On 07/15/2015 04:51 PM, Pierre-Yves Chibon wrote:
On Wed, Jul 15, 2015 at 04:07:31PM +0200, Ralf Corsepius wrote:
> Hi
>
> I am facing a bizarre f23-koji build breakdown:
>
http://koji.fedoraproject.org/koji/taskinfo?taskID=10366596
>
> The interesting part seems to be hidden in this file:
>
https://kojipkgs.fedoraproject.org//work/tasks/6598/10366598/checkout.log:
>
> $ git clone -n
git://pkgs.fedoraproject.org/ht
> /var/lib/mock/f23-build-3655660-501262/root/tmp/scmroot/ht
> Cloning into '/var/lib/mock/f23-build-3655660-501262/root/tmp/scmroot/ht'...
> fatal: read error: Connection reset by peer
>
>
> Same result, when retrying:
>
http://koji.fedoraproject.org/koji/taskinfo?taskID=10366729
>
> An f24-build succeeded ca. 1/2 hour ago:
>
http://koji.fedoraproject.org/koji/taskinfo?taskID=10366109
We had a weird SELinux issue that I fixed this morning and that has apparently
surfaced again this afternoon.
So I re-labeled the tree again and it should be working again.
I don't know, whether this SELinux issue was the culprit ;)
All I can say, another 3rd build attempt (ca. yet another 1/2 hour
later) succeeded:
http://koji.fedoraproject.org/koji/taskinfo?taskID=10367098
Let us know if that's not the case
Well, ... I really
don't know the cause.
Ralf