| 1 |
Michael Haubenwallner <mailto:haubi@g.o> wrote: |
| 2 |
> On Sat, 2008-04-19 at 23:11 +0200, Fabian Groffen wrote: |
| 3 |
>> On 19-04-2008 16:07:19 -0500, matt hull wrote: |
| 4 |
>>>> Due to numerous occasions of b0rkerage in the bootstrap snapshots, |
| 5 |
>>>> bootstrapping has failed. A selection of issues that come up: |
| 6 |
>>>> - XXXX/YYYYY-p.q is the latest version in the snapshot, but this |
| 7 |
>>>> version has been removed, and its distfiles have become |
| 8 |
>>>> unavailable (e.g. rsync-3.0.0_pre2) - XXXX/YYYYY-p.q was added |
| 9 |
>>>> in the snapshot, but breaks several packages (e.g. gcc-4.3.0) |
| 10 |
>>> |
| 11 |
>>> perhaps we need to keep older versions for bootstrap ? |
| 12 |
>> |
| 13 |
>> I don't think that's a problem. In general the highest version in |
| 14 |
>> the tree is used at the moment due to our single keywordness. |
| 15 |
>> |
| 16 |
>>>> To aid the pre-sync stage, I am considering to switch to usign |
| 17 |
>>>> stable keywords for the system packages *only*. That is, the |
| 18 |
>>>> bootstrap process is done with stable keywords, all other packages |
| 19 |
>>>> remain ~arch and hence a user has to add ~arch to ACCEPT_KEYWORDS |
| 20 |
>>>> after bootstrapping finishes. |
| 21 |
>>> |
| 22 |
>>> i would think it would be better to use stable and testing like |
| 23 |
>>> normal. i would like to try to keep stable here. |
| 24 |
>> |
| 25 |
>> That's very unlikely to happen from my point of view. For that we |
| 26 |
>> really need a bunch of devs that are going to maintain the stable |
| 27 |
>> keywords like the arch teams currently do for gentoo-x86. |
| 28 |
> |
| 29 |
> As we have approximately less than 1 dev per arch, IMO for the |
| 30 |
> bootstrap _only_ it could be possible to have "stable and testing |
| 31 |
> like normal", but we might need to reduce the definition of "normal" |
| 32 |
> (drop stablereq bugs, ...), but then we cannot call it "normal" any |
| 33 |
> more ;) |
| 34 |
|
| 35 |
+1. For me "stable" is what i managed to bootstrap with, and my prefix |
| 36 |
works without constantly core-ing :) I think it would suffice to mark |
| 37 |
all packages in a fresh bootstrapped prefix to be "bootstrap safe" |
| 38 |
(which of course may look the same as main's stable keywords), after |
| 39 |
playing a little with the prefix and not seeing serious issues. |
| 40 |
|
| 41 |
At least for platforms like interix i see nothing else making sense, |
| 42 |
since for example i'm the only person right now on interix ... if i |
| 43 |
can't see any issue, and bootstrapping went ok, it will probably go ok |
| 44 |
again, and after that i can still update packages where there are fixes |
| 45 |
for not so critical issues. |
| 46 |
|
| 47 |
Cheers, Markus |
| 48 |
|
| 49 |
> |
| 50 |
> /haubi/ |
| 51 |
> -- |
| 52 |
> Michael Haubenwallner |
| 53 |
> Gentoo on a different level |
| 54 |
|
| 55 |
-- |
| 56 |
gentoo-alt@l.g.o mailing list |
Archives