Re: [gentoo-alt] Re: 'Continuous Integration' for Gentoo Prefix?

[Fabian Groffen]

Hi Sam,

Thanks!!!!

I'm currently assembling my own builds (work in progress) and am
planning to make some log analysis to point out what's wrong.  Perhaps
we can integrate the CI runs with the bootstraps on my "hardware".

I'm planning on releasing details about buildlogs, distfiles used
(possibly allow bootstraps to use it, so we work around the missing
distfiles problem), and have some status overview.  I considered email,
but I think it shouldn't be to spammy if sent to a list.  Perhaps we can
create a separate list for it though.

Fabian

On 17-01-2019 00:59:09 +1100, Sam Pfeiffer wrote:
> Hello,
>
> Given it's the first time I touch Fedora it took a bit of time to get it working
> (also I had some DNS issues where I couldn't use dnf).
>
> But it's done.
>
> You can check out the jobs at:
>
> For amd64 (Job called bootstrap_on_fedora_rap_off, [1]Dockerfile of the
> job): [2]https://dev.azure.com/12719821/12719821/_build/results?buildId=451
>
> For x86 (Job called bootstrap_on_fedora_rap_off, [3]Dockerfile of the
> job): [4]https://dev.azure.com/12719821/12719821/_build/results?buildId=450
>
> In 5-6h they should be finished.
>
> In general all the builds are
> here: [5]https://dev.azure.com/12719821/12719821/_build?definitionId=2&_a=summary
>
> Given this seems to be running quite nicely. Should I add automated emails when
> builds fail?
>
> I can do it either adding a script that does it or via the CI interface (I
> think). I won't always be able to check it out quick and make a bug report, so
> maybe someone else would like to receive the emails about the failures.
>
> On Wed, Jan 16, 2019 at 11:28 PM Michael Haubenwallner <[6]haubi@g.o>
> wrote:
>
> > On 1/16/19 2:23 AM, Sam Pfeiffer wrote:
> > > Hello Michael,
> > >
> > > Yeah, just tell me which base distro do you want and I'll add a nightly job
> > with that one. I just need to find a Docker image for it.
>
> > It doesn't really matter, just not Ubuntu (or anything else that does
> > 'multiarch').
> > I would suggest Fedora though...
>
> > >
> > > Will the final Gentoo Prefix, once bootstrapped, be any different from the
> > current one I'm bootstrapping?
>
> > Yes: It does not contain sys-libs/glibc and sys-kernel/linux-headers.
>
> > > (To know if I should also publish automated releases of the bootstrapped
> > Gentoo Prefix).
>
> > Haven't recognized that you do 'publish automated releases', nice!
>
> > As Prefix/Guest is stronger bound to the host OS compared to Prefix/RAP,
> > binary releases don't feel that useful here - at least to myself.
>
> > Instead, besides x86_64, also x86 (32bit) would be nice - simply using
> > 'linux32':
> > $ PREFIX_DISABLE_RAP=yes linux32 ./bootstrap-prefix.sh /target/prefix
> > noninteractive
>
> > Thanks a lot!
> > /haubi/
>
> > >
> > >
> > > On Wed, Jan 16, 2019, 04:38 Michael Haubenwallner <[7]haubi@g.o
> > <mailto:[8]haubi@g.o> wrote:
> > >
> > >     Hi Sammy,
> > >
> > >     because of Ubuntu inside these build slaves I do understand you
> > currently
> > >     perform Prefix RAP bootstraps only - as this is the default anyway.
> > >
> > >     Do you see a chance to perform Prefix Guest bootstraps as well,
> > >     even if that would require something other distro than Ubuntu?
> > >
> > >     Otherwise, the only difference is to set the PREFIX_DISABLE_RAP=yes
> > >     environment variable when executing bootstrap-prefix.sh.
> > >
> > >     Thanks!
> > >     /haubi/
> > >
> > >
>
> --
>
> Sammy Pfeiffer
> PhD Candidate at The Magic Lab within UTS.
>
>
>
>  References:
>    1. https://github.com/awesomebytes/gentoo_prefix_ci/blob/master/initial_bootstrap/Dockerfile.fedora
>    2. https://dev.azure.com/12719821/12719821/_build/results?buildId=451
>    3. https://github.com/awesomebytes/gentoo_prefix_ci_32b/blob/master/initial_bootstrap/Dockerfile.fedora
>    4. https://dev.azure.com/12719821/12719821/_build/results?buildId=450
>    5. https://dev.azure.com/12719821/12719821/_build?definitionId=2&amp;_a=summary
>    6. mailto:haubi@g.o
>    7. mailto:haubi@g.o
>    8. mailto:haubi@g.o
>
> read_char: errno==EILSEQ; invalid byte sequence for UTF-8:
-- 
Fabian Groffen
Gentoo on a different level