1 |
On Mon, May 21, 2012 at 10:13 AM, Markos Chandras <hwoarang@g.o> wrote: |
2 |
<SNIP> |
3 |
>> if you use testing, you have to deal with such kind of situations. |
4 |
>> Using a known broken version is just stupid. There isn't a choice |
5 |
>> between those two. There is only a choice between: use unstable or |
6 |
>> stable. And if you use unstable, don't complain about things being |
7 |
>> fluid. |
8 |
>> |
9 |
|
10 |
Typical Volker. Not worth a response. |
11 |
|
12 |
> - From what I can tell, he is not using ~amd64 (testing). He uses a |
13 |
> mixed system (stable with a few packages in package.keywords) which |
14 |
> sometimes is even worse :) |
15 |
> |
16 |
|
17 |
That's it precisely. All my systems are 'stable' as best they can be. |
18 |
The only things that are more or less permanently testing are portage |
19 |
& eix. After that the only entries in package.keywords are specific |
20 |
applications or _very_ targeted things that address hardware issues. |
21 |
(Virtualbox, VMWare Player, nvidia-drivers) That's it. In general, the |
22 |
only reason ANYTHING goes in package.keywords is to address a |
23 |
temporary package problem, and typically they aren't there that long. |
24 |
For instance, if say Virtualbox had a problem I might keyword it for a |
25 |
month or two to get a new release that fixes the problem. After that |
26 |
version goes stable I remove the keyword entry and am running the same |
27 |
version I've been running. My keyword file is currently only about 12 |
28 |
lines. |
29 |
|
30 |
In the case of skype there was no stable version to use. Everything is |
31 |
marked as testing. |
32 |
|
33 |
While I agree that the emulation libraries are probably low risk I |
34 |
don't want to add 8 new things to package.keywords. It's just not the |
35 |
way I work here. |
36 |
|
37 |
> The best way to deal with your problem (and avoid seeing your |
38 |
> package.keywords getting bigger and bigger) is to grab -r1, mark it |
39 |
> stable, put it in your local overlay and keep using that indefinitely. |
40 |
> |
41 |
|
42 |
Yeah, most likely the best medium-term solution although what I've |
43 |
done for now is emerge -C skype. |
44 |
|
45 |
> - -- |
46 |
> Regards, |
47 |
> Markos Chandras / Gentoo Linux Developer / Key ID: B4AFF2C2 |
48 |
|
49 |
|
50 |
Thanks Markos! I really appreciate your inputs as well as all the work |
51 |
you guys do. We don't say thanks enough. |
52 |
|
53 |
(Ugh Volker. Keep the attitude to yourself dude) ;-( |
54 |
|
55 |
Cheers, |
56 |
Mark |