So curious now, is this duplicable? do you havea working iso image?
did it install properly? it is functional ?

On Fri, Sep 22, 2017 at 5:41 PM, Marek Marczykowski-Górecki
<[email protected]> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
> On Fri, Sep 22, 2017 at 02:23:14PM -0700, Frédéric Pierret (Epitre) wrote:
>> I finally succeeded to build Qubes with Fedora 26 as dom0 and template.
>
> Yay!
>
>> Globally, it was a matter of some fixes with respect to GCC7 for xen and
>> other small minor corrections in qubes packages.
>> There is quite a lot small changes in codes between Fedora 25 and 26+ with
>> respect the installer packages (Pungi, Lorax, Pykickstart and Anaconda).
>
> One very important part of patches there is checking signatures of
> downloaded packages. Unfortunately, upstream maintainers don't care about
> it. Besides porting patches, it's a good idea to test it. See
> .travis.yml and conf/travis*.ks - there are configs intentionally
> including repositories with unsigned packages.
>
>> So to debug it was (still) long because I have not found a shortcut to
>> bypass the whole build of the iso (which is about ~20min on my servers each
>> time), neither a way to debug easily Anaconda.
>
> I don't have way around it. But for testing anaconda, it is useful to
> have:
> 1. PXE boot, like here: 
> https://unix.stackexchange.com/questions/186302/fedora-network-install-via-pxe-boot
> 2. kickstart file - look at /root/anaconda-ks.cfg after installation
> 3. wake on lan.
>
> This way, it's possible to automate the whole workflow, so you hit
> "build" and come back half an hour later to see result of the test.
>
> For trivial fixes, you can do that from tty2 (edit files in
> /usr/...). Make sure to kill crashed
> anaconda with SIGKILL, otherwise it will shutdown the system. And AFAIR
> you also need to kill X server before starting new one.
>
>> In other words, in order to not do all the stages of Pungi, it seems that
>> we have to comment several things it the Makefile. Notably, I rewrite all
>> the necessary code (patches) for Anaconda with respect to the previous
>> commits to have, like the others, a source package, a spec file and patches
>> (In fact, I have done it for Fedora 26 and 27). In my opinion, it is more
>> clear and consistent.
>>
>> For the Fedora 26 template (Qubes 3.2 and 4.x), one has to take into
>> account a very slight modification in the qubes dnf plugin due to newer
>> version (a temporarly fix
>> https://github.com/fepitre/qubes-core-agent-linux/commit/8fffaadb41b9fb5ef5231287483040b31e02d57b).
>>
>> I should push everything on github on devel branches in the next days, a
>> time to polish and debug a bit more. Stay tuned !
>>
>> Le dimanche 10 septembre 2017 01:43:33 UTC+2, Marek Marczykowski-Górecki a
>> écrit :
>> >
>> > -----BEGIN PGP SIGNED MESSAGE-----
>> > Hash: SHA256
>> >
>> > On Sun, Sep 10, 2017 at 12:31:03AM +0200, Frédéric Epitre wrote:
>> > > Hi,
>> > >
>> > > Just to give you some news about what I have done these days on Qubes:
>> >
>> > Thanks! Please cc qubes-devel ML for such things. I've added it here
>> > now.
>> >
>> > > - I succeed to build all the components (dom0 and vm) for the release
>> > > 4.0 with Fedora 26. For xen-4.8 I had to fix few lines in the patches of
>> > > xen-4.6 for GCC7 and also to add two patches of the version 4.8.2 for
>> > > the current 4.8.1.
>> >
>> > Oh, there is already 4.8.2. I'll check how feasible is updating to
>> > 4.8.2.
>> >
>> > > - I solved one more fallthrough in the core-admin and everything is
>> > good.
>> > >
>> > > - The last thing I'm currently stuck is the installer. I have to create
>> > > a fedora 26 branch and import the new version of anaconda and also new
>> > > version of pykickstart(>=1.33 for fedora 26 support) :
>> >
>> > This isn't needed for fc26 template, and it's too late to change dom0
>> > distribution version for 4.0. So, don't bother. For Qubes 4.1, there
>> > will be probably newer version at that time.
>> >
>> > > * I tried to build pykickstart but tests create errors (need to add more
>> > > than you first patch to skip all the errors). So skipping errors, it
>> > > allows to obtain the packages.
>> > >
>> > > * Then, PUNGI...........During the nosetests, it bothers me with
>> > > something like Pungi:INFO: <type 'exceptions.AttributeError'>
>> > > AttributeError("'F27_RepoData' object has no attribute 'gpgkey'",), or
>> > > other erros always related to "'F27_RepoData' object has no attribute
>> > > 'gpgkey'". I don't understand neither why it is talking about FC27 nor
>> > > gpgkey because this supplementary parameter added should not be
>> > > considered in the test right?
>> >
>> > Looks related to a patch adding package signature verification. Probably
>> > needs to be updated. But as said above - don't worry about it now.
>> >
>> > > I will debug in the next days with a new fresh branch for fedora 26 in
>> > > the installer.
>> > >
>> > > Best,
>> > >
>> > > Frédéric
>> > >
>> > >
>> >
>> > - --
>> > Best Regards,
>> > Marek Marczykowski-Górecki
>> > Invisible Things Lab
>> > A: Because it messes up the order in which people normally read text.
>> > Q: Why is top-posting such a bad thing?
>> > -----BEGIN PGP SIGNATURE-----
>> > Version: GnuPG v2
>> >
>> > iQEcBAEBCAAGBQJZtHyeAAoJENuP0xzK19csP64H/RHDPGvk2toK18ibmP9MZV5c
>> > vLYp6NtHSU6zZrIxqcc+6/yXWEykhYT2FbNx+elJeB+k5dRpJoeZ0e9CnZtFSV9X
>> > CsfN9j5IH59xm2mMUeXxWC4ObgHYVQDcxEtPWeOU4IYq02G45XBdYd4TVk/COtmR
>> > NFJV5lHpzGqeCiDY3O1Sq2VKHbU4DMbXOlz00x6PADAT3GTErC/JcCiEVyoXzN6b
>> > ItiG/qgP8QsqA6VkVDazjvyopRyuNsn3HkLMl/NK8iUlOwIdoDyGgaYT2zgeah38
>> > 3pnWcUAJ8LmV79JGH9i39NHI2Ww+px7hVAVTbT7eXQgdn3sta9CZ3ZKAae2iHbQ=
>> > =3RbI
>> > -----END PGP SIGNATURE-----
>> >
>>
>
>
> - --
> Best Regards,
> Marek Marczykowski-Górecki
> Invisible Things Lab
> A: Because it messes up the order in which people normally read text.
> Q: Why is top-posting such a bad thing?
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v2
>
> iQEcBAEBCAAGBQJZxYOYAAoJENuP0xzK19cs1D4IAIQRz3s3DNlCjsNlhcliMD1i
> bVt0SOIFpG6QLMIzADHq5m7vcEzvBUdfn/ULEdyrulKgPZkj+Tx+ZbHmMP4+SLqi
> DypdXyvjBOJLAjMHfhTgVQ4D36ywN+ZYjBtwDaJAb2oYHMBmO1+fYPPM3ZJpSI23
> Q3xYDFv0DUqQBuNL39uHN+VJdnWZxbtay+F4DfgptlajiLmsax0tbwcqrx6qGtfO
> 1s3uxTvASv8MojE5Lpw3ZpFFV3kuLzKx6C8862KK1/vsXqsfCtLrKhuYp22X+d/8
> +gGixjaIRUhCL/EZiDE+Tot7A+Eq5Rh2meryrg01y4wRqO3BtI00vZE6B/imLyw=
> =/9AM
> -----END PGP SIGNATURE-----
>
> --
> You received this message because you are subscribed to the Google Groups 
> "qubes-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to [email protected].
> To post to this group, send email to [email protected].
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/qubes-devel/20170922214145.GV1116%40mail-itl.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"qubes-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/qubes-devel/CAKYr3zzkQjDf6kj0ojguLLN91%2B-j3-hWFSX%3D7jnrGAwDS2Sicg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to