The Spin SIG meeting was held today and this has a couple of impacts on
the games spin.
For the alpha release it was decided to keep the iso's under 4 GiB. This
will be looked at again before the beta. I took a look at package sizes
and I either needed to get rid of one large game or many other non-game
packages. And at this point removing one game seems like a better option.
The game I have selected to remove for the alpha (pending my spin compose
test coming in under 4 GiB) is openarena. Of the large games it seemed
to be the one that was similar to other games and hence most expendable.
Spins will now not be features. I need to create a spin page with a slightly
different format and move over all of my data from the feature page. This
won't be done tonight, but I will send another update when it is complete
that includes the new URL.
I plan to continue testing livecd-creator patches to allow larger spins
to make sure this feature is in place and provides an option for larger
I am still having problems with X on F11, which limits the amount of testing
I can do.
I noticed that the Fedora Games SIG was not mentioned at
I fixed that.. so perhaps now more people will be coming you're way.
I also saw that more contributors was listed as a general issue at
you're Wiki page, perhaps these two things went hand in hand.
Magnus 'magnusg' Glantz
E-mail: mg -at- hacka -dot- net
GPG Fingerprint: DDCE BD02 793C 5C2E 8C5C 3FE9 1E1E 7593 0DB5 3317
The rawhide update last night was pretty good and I was able to retest the
dependencies bugs. The good news there is that we are down to just nethack
(not nethack-vultures which got fixed). The bad news is that I looked at the
scripts for nethack, and they involve setting up links to font directories.
So this is a bug that appears to have more impact than just an error message
and would be nice to get fixed.
X in F11 hasn't been working for me for about a week now. I don't really
need it for the build testing, but for making sure the build does more
than boot, I could use a graphical display. (Actually X has been broken
a lot longer than that, but I apply a patch locally to force the ati
driver to acknowledge the existance of the LCD which doesn't do EDID
correctly. Normally that gets me by, but now something else is broken.)
I updated the kickstart file in the git repository and it should make its
way into rawhide soon.
The livecd-tools update is also in the git repository and rel-eng might
use that copy to do the spin. (Also note the F10 version of livecd-tools
has a higher NVR than the F11 version.)
The 4 GiB stuff is still up in the air. I'd like to try > 4 GiB for the
alpha and if there are a lot of problems switch back for the beta. I'll
need to get a backup kickstart file ready in case the Spins SIG says stay
under 4 GiB.
poker2d got updated.
Jon Ciesla helped get the spec file changes in for nethack-vultures.
Jeremy Katz applied the livecd-creator patch to allow for 4+ GiB
squashfs images. This isn't in rawhide yet, but Jeremy has said if it
isn't by the alpha they can use the git version to do the alpha spin.
There will probably be some discussion at FudCon about the Spin process
and the Games Spin in particular. Approval still isn't guaranteed.
I plan to attend the January 16th FESCO meeting either for a normal
approval request if things fall into place for the Spin process or
to ask for a temporary exception to get a games spin made for the
alpha release while the process requirements are being figured out.
Those are all of the issues that were left blocking the games spin for
the alpha release. So while it isn't for sure yet, it is looking pretty
There are a couple of other things that produce error messages during the
livecd-creator run, but I don't think any of them actually result in a
I have been bugging people to try to get info on what I need to do to get
a Games Spin published as part of the alpha release.
If necessary I'll temporarily drop a couple of games in order to have
a working kickstart file for the alpha release. (Either to keep broken
packages, such as poker2d, out of the kickstart file or to get below the 4 GiB
limit if my livecd-creator patch isn't in rawhide in time for the alpha
I was able to test the livecd-creator patch on a spin that was under the
4 GiB limit and it also seemed to work. So I am pretty much just waiting
for review there.
I ended up filing a bug for poker2d about the dependency issue triggered
by the upgrade of poker-network.
I plan on doing a change request bug for the kickstart file tonight. Things
are getting close enough that I want to start things moving their. The current
kickstart file is broken in F11 anyway as it doesn't pull in a display
manager since it is based of the base spin instead of the desktop spin.
It would be nice if someone could work on either the nethack-vultures or
poker2d bugs. The nethack-vultures one is easy and pretty much any maintainer
could take care of it.
As usual, pretty much everything I have been putting in these status reports
is also being tracked (with links to all of the related bugzilla entries) on
the proposed feature page at:
Things have been moving slowly getting ready for rawhide testing. I wanted to
relayout my disks and get both rawhide and F10 resinstalled. There is still
more more to do with this. I expect to be testing on rawhide on Wednesday
The first thing I will be doing after I have a working rawhide install is
confirming fixes for the dependencies bugs that I filed and have new updates.
I think most of them have been fixed. In one case the maintainer's key
doesn't work any more and they don't know how to proceed. (This is bug
475919, if someone here can help them out.)
Assuming no new problems show up, I'll move on to working on extending
my patches to livecd-tools for switching to udf when the squashfs file
system reaches 4GiB. I am off this week, so I think I'll have a good
chance of having something (a patch for livecd-tools) reviewable by the end
of this weekend (the 4th).
The Spins SIG still needs to work on what the policy will be for Spins.
(So that the Feature Wrangler can properly evaluate the feature.)
I don't expect this to be a problem, but if something ends up needing
to absolutely done by the alpha there could be a problem.
If the spin's squashfs file system still fits under 4GiB, then I don't
think should be a problem having a spin available for the alpha. (Though
I am not sure how that is going to work procedurely.) If it is over
4GiB, then it will depend on having an update applied to livecd-tools before
the alpha. I may not have an acceptable patch by that time. This potentially
could also hit a policy issue regarding whether or not spins can use
udf even if they still fit on a DVD as that may limit the hardware they
can be used on. (I wouldn't expect their to be many systems that can boot
off a DVD, that couldn't also use udf.)