| 1 |
Hi; |
| 2 |
|
| 3 |
Im running the default-1.0 profile, and keeping current |
| 4 |
near daily. |
| 5 |
After upgrading to python-2.2-r6, I unmerged an older |
| 6 |
version (probably -r5). |
| 7 |
This had the unfortunate effect of breaking portage. |
| 8 |
(The symptom was that it complained about importing the module |
| 9 |
'output' when you ran emerge.) |
| 10 |
I had just also upgraded portage, and unmerged the old |
| 11 |
one in the same operation (using pkg-clean), so I initially |
| 12 |
suspected portage was the problem. I installed the rescue portage |
| 13 |
as described in /usr/portage/sys-apps/portage/files/README.RESCUE |
| 14 |
But that didnt fix it. However it gave me a more informative |
| 15 |
traceback-- instead of just complaining about 'output' module, |
| 16 |
it went deeper and ended up unable to find 'fchksum'. |
| 17 |
After looking at -r5 and -r6 python ebuilds, I noticed that |
| 18 |
-r5 installed python-fchksum on its build, which -r6 doesnt (its |
| 19 |
now in a separate ebuild: dev-python/python-fchksum ) |
| 20 |
I assume this has something to do with the problem, |
| 21 |
though emerge shouldnt harm newer versions of something it |
| 22 |
is umerging... I wish I could provide an explaination, but |
| 23 |
most of the evidence was destroyed in my efforts. |
| 24 |
So, long story short, I untarred python-fchksum-1.6.1.tar.gz |
| 25 |
which was still in /usr/portage/distfiles, otherwise available |
| 26 |
from: |
| 27 |
|
| 28 |
http://www.azstarnet.com/~donut/programs/fchksum/python-fchksum-1.6.1.tar.gz |
| 29 |
|
| 30 |
and following the README, built and installed it manually. After that, |
| 31 |
emerge worked again, and I was back in business... |
| 32 |
|
| 33 |
Paul |
| 34 |
set@×××××.com |
Archives