1 |
Mark Knecht wrote: |
2 |
> On Tue, Jul 13, 2010 at 11:29 AM, Dale<rdalek1967@×××××.com> wrote: |
3 |
> |
4 |
>> Mark Knecht wrote: |
5 |
>> |
6 |
>>> Right. I misspoke. New use flags, not new a package. |
7 |
>>> |
8 |
>>> However the outcome is consistent. It says it's 'fetching' because |
9 |
>>> it's responding to the idea that there's a new use flag, but then |
10 |
>>> fetches nothing because the file is already here and builds nothing |
11 |
>>> because it's already been built. |
12 |
>>> |
13 |
>>> It's not a big deal. The machine is fine. Everything is consistent as |
14 |
>>> far as I know. However some portion of the download logic in emerge is |
15 |
>>> confused. |
16 |
>>> |
17 |
>>> I've seen this sort of thing a few times in the past but it's always |
18 |
>>> cleared up in a day or two. This one has gone on for weeks. |
19 |
>>> |
20 |
>>> Note that I do use ~amd64 on portage& eix. Maybe it's a bug that |
21 |
>>> hasn't been cleaned up yet. |
22 |
>>> |
23 |
>>> So, following through, there does appear to be a new flag setting |
24 |
>>> (-gdu) however it's not triggered emerging @world even though packages |
25 |
>>> on the system require gvfs according to equery. |
26 |
>>> |
27 |
>>> Somethings inconsistent. (Possibly my brain!) |
28 |
>>> |
29 |
>>> Cheers, |
30 |
>>> Mark |
31 |
>>> |
32 |
>>> c2stable ~ # emerge -pvDuN gvfs |
33 |
>>> |
34 |
>>> These are the packages that would be merged, in order: |
35 |
>>> |
36 |
>>> Calculating dependencies... done! |
37 |
>>> [ebuild R ] gnome-base/gvfs-1.4.3-r1 USE="gnome http udev |
38 |
>>> -archive -avahi -bash-completion -bluetooth -cdda -doc -fuse -gdu* |
39 |
>>> -gnome-keyring -gphoto2 -hal -samba" 0 kB |
40 |
>>> |
41 |
>>> Total: 1 package (1 reinstall), Size of downloads: 0 kB |
42 |
>>> c2stable ~ # emerge -pvDuN @world |
43 |
>>> |
44 |
>>> These are the packages that would be merged, in order: |
45 |
>>> |
46 |
>>> Calculating dependencies... done! |
47 |
>>> |
48 |
>>> Total: 0 packages, Size of downloads: 0 kB |
49 |
>>> c2stable ~ # equery depends gvfs |
50 |
>>> * These packages depend on gvfs: |
51 |
>>> app-cdr/brasero-2.28.3 (gnome-base/gvfs) |
52 |
>>> gnome-base/gnome-2.28.2 (>=gnome-base/gvfs-1.4.3[gdu]) |
53 |
>>> gnome-base/libgnome-2.28.0 (gnome-base/gvfs) |
54 |
>>> gnome-base/nautilus-2.28.4-r1 (>=gnome-base/gvfs-0.1.2) |
55 |
>>> c2stable ~ # |
56 |
>>> |
57 |
>>> |
58 |
>>> Just curious. |
59 |
>>> |
60 |
>>> Cheers, |
61 |
>>> Mark |
62 |
>>> |
63 |
>>> |
64 |
>>> |
65 |
>> What does it do when you run: |
66 |
>> |
67 |
>> emerge -vDuN gvfs |
68 |
>> |
69 |
>> It should start to compile the package. Keep in mind, if you use the -f |
70 |
>> option, it will fetch but not compile or anything else. I used to use that |
71 |
>> back when I was on dial-up, which sucked by the way. Even this slow DSL is |
72 |
>> about 30 times faster than my old dial-up. lol It *should* only fetch it |
73 |
>> once tho unless it fails the checksum thingy. Portage should give you a |
74 |
>> error message tho. |
75 |
>> |
76 |
>> Dale |
77 |
>> |
78 |
> That probably does build. (The machine is now powered down so I'll |
79 |
> save that for tomorrow.) |
80 |
> |
81 |
> I understand (or thought I did!) ;-) the use of -f. I use it every day |
82 |
> to download any packages that are needed to update the system before |
83 |
> starting the actual updates. I don't like the updates to get held up |
84 |
> by some server that's not responding and giving me the files. |
85 |
> |
86 |
> So, every day I do updates (2-3 days a week maybe) the basic set of |
87 |
> commands I consider running through looks like: |
88 |
> |
89 |
> eix-sync |
90 |
> emerge -fDuN @world |
91 |
> emerge -pvDuN @system |
92 |
> emerge -DuN @system |
93 |
> emerge -pvDuN @world |
94 |
> emerge -DuN @world |
95 |
> revdep-rebuild -ip |
96 |
> eix-test-obsolete -d |
97 |
> emerge -p --depclean |
98 |
> |
99 |
> Usually it takes 10-15 minutes if it's not a big day and it has ALWAYS |
100 |
> resulted in |
101 |
> |
102 |
> emerge -fDuN @world |
103 |
> |
104 |
> fetching NOTHING new, at least until gvfs came along... |
105 |
> |
106 |
> Cheers, |
107 |
> Mark |
108 |
> |
109 |
> |
110 |
|
111 |
Well, I do mine a bit differently. I do this way: |
112 |
|
113 |
eix-sync && emerge -uvDNa world |
114 |
|
115 |
You may want to add the -f to that then come back and do the actual |
116 |
emerge. If you use world instead of @world, you don't need to worry |
117 |
about @system since it is included in the world file. |
118 |
|
119 |
Then when that is done you can run emerge -p --depclean and |
120 |
revdep-rebuild. I would run revdep-rebuild last since --depclean could |
121 |
cause a problem. I only had that once but it may be even safer to run |
122 |
revdep-rebuild then --depclean then revdep-rebuild again just to |
123 |
alleviate any paranoia. |
124 |
|
125 |
Making sure portage didn't tell you to fix anything broken is a good |
126 |
idea too. I use either elogv or elogviewer to check those. |
127 |
|
128 |
Dale |
129 |
|
130 |
:-) :-) |