolution,
> LN and Thunderbird), and all of them mangle the inline patches, so

You used the MTAs with a SMTP server, right? You can configure a SMTP
server for 'git send-email' and send the patches directly. Check the man
page for git-send-email. There is an example at the end.

> I've included the (very simple) patch as an attachment. 

Well, applying the patch works. Try to send it with type text/
next time (if you can). Then I should be able to actually comment on it.

> One note: The OpenVPN Download page[1] gives me sometimes a 404 for
> the source.tar.xz, even though the source itself[2] *can* be reached by
> curl. Maybe some weird caching issue.

Works for me.

Michael


-- 
Pengutronix e.K.   | |
Industrial Linux Solutions | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0|
Amtsgericht Hildesheim, HRA 2686   | Fax:   +49-5121-206917- |

___
ptxdist mailing list
ptxdist@pengutronix.de

Re: [ptxdist] ptxdist compile redirects stderr to stdout

[Michael Olbrich]

Hi,

On Fri, Dec 02, 2016 at 12:25:01PM -0500, Jon Ringle wrote:
> On Thu, 1 Dec 2016, Michael Olbrich wrote:
> > On Wed, Nov 30, 2016 at 03:46:14AM -0500, Jon Ringle wrote:
> > > I found that this is being caused by commit
> > > 30b9267e35eea1c2edb4da0231a428bfa25b6766
> > > 
> > > Is there a way that the stderr to stdout redirect could be implemented
> > > conditionally so that I can still have stderr segregated if I want to?
> > 
> > Yes. It's really only necessary if output synchronization is enabled. So,
> > we could set some variable in setup_parallel() to '1' or '2' depending on
> > PTXDIST_OUTPUT_SYNC and then use it instead of '2>&1'.
> 
> I did some experimenting with this, and when I use a variable to hold 
> "2>&1", it works fine for the *) case, but for the python*) case where the 
> ptxd_eval is surrounded by ( ), bash chokes.
> 
> Here's what I was trying:
> ptxd_make_world_compile() {
> 
> output_redirect="2>&1";
> 
> ptxd_make_world_init &&
> 
> if [ -z "${pkg_build_dir}" ]; then
> # no build dir -> assume the package has nothing to build.
> return
> fi &&
> case "${pkg_conf_tool}" in
> python*)
> (
> ptxd_eval \
> cd "${pkg_build_dir}" '&&' \
> "${pkg_path}" \
> "${pkg_env}" \
> "${pkg_make_env}" \
> "${ptx_build_python}" \
> setup.py \
> "${pkg_make_opt}"
> ) ${output_redirect}
> ;;
> *)
> ptxd_eval \
> "${pkg_path}" \
> "${pkg_env}" \
> "${pkg_make_env}" \
> "${MAKE}" -C "${pkg_build_dir}" \
> "${pkg_make_opt}" \
> "${pkg_make_par}" ${output_redirect}
> ;;
> esac
> }
> 
> $ ~/git/ptxdist/bin/ptxdist --force compile testprog >/dev/null
> /home/jringle-admin/git/ptxdist/scripts/lib/ptxd_make_world_compile.sh: 
> line 35: syntax error near unexpected token `${output_redirect}'
> /home/jringle-admin/git/ptxdist/scripts/lib/ptxd_make_world_compile.sh: 
> line 35: `  ) ${output_redirect}'
> ptxdist: error: failed to source lib: 
> /home/jringle-admin/git/ptxdist/scripts/lib/ptxd_make_world_compile.sh
> 
> 
> If I remove the variable redirect from the offending line and leave the 
> variable redirect in the *) case, bash is fine with it and it does what is 
> expected.

It would work if you move it inside the brackets. But I think it would be
better to just have one redirect after the 'esac'.

Michael

-- 
Pengutronix e.K.   | |
Industrial Linux Solutions | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0|
Amtsgericht Hildesheim, HRA 2686   | Fax:   +49-5121-206917- |

___
ptxdist mailing list
ptxdist@pengutronix.de

Re: [ptxdist] systemd 232

[Michael Olbrich]

Hi,

On Thu, Dec 01, 2016 at 05:00:19PM +0100, Clemens Gruber wrote:
> On Tue, Nov 29, 2016 at 09:46:14AM +0100, Michael Olbrich wrote:
> > On Mon, Nov 28, 2016 at 06:02:51PM +0100, Clemens Gruber wrote:
> > > did you already start working on updating systemd to version 232?
> > 
> > I've started with it. But new systemd versions require more testing than
> > most packages.
> 
> I just tried the ptxdist master version of systemd 232.
> 
> Looking good except for one problem:
> rc-once.service: Main process exited, code=killed, status=15/TERM
> rc-once.service: Unit entered failed state.
> rc-once.service: failed with result 'signal'.
> 
> Because I am not doing anything different in /etc/rc.once.d/*, I assume
> this has something to do with the update from 231 to 232.
> 
> Can you reproduce this?
> 
> For the record, in rc.once.d I have: ldconfig, machine-id and openssh.
> 
> Is there something in /lib/systemd/systemd-rc-once which could cause
> this on systemd 232?

I'm not aware of anything that would cause this. I just tried this
combination in my qemu test BSP and it works without problems.

Do you get a emergency shell? Anything else in the journal? How long does
it take to get the error? Maybe the service is killed by a timeout.

Michael

-- 
Pengutronix e.K.   | |
Industrial Linux Solutions | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0|
Amtsgericht Hildesheim, HRA 2686   | Fax:   +49-5121-206917- |

___
ptxdist mailing list
ptxdist@pengutronix.de