1 |
On Sunday 14 September 2014 15:40:06 Davide Pesavento wrote: |
2 |
> On Sun, Sep 14, 2014 at 2:03 PM, Michał Górny <mgorny@g.o> wrote: |
3 |
> > We have main developer repo where developers work & commit and are |
4 |
> > relatively happy. For every push into developer repo, automated magic |
5 |
> > thingie merges stuff into user sync repo and updates the metadata cache |
6 |
> > there. |
7 |
> |
8 |
> How long does the md5-cache regeneration process take? Are you sure it |
9 |
> will be able to keep up with the rate of pushes to the repo during |
10 |
> "peak hours"? If not, maybe we could use a time-based thing similar to |
11 |
> the current cvs->rsync synchronization. |
12 |
|
13 |
Best case only one package is affected - a few seconds |
14 |
Worst case someone touches an eclass like eutils, then it expands to something |
15 |
on the order of one or two CPU-hours. |
16 |
|
17 |
> Are we going to disallow merge commits and ask devs to rebase local |
18 |
> changes in order to keep the history "clean"? |
19 |
|
20 |
Is that going to be sane with our commit frequency? |