----- "James Laska" jlaska@redhat.com wrote:
On Fri, 2010-08-20 at 17:42 +0200, Kamil Páral wrote:
lib/python/koji_utils.py | 51
++++++++++++++++++++-------------------------
1 files changed, 23 insertions(+), 28 deletions(-)
diff --git a/lib/python/koji_utils.py b/lib/python/koji_utils.py index f8ac077..87b32e9 100644 --- a/lib/python/koji_utils.py +++ b/lib/python/koji_utils.py
<snip>
Thanks Kamil,
As we discussed in the Fedora QA meeting today. I have applied this patch to the running AutoQA instance and rescheduled several previously failed rpmguard jobs. I consider the results a PASS, but welcome any additional tests you'd like me to run.
First, a package that does *not* have previous builds ... gtk-nodoka-engine-0.7.5-1.fc15 * Without the patch, the job (#45485) failed (see http://fpaste.org/cOJh/). * With the patch, the job (#45506) runs to completion, but is not able to find a previous build to compare against in dist-f15. This is accurate, looking in 'koji' I cannot find another dist-f15 build to compare against
(http://koji.fedoraproject.org/koji/packageinfo?packageID=4871).
N: There is no previous build of gtk-nodoka-engine-0.7.5-1.fc15 in parents of dist-f15 tag.Next, a package that does have previous builds ... schroot-1.4.2-5.fc14 * Without patch - rpmguard result -
https://fedorahosted.org/pipermail/autoqa-results/2010-July/028449.html * With patch - rpmguard result -
https://fedorahosted.org/pipermail/autoqa-results/2010-August/030026.html
The results match, and the confirm what I'd expect to see based on koji builds (http://koji.fedoraproject.org/koji/packageinfo?packageID=8929).
Yes, I also believe that the behaviour is correct now. Thanks for testing it, James, and thanks for pushing it, Will. Closing ticket #217.
autoqa-devel@lists.fedorahosted.org
-
Kamil Paral