#1680 [freeze break] Tag Fixes for F11 Mass Rebuild Failures
Closed: Fixed None Opened 14 years ago by mjakubicek.

I've fixed a whole bunch of packages that failed in February during the F11 Mass Rebuild. As for the goals of this rebuild, I think that those should be tagged for dist-f11:

kdetv-0.8.9-12.fc11[[BR]]
cone-0.75-5.fc11[[BR]]
showimg-0.9.5-22.fc11[[BR]]
ruby-ldap-0.9.7-8.fc11[[BR]]
boo-0.8.1.2865-6.fc11[[BR]]
darkice-0.19-5.fc11[[BR]]
libresample-0.1.3-11.fc11[[BR]]
Sprog-0.14-16.fc11[[BR]]
R-GeneR-2.13.1-2.fc11[[BR]]
libgeotiff-1.2.5-4.fc11[[BR]]
itpp-4.0.6-1.fc11[[BR]]
kadu-0.6.5-10.fc11[[BR]]
postgis-1.3.5-3.fc11[[BR]]
postgresql-pgpool-ha-1.1.0-7.fc11[[BR]]
gnomesword-2.4.1-3.fc11[[BR]]
reciteword-0.8.4-5.fc11[[BR]]
velocity-1.4-8.4.fc11[[BR]]
fldigi-3.10-3.fc11[[BR]]
stardict-3.0.1-17.fc11[[BR]]
rxvt-unicode-9.06-3.fc11[[BR]]
!MyPasswordSafe-0.6.7-8.20061216.fc11[[BR]]
log4cxx-0.10.0-7.fc11[[BR]]
javahelp2-2.0.05-7.fc11[[BR]]
jakarta-commons-el-1.0-10.5.fc11[[BR]]
trousers-0.3.1-15.fc11[[BR]]
incollector-1.0-8.fc11[[BR]]
python-shout-0.2.1-3.fc11[[BR]]
rlog-1.4-5.fc11[[BR]]
ctapi-cyberjack-3.3.0-6.fc11[[BR]]
pdf2djvu-0.5.0-3.fc11[[BR]]


gnome-specimen-0.3-4.fc11[[BR]]
seedit-2.2.0-5.fc11[[BR]]
ws-commons-util-1.0.1-11.fc11[[BR]]
ws-jaxme-0.5.1-3.4.fc11[[BR]]
metamonitor-0.4.5-7.fc11[[BR]]
perl-Hardware-Vhdl-Tidy-0.8-4.fc11[[BR]]
mx4j-3.0.1-8.9.fc11[[BR]]
latex2html-2008-2.fc11[[BR]]
jline-0.9.94-0.3.fc11[[BR]]
xdoclet-1.2.3-10.4.fc11[[BR]]
modello-1.0-0.2.a8.4.4.fc11[[BR]]
batik-1.7-4.fc11[[BR]]

(The packages under the line could have been built without any modifications, i.e. previously they probably failed because of some deps).

Cheers,
Milos


My that's a lot. I'd tend to think that shipping packages that are actually buildable (and benefit from the mass rebuild and all) is generally a good thing, despite the lack of testing here (in alpha, beta).

+1 (ish)

As an aside, I'm concerned about the maintainers of these packages not being responsive during f-11 development...

Replying to [comment:1 rdieter]:

My that's a lot. I'd tend to think that shipping packages that are actually buildable (and benefit from the mass rebuild and all) is generally a good thing, despite the lack of testing here (in alpha, beta).

I should note here that I believe that the fixes I've made were indeed rather small (however sometimes not trivial to find out:), they mostly consist of following items:

  • gcc 4.4 header includes & glibc changes (const char vs. char etc.)
  • mono one-line patches
  • java patches to avoid naming variables as "enum" since this is a reserved word in java >= 1.5
  • java missing BR (if there is another way how to find out what package provides foo.bar.baz java package than querying java-people like lkundrak and dbhole who then grep all they .jars for the given string, I (and they too:) would be happy to know about)
  • other missing BR.

I've also tried to fix trivial bugs that were assigned to the given packages, like missing requires and similar.

...and yes, all the patches I've made have been upstreamed (bugtracker, mailing list, author e-mail, in this availability order) except of packages which were really outdated, in that case, this is clearly noted along with the PatchX: line in the spec file ("Not sent upstream, please update the package first." -- or similar comments).

One exception is itpp-4.0.6-1.fc11, which (as can be seen from the release number) has been updated to the next maintenance release (4.0.0 => 4.0.6).

To sum up: I don't think that these new builds contain something bad that was not there before;)

+1 (ish)

As an aside, I'm concerned about the maintainers of these packages not being responsive during f-11 development...

Me too -- some of them have currently inactive accounts as I received undeliverable e-mail notification after the commit, but many probably just don't care, or even don't know there is something bad with their packages (even some which are @redhat.com -- I tried to contact those people via IRC, some of them promised to fix their packages).

+1, just to get these into F-11 final. All of these are tagged. Thanks for the effort in getting these cleaned up.

I'm reopening this to continue with a smaller set consisting of:

avr-libc-1.6.4-3.fc11
nx-3.3.0-34.fc11
guimup-0.1.4-6.c.fc11
polyxmass-bin-0.9.8-1.fc11
scrip-1.4-13.fc11
javassist-3.9.0-4.fc11
licq-1.3.5-7.fc11
gnome-applet-grandr-0.4.1-1.fc11
cowbell-0.3-0.svn47.1.fc11.2
tagsoup-1.0.1-3.3.fc11 [<= ExcludeArch: ppc64]
busybox-1.13.2-4.fc11 [<= ExcludeArch: ppc64]

As we're very close to F11 GA, I'm not sure whether these should be tagged for f11-final, or I should rather push them all as one update (...or do nothing) -- please decide what you think it's the best now.

Sorry for bad formatting, here is the list once more:

avr-libc-1.6.4-3.fc11[[BR]]
nx-3.3.0-34.fc11[[BR]]
guimup-0.1.4-6.c.fc11[[BR]]
polyxmass-bin-0.9.8-1.fc11[[BR]]
scrip-1.4-13.fc11[[BR]]
javassist-3.9.0-4.fc11[[BR]]
licq-1.3.5-7.fc11[[BR]]
gnome-applet-grandr-0.4.1-1.fc11[[BR]]
cowbell-0.3-0.svn47.1.fc11.2[[BR]]
tagsoup-1.0.1-3.3.fc11 [<= !ExcludeArch: ppc64][[BR]]
busybox-1.13.2-4.fc11 [<= !ExcludeArch: ppc64]

We're so close that I don't wish to push them. We know they build now, so next time they need an update they'll build, and they'll be rebuilt in the f12 rawhide stream. Thank you very much for your efforts!

Metadata Update from @mjakubicek:
- Issue set to the milestone: Fedora 11 Final

7 years ago

Login to comment on this ticket.

Metadata