I'm starting to see koschei build failures, but no particular info as to why it failed. See: https://kojipkgs.fedoraproject.org/work/tasks/9874/11749874/root.log
DEBUG util.py:393: Running transaction DEBUG util.py:515: Child return code was: 1 DEBUG util.py:172: kill orphans
On Sat, 7 Nov 2015 19:55:56 -0700 Orion Poplawski orion@cora.nwra.com wrote:
I'm starting to see koschei build failures, but no particular info as to why it failed. See: https://kojipkgs.fedoraproject.org/work/tasks/9874/11749874/root.log
DEBUG util.py:393: Running transaction DEBUG util.py:515: Child return code was: 1 DEBUG util.py:172: kill orphans
please please please give a link to the top level task?
it's anoying to have to track back to the top level.
So, this failed on buildvm-08. I moved 01-09 to fedora 23 eariler today, so thats likely related.
I'll see if I can figure it out, and if not take them back out of rotation.
kevin
On 11/07/2015 08:15 PM, Kevin Fenzi wrote:
On Sat, 7 Nov 2015 19:55:56 -0700 Orion Poplawski orion@cora.nwra.com wrote:
I'm starting to see koschei build failures, but no particular info as to why it failed. See: https://kojipkgs.fedoraproject.org/work/tasks/9874/11749874/root.log
DEBUG util.py:393: Running transaction DEBUG util.py:515: Child return code was: 1 DEBUG util.py:172: kill orphans
please please please give a link to the top level task?
it's anoying to have to track back to the top level.
So, this failed on buildvm-08. I moved 01-09 to fedora 23 eariler today, so thats likely related.
I'll see if I can figure it out, and if not take them back out of rotation.
kevin
Apologies - here's a koschei example that links to the tasks:
https://apps.fedoraproject.org/koschei/package/plplot
https://apps.fedoraproject.org/koschei/package/octave
octave went from failing on buildvm-07 to working on buildvm-16.phx2
On Sat, 7 Nov 2015 20:19:26 -0700 Orion Poplawski orion@cora.nwra.com wrote:
On 11/07/2015 08:15 PM, Kevin Fenzi wrote:
On Sat, 7 Nov 2015 19:55:56 -0700 Orion Poplawski orion@cora.nwra.com wrote:
I'm starting to see koschei build failures, but no particular info as to why it failed. See: https://kojipkgs.fedoraproject.org/work/tasks/9874/11749874/root.log
DEBUG util.py:393: Running transaction DEBUG util.py:515: Child return code was: 1 DEBUG util.py:172: kill orphans
please please please give a link to the top level task?
it's anoying to have to track back to the top level.
So, this failed on buildvm-08. I moved 01-09 to fedora 23 eariler today, so thats likely related.
I'll see if I can figure it out, and if not take them back out of rotation.
kevin
Apologies - here's a koschei example that links to the tasks:
https://apps.fedoraproject.org/koschei/package/plplot
https://apps.fedoraproject.org/koschei/package/octave
octave went from failing on buildvm-07 to working on buildvm-16.phx2
Yeah, unfortunately, there's not much in the log... it's running the transation with the dnf builddep results and then says 'exit 1' but doesn't say why. ;(
kevin
ok. I think this might be: https://bugzilla.redhat.com/show_bug.cgi?id=1264198 which is an rpm bug fixed in rawhide, but not yet in f23.
We need a f23 update with that fixed and we can try again.
kevin
On Sun, 8 Nov 2015 13:40:16 -0700 Kevin Fenzi kevin@scrye.com wrote:
ok. I think this might be: https://bugzilla.redhat.com/show_bug.cgi?id=1264198 which is an rpm bug fixed in rawhide, but not yet in f23.
We need a f23 update with that fixed and we can try again.
There's now a f23 update, I have applied it to buildvm-01 to 09 and re-added them.
Please let me know if you see any build failures on these builders that cannot be duplicated on others.
kevin
On Mon, 9 Nov 2015 12:19:18 -0700 Kevin Fenzi kevin@scrye.com wrote:
On Sun, 8 Nov 2015 13:40:16 -0700 Kevin Fenzi kevin@scrye.com wrote:
ok. I think this might be: https://bugzilla.redhat.com/show_bug.cgi?id=1264198 which is an rpm bug fixed in rawhide, but not yet in f23.
We need a f23 update with that fixed and we can try again.
There's now a f23 update, I have applied it to buildvm-01 to 09 and re-added them.
Please let me know if you see any build failures on these builders that cannot be duplicated on others.
I now have moved:
buildvm* arm02* arm04* buildhw*
over to being Fedora 23.
Next week I will move the bkernel builders over and hopefully we will also retire the old buildppc instances in Favor of some new Fedora 23 ones soon. Then we should be 100% Fedora 23 builders.
One issue to note: dnf is not that verbose when running builddep. It will tell you if the package you needed cannot be installed, but it won't tell you that the reason is 5 more layers deep. ;( I've filed https://bugzilla.redhat.com/show_bug.cgi?id=1281596 for this.
kevin
kevin