Yum goes yuck

[Jonathan Corbet]

Another day, another attempted update...

[...]
Checking deps for mesa-libGLU.i386 0-6.5.2-6.fc7 - u
Checking deps for e2fsprogs.x86_64 0-1.39-11 - u
Traceback (most recent call last):
  File "/usr/bin/yum", line 29, in <module>
    yummain.main(sys.argv[1:])
  File "/usr/share/yum-cli/yummain.py", line 135, in main
    (result, resultmsgs) = base.buildTransaction() 
  File "/usr/lib/python2.5/site-packages/yum/__init__.py", line 451, in
buildTransaction
    (rescode, restring) = self.resolveDeps()
  File "/usr/lib/python2.5/site-packages/yum/depsolve.py", line 1104, in
_resolveDeps
    deps = self._mytsCheck()
  File "/usr/lib/python2.5/site-packages/yum/depsolve.py", line 1082, in
_mytsCheck
    ret.extend(self._checkUpdate(txmbr))
  File "/usr/lib/python2.5/site-packages/yum/depsolve.py", line 1265, in
_checkUpdate
    ret = self._checkInstall(txmbr)
  File "/usr/lib/python2.5/site-packages/yum/depsolve.py", line 1261, in
_checkInstall
    member.setAsDep(txmbr.po)
UnboundLocalError: local variable 'member' referenced before assignment

A look at the function makes it clear that there are paths were "member"
could fail to be set, but it didn't yield a whole lot more light than
that... 

FWIW, the problem was 100% repeatable.  Updating e2fsprogs separately
made it go away, though.

jon

[Michal Jaegermann]

On Wed, Feb 28, 2007 at 03:32:34PM -0700, Jonathan Corbet wrote:
>  Another day, another attempted update... 
Yum seems to get overwhelmed by an amount of updates which showed
up.  Repositories also appear to be in a not entirely sane state.
See, for example:
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=230477
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=230485

   Michal

[Jonathan Corbet]

Michal Jaegermann <michal(a)harddata.com&gt; wrote:

>  https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=230485 
Interesting, I was just trying to track that one down myself.  Now that
yum doesn't crash, it seems to want to populate my x86-64 system with
i386 packages.  Or, more to the point, I thinks somehow I have already
done so.  

For example, it wants to "update" GConf2.i386.  Investigating this
yields some interesting results:

	[root@bike corbet]# rpm -q GConf2
	GConf2-2.16.0-6.fc7
	[root@bike corbet]# rpm -q GConf2.i386
	[root@bike corbet]# rpm -q GConf2.x86_64
	GConf2-2.16.0-6.fc7
	[root@bike corbet]# rpm -e GConf2.i386
	[root@bike corbet]# rpm -e GConf2.i386
	[root@bike corbet]# rpm -e GConf2.i386
	[root@bike corbet]# rpm -e ThisIsWeird.i386
	error: package ThisIsWeird.i386 is not installed

Any attempt to do anything with the "i386 version" yields silence, even
though that package certainly does not exist on my system.  Something is
very confused in rpmland...

jon

[Jeremy Katz]

On Wed, 2007-02-28 at 15:32 -0700, Jonathan Corbet wrote:
>  Another day, another attempted update... 
[snip]
>    File "/usr/lib/python2.5/site-packages/yum/depsolve.py",
> line 1261, in _checkInstall
>      member.setAsDep(txmbr.po)
>  UnboundLocalError: local variable 'member' referenced before assignment 
Thanks for the report -- fixed in CVS and will be in 3.1.3

Jeremy

[Kwan Lowe]

>  On Wed, 2007-02-28 at 15:32 -0700, Jonathan Corbet wrote:
> > Another day, another attempted update...
>  [snip]
> >   File "/usr/lib/python2.5/site-packages/yum/depsolve.py", line 1261, in
> > _checkInstall
> >     member.setAsDep(txmbr.po)
> > UnboundLocalError: local variable 'member' referenced before assignment
>
>  Thanks for the report -- fixed in CVS and will be in 3.1.3 
Probably a dumb question... Not certain if it's related:
Is there any sort of flag set while the repository mirrors are being updated? I.e.,
during their synchronization, is there a way to tell if it's not fully updated?  The
reason I'm asking is that I rsync to my own repository and sometimes get incomplete
pulls.

-- 
* The Digital Hermit   http://www.digitalhermit.com
* Unix and Linux Solutions   kwan(a)digitalhermit.com