1 |
On Wed, 25 Apr 2012 09:16:05 +0200 |
2 |
Corentin Chary <corentin.chary@×××××.com> wrote: |
3 |
|
4 |
> On Tue, Apr 24, 2012 at 6:38 PM, Michał Górny <mgorny@g.o> |
5 |
> wrote: |
6 |
> > On Tue, 24 Apr 2012 16:19:11 +0000 |
7 |
> > "Robin H. Johnson" <robbat2@g.o> wrote: |
8 |
> > |
9 |
> >> On Tue, Apr 24, 2012 at 04:50:49PM +0200, Corentin Chary wrote: |
10 |
> >> > >> $ ./mirrors.py --all --count |
11 |
> >> > >> 297 ?? ?? http://pear.php.net |
12 |
> >> > >> 297 ?? ?? http://pear.php.net/get |
13 |
> >> > >> 88 ?? ?? ??http://pecl.php.net |
14 |
> >> > >> 88 ?? ?? ??http://pecl.php.net/get |
15 |
> >> > > These are already mirror bouncers. If you visit the above, |
16 |
> >> > > you'll get the closest mirror for downloading. |
17 |
> >> > And since there is already ~10 "mirrors" with only one actual |
18 |
> >> > backend, should they go to thirdpartymirrors or not ? If not, |
19 |
> >> > what about this pseudo-mirrors already present in |
20 |
> >> > thirdpartymirrors ? |
21 |
> >> I think we should add the pseudo-mirrors, but explicitly mark them |
22 |
> >> as such in the file, so that they don't get duplicate entries |
23 |
> >> added (eg adding us.pear, de.pear and the pear bouncer is bad. |
24 |
> >> Should have just the bouncer). |
25 |
> > |
26 |
> > It'd be great if we could add some kind of additional mirror |
27 |
> > entries, which would be used by repoman to signal missing mirror:// |
28 |
> > entries but won't be used for downloads. |
29 |
> |
30 |
> Yep, we could put that in it too: |
31 |
> github http://github.com/downloads/ |
32 |
> https://github.com/downloads/ |
33 |
|
34 |
Per spec, portage can choose a random mirror of the list. If we put |
35 |
entries like that, these two will be equally possible as the preferred |
36 |
cloud. URL -- while they redirect one to another. |
37 |
|
38 |
We might decide on some common syntax like preceding all extra entries |
39 |
with '-' but I don't want to be the one deciding here. |
40 |
|
41 |
-- |
42 |
Best regards, |
43 |
Michał Górny |