[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gentoo-amd64
Subject:    [gentoo-amd64]  Re: mldonkey update issues
From:       Duncan <1i5t5.duncan () cox ! net>
Date:       2007-02-15 6:54:46
Message-ID: er103m$7r3$2 () sea ! gmane ! org
[Download RAW message or body]

Mauro Maroni <mmaroni@fi.uba.ar> posted
200702142151.54947.mmaroni@fi.uba.ar, excerpted below, on  Wed, 14 Feb
2007 21:51:54 -0300:

> While doing a general upgrade yesterday, I noticed that mldonkey fails
> almost at the end (see below) and emerge ends. I tried again today but got
> the same result.  I did not find any bug related to this. Did anyone know
> where is the problem?
>

>>>> Merging net-p2p/mldonkey-2.8.2-r1 to /
> --- /usr/
> --- /usr/bin/
>>>> /usr/bin/mlnet
>>>> /usr/bin/mld_hash
>>>> /usr/bin/get_range
>>>> /usr/bin/copysources
>>>> /usr/bin/make_torrent
>>>> /usr/bin/subconv
>>>> /usr/bin/mldonkey
> --- /etc/
> --- /etc/conf.d/
> Traceback (most recent call last):
>   File "/usr/bin/emerge", line 5299, in ?
>     retval = emerge_main()
>   File "/usr/bin/emerge", line 5294, in emerge_main
>     myopts, myaction, myfiles, spinner)
>   File "/usr/bin/emerge", line 4761, in action_build
>     retval = mergetask.merge(pkglist, favorites, mtimedb)
>   File "/usr/bin/emerge", line 2999, in merge
>     prev_mtimes=ldpath_mtimes)
>   File "/usr/lib/portage/pym/portage.py", line 3699, in doebuild
>     vartree=vartree, prev_mtimes=prev_mtimes)
>   File "/usr/lib/portage/pym/portage.py", line 3907, in merge
>     mydbapi=mydbapi, prev_mtimes=prev_mtimes)
>   File "/usr/lib/portage/pym/portage.py", line 7768, in merge
>     cleanup=cleanup, mydbapi=mydbapi, prev_mtimes=prev_mtimes)
>   File "/usr/lib/portage/pym/portage.py", line 7364, in treewalk
>     if
> self.mergeme(srcroot,destroot,outfile,secondhand,"",cfgfiledict,mymtime):
>   File "/usr/lib/portage/pym/portage.py", line 7640, in mergeme
>     join(offset, x), cfgfiledict, thismtime):
>   File "/usr/lib/portage/pym/portage.py", line 7640, in mergeme
>     join(offset, x), cfgfiledict, thismtime):
>   File "/usr/lib/portage/pym/portage.py", line 7661, in mergeme
>     if self.isprotected(mydest):
>   File "/usr/lib/portage/pym/portage.py", line 7116, in isprotected
>     mydata = self._installed_instance.getcontents().get(filename, None)
> AttributeError: 'NoneType' object has no attribute 'get' amd64 ~ #
> amd64 ~ #

I track portage updates and watch what bugs they are fixing.  As a result, I
see a lot of portage bugs, and this would appear to be one -- at least
indirectly.  Even if a file is corrupt or something, portage should quit
with an intelligent error.  That it crashes with a traceback therefore
indicates either a direct portage bug, or at minimum (and perhaps more
likely), a bug in that a file or something was corrupt that it should have
been prepared to handle but wasn't.

I'd suggest filing it as a portage bug (of course, make sure your portage is
up to date and try it again before you file, if it's not, in case the bug
has already been fixed) and letting the experts deal with it.  They are
likely to tell you something either in your currently merged version or in
portage's record of it is corrupt, but will be able to point you to what it
is, and in the process likely fix the traceback such that portage quits with
a coherent error that at least lets the user know what was wrong so they
know where to start looking to fix it.

You may be able to fix the issue yourself by unmerging whatever version you
have currently merged, or possibly deleting the record of it in /var/db/
portage, then remerging the old version to give portage a correct record so
it can then properly unmerge it when you upgrade.  However, that will likely
destroy the evidence that the portage folks need to trace and fix their bug,
and there's a tiny chance it could make things worse.  It'd be much better
to file the bug and let the portage experts have at it.  They should be able
to tell you what's happening and how to fix it, tho from the bugs I've seen,
they do sometimes have you run an intermediate patch that doesn't fix it,
but helps them to better pin down the problem.

--
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman

-- 
gentoo-amd64@gentoo.org mailing list

[prev in list] [next in list] [prev in thread] [next in thread]