polkit broken?

polkit broken?

[Steve Sakoman]

I find that polkit is failing after a pull this morning:

ERROR: Function failed: BUILDSPEC (see
/media/Work/yocto/tmp/work/armv7a-vfp-neon-poky-linux-gnueabi/polkit-0.104-r0/temp/log.do_package_write_rpm.25212
for further information)
ERROR: Logfile of failure stored in:
/media/Work/yocto/tmp/work/armv7a-vfp-neon-poky-linux-gnueabi/polkit-0.104-r0/temp/log.do_package_write_rpm.25212
Log data follows:
| NOTE: Creating RPM package for polkit-dbg
| NOTE: Creating RPM package for polkit
| NOTE: Creating RPM package for polkit
| NOTE: Creating RPM package for polkit-doc
| NOTE: Creating RPM package for polkit-dev
| NOTE: Creating RPM package for polkit-staticdev
| NOTE: Creating RPM package for polkit-locale-da
| DEBUG: SITE files ['endian-little', 'bit-32', 'arm-common',
'common-linux', 'common-glibc', 'arm-linux', 'arm-linux-gnueabi',
'common']
| ERROR: Function failed: BUILDSPEC (see
/media/Work/yocto/tmp/work/armv7a-vfp-neon-poky-linux-gnueabi/polkit-0.104-r0/temp/log.do_package_write_rpm.25212
for further information)
| error: line 24: Illegal char '-' in EVR: Requires: systemd >= v38-53-ga3f914b
| error: Package has no %description: polkit-0.104-r0.armv7a
| Building target platforms: armv7a-vfp-neon-poky-linux-gnueabi
NOTE: package polkit-0.104-r0: task do_package_write_rpm: Failed
ERROR: Task 1059
(/home/sakoman/source/yocto/poky/meta/recipes-extended/polkit/polkit_0.104.bb,
do_package_write_rpm) failed with exit code '1'

My image does include systemd from meta-openembedded so I suspect that
may be a factor.

For those who are building with sysvinti, do you see this issue?

Steve

Re: polkit broken?

[Richard Purdie]

On Fri, 2012-01-27 at 11:33 -0800, Steve Sakoman wrote:
> I find that polkit is failing after a pull this morning:
> 
> ERROR: Function failed: BUILDSPEC (see
> /media/Work/yocto/tmp/work/armv7a-vfp-neon-poky-linux-gnueabi/polkit-0.104-r0/temp/log.do_package_write_rpm.25212
> for further information)
> ERROR: Logfile of failure stored in:
> /media/Work/yocto/tmp/work/armv7a-vfp-neon-poky-linux-gnueabi/polkit-0.104-r0/temp/log.do_package_write_rpm.25212
> Log data follows:
> | NOTE: Creating RPM package for polkit-dbg
> | NOTE: Creating RPM package for polkit
> | NOTE: Creating RPM package for polkit
> | NOTE: Creating RPM package for polkit-doc
> | NOTE: Creating RPM package for polkit-dev
> | NOTE: Creating RPM package for polkit-staticdev
> | NOTE: Creating RPM package for polkit-locale-da
> | DEBUG: SITE files ['endian-little', 'bit-32', 'arm-common',
> 'common-linux', 'common-glibc', 'arm-linux', 'arm-linux-gnueabi',
> 'common']
> | ERROR: Function failed: BUILDSPEC (see
> /media/Work/yocto/tmp/work/armv7a-vfp-neon-poky-linux-gnueabi/polkit-0.104-r0/temp/log.do_package_write_rpm.25212
> for further information)
> | error: line 24: Illegal char '-' in EVR: Requires: systemd >= v38-53-ga3f914b
> | error: Package has no %description: polkit-0.104-r0.armv7a
> | Building target platforms: armv7a-vfp-neon-poky-linux-gnueabi
> NOTE: package polkit-0.104-r0: task do_package_write_rpm: Failed
> ERROR: Task 1059
> (/home/sakoman/source/yocto/poky/meta/recipes-extended/polkit/polkit_0.104.bb,
> do_package_write_rpm) failed with exit code '1'
> 
> My image does include systemd from meta-openembedded so I suspect that
> may be a factor.

I suspect so too.

> For those who are building with sysvinti, do you see this issue?

No, this looks like the version string for systemd is upsetting rpm
somehow, probably the inclusion of a "-". The rpm backend is supposed to
remap that character iirc...

Cheers,

Richard

Re: polkit broken?

[Steve Sakoman]

On Fri, Jan 27, 2012 at 3:47 PM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
>
>> My image does include systemd from meta-openembedded so I suspect that
>> may be a factor.
>
> I suspect so too.
>
>> For those who are building with sysvinti, do you see this issue?
>
> No, this looks like the version string for systemd is upsetting rpm
> somehow, probably the inclusion of a "-". The rpm backend is supposed to
> remap that character iirc...

I tested this on a second build machine and didn't encounter the error.

I then went back to the first build machine, did a 'bitbake -c
cleansstate avahi systemd' and then rebuilt the image.

Of course it built with no problems, so I'm not really sure what was
going on!  I'll try a clean build on another machine to see if it
re-occurs.  If not, it will be yet another addition to my "mysteries
of the universe" list :-)

Steve