autoqa-devel September 2010

autoqa-devel@lists.fedorahosted.org

8 participants
88 discussions

Tests storing results into the ResultsDB - first spin
by Josef Skladanka 07 Sep '10

07 Sep '10

Hello, gang, I've created a patch, which allows AutoQA tests to store results in resultsdb. Because of the quite nice design provided by the AutoQATest base class, it's only new files, and few changes in the base class (23 lines to be exact) :) Not that I'd want to push it to master or so - just happy that it's working and the changes were as simple as i hoped :) Either have a look at the jskladan branch in git, or at the attached patch. Joza

2 2

Re: New branch - multihook_tests
by Kamil Paral 07 Sep '10

07 Sep '10

----- "James Laska" <jlaska(a)redhat.com> wrote: > > Thanks Kamil, > > I'm pulling from master for the new 0.4 content. As long as this is > in > a different branch until it's cleaned up, I think we'll be good. > It's already pushed into master. I would recommend basing the new release on 5251bdccebb9a9612f19fa008276ebc5a07ae29c or wait a few days until I provide few patches. Thanks, Kamil

1 0

New branch - multihook_tests
by Vojtěch Aschenbrenner 07 Sep '10

07 Sep '10

Hiya autoqa friends, I created a new git branch named 'multihook_tests'. There are some modified tests which are now able to run with both post-koji-build and post-bodhi-update hooks. It's first dirty version, but it should be function (it's tested with a few packages).

4 3

[AutoQA] #203: make rpmguard run as a post-bodhi-update test too
by fedora-badges 06 Sep '10

06 Sep '10

#203: make rpmguard run as a post-bodhi-update test too --------------------+------------------------------------------------------- Reporter: wwoods | Owner: Type: task | Status: new Priority: major | Milestone: Multi-hook tests Component: tests | Version: 1.0 Keywords: | --------------------+------------------------------------------------------- Using the multihook capability described elsewhere, make rpmguard run for the post-bodhi-update hook. -- Ticket URL: <https://fedorahosted.org/autoqa/ticket/203> AutoQA <http://autoqa.fedorahosted.org> Automated QA project

1 5

AutoQA 0.4.0 update
by Will Woods 03 Sep '10

03 Sep '10

So! We've made some impressive progress on autoqa in recent weeks, and (as discussed in the QA meeting earlier this week) it'd be a good idea to push a new package into production. We agreed in the meeting that current master is ready to be blessed as 0.4.0 - which means the following pending patches will be queued for 0.4.1 or later: - multihook_tests branch - ResultsDB support (in jskladan's branch) - depcheck test (in depcheck branch) - upgradepath polish patch I believe jlaska is working on the autoqa-0.4.0-1 build, which should appear soon. Thanks for all your hard work, everyone! -w

2 1

Re: [PATCH] Polished upgradepath code - minor fixes.
by Josef Skladanka 03 Sep '10

03 Sep '10

It's just that __init__ should be the place to create instance-variables - nothing more, nothing less. I don't have any problem with creating these variables in the run_once, I just made the first change according to my customs. Feel free to change it, and apply to master. Thanks Joza ----- Original Message ----- From: "Will Woods" <wwoods(a)redhat.com> To: "List for development discussion of the AutoQA project" <autoqa-devel(a)lists.fedorahosted.org> Sent: Tuesday, August 31, 2010 6:28:44 PM GMT +01:00 Amsterdam / Berlin / Bern / Rome / Stockholm / Vienna Subject: Re: [PATCH] Polished upgradepath code - minor fixes. This patch looks OK, with one comment: On Tue, 2010-08-31 at 13:07 +0200, Josef Skladanka wrote: > - def initialize(self, config): > - self.config = config_loader(config, self.tmpdir) > - #URL of logs/results stored on autotest-server > - self.autotest_url = autoqa.util.make_autotest_url(self.config) > - > - def setup(self): > - pass > + def __init__(self, *args, **kwargs): > + super(upgradepath, self).__init__(*args, **kwargs) > + self.log = [] > + self.envr_list = set() This is kind of confusing - is there a reason you need to use __init__() and not initialize(), as suggested by the test template? For example: def initialize(self, *args, **kwargs): super(upgradepath, self).initialize(config) self.log = [] self.envr_list = set() Even simpler - why not just set self.log and self.envr_list at the top of run_once()? Then you don't have to worry about running super(...) at all. -w _______________________________________________ autoqa-devel mailing list autoqa-devel(a)lists.fedorahosted.org https://fedorahosted.org/mailman/listinfo/autoqa-devel

2 1

[PATCH] be more verbose when autotest scheduling fails
by Kamil Paral 03 Sep '10

03 Sep '10

--- autoqa | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/autoqa b/autoqa index 8fd7dee..e0848ea 100755 --- a/autoqa +++ b/autoqa @@ -128,7 +128,7 @@ def maybe_call(cmdlist, dryrun=False, verbose=False): try: return call(cmdlist) except OSError, e: - print "Command failed: %s" % str(e) + print "Error: Command failed: %s\n %s" % (' '.join(cmdlist), e) return None def schedule_job(controlfile, required_arch=None, email=None, name=None, dryrun=False, labels=[]): -- 1.7.2.2

2 1

[AutoQA] #222: rpmguard: check for dropped arch on multilib updates
by fedora-badges 02 Sep '10

02 Sep '10

#222: rpmguard: check for dropped arch on multilib updates --------------------+------------------------------------------------------- Reporter: wwoods | Owner: Type: task | Status: new Priority: major | Milestone: Package Update Acceptance Test Plan Component: tests | Version: 1.0 Keywords: | --------------------+------------------------------------------------------- To properly prevent the `nss-softokn` problem from recurring (see ticket #201), we need to ensure that new multilib updates do ''not'' mistakenly drop one of their arches. -- Ticket URL: <https://fedorahosted.org/autoqa/ticket/222> AutoQA <http://autoqa.fedorahosted.org> Automated QA project

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

autoqa-devel September 2010