| 1 |
On Tue, Jul 19, 2016 at 4:31 AM, Consus <consus@×××.com> wrote: |
| 2 |
|
| 3 |
> On 18:03 Fri 15 Jul, Robin H. Johnson wrote: |
| 4 |
> > Hi all, |
| 5 |
> > |
| 6 |
> > In tracing down problems with the git->rsync path, it has been noticed |
| 7 |
> > that some developers have significant clock drift on their local systems |
| 8 |
> > (up to one case of 14 days wrong), and it's potentially contributing to |
| 9 |
> > problems in generating the rsync tree. |
| 10 |
> > |
| 11 |
> > I have implemented a check as part of the hook that validates Git push |
| 12 |
> > certificates (require-signed-push). It looks for clock drift or an |
| 13 |
> > overly long push, and aborts if needed. |
| 14 |
> > |
| 15 |
> > The tolerances are presently set to: |
| 16 |
> > - 5 seconds of clock drift. |
| 17 |
> > - 'git push' must be completed in 60 seconds. |
| 18 |
> |
| 19 |
> AFAIR Windows AD servers allow 5 minutes clock drift. How about at least |
| 20 |
> a minute or so? |
| 21 |
> |
| 22 |
> |
| 23 |
I would also argue for 5 minutes. |
| 24 |
|
| 25 |
-A |
Archives