| 1 |
On Wednesday 10 November 2004 14:52, Kurt Lieber wrote: |
| 2 |
> Anyway, enough preaching. This thread has gone on long enough. The |
| 3 |
> solution that's been agreeed upon is signing the daily snapshots that |
| 4 |
> we provide for users who can't use rsync. (/snapshots directory on |
| 5 |
> your favorite source mirror) |
| 6 |
|
| 7 |
All right, repeating it is not usefull. |
| 8 |
|
| 9 |
> |
| 10 |
> This provides the ability to verify the integrity of every single file |
| 11 |
> under /usr/portage/ and requires very little changes to our existing |
| 12 |
> infrastructure. emerge-webrsync will be hacked up to provide |
| 13 |
> verification support for it. I don't have any commitments from the |
| 14 |
> portage devs that these changes will be included (emerge-webrsync is |
| 15 |
> part of portage) so this may end up being an unsupported, |
| 16 |
> use-at-your-own-risk solution. It does not take away from or alter the |
| 17 |
> plans to implement a much better, more robust verification solution in |
| 18 |
> portage itself. |
| 19 |
|
| 20 |
Well, finally some useable solution. I'm fairly confident that the portage |
| 21 |
devs will support it. I think it can be an acceptable measure until the |
| 22 |
final measures are finalized. |
| 23 |
|
| 24 |
Paul |
| 25 |
|
| 26 |
> P.S. I do not want anyone to think that this solution is being |
| 27 |
> implemented because of the bitching and screaming that occurred. If |
| 28 |
> someone had posted a message to the list before all this broke out |
| 29 |
> suggesting this solution and volunteering to write the code for it, it |
| 30 |
> would be in place by now. That's another way of saying that we didn't |
| 31 |
> have to go through all this unpleasantness... |
| 32 |
|
| 33 |
ps. I'm fairly confident that all the bashing has in general been |
| 34 |
counterproductive. I certainly have still about 100 mails on the mailing |
| 35 |
list laying about, which I don't intend to read. I don't care much about |
| 36 |
flamewars, and might certainly have missed productive suggestions. |
| 37 |
|
| 38 |
At least now there is a good temporary measure, and we can now focus on |
| 39 |
how the keychain maintenance can be handled (for the final solution) |
| 40 |
|
| 41 |
-- |
| 42 |
Paul de Vrieze |
| 43 |
Gentoo Developer |
| 44 |
Mail: pauldv@g.o |
| 45 |
Homepage: http://www.devrieze.net |
Archives