Comments on: Daily automated Mozmill test-runs in the QA Lab

By: Henrik Skupin

Henrik Skupin — Sun, 31 Jan 2010 20:16:06 +0000

Setting up such a machine isn’t so complicated. It’s easier now with the latest Mozmill release. The main problem are the tests itself. Nearly all the tests we have are browser chrome tests and have been written for testing the Firefox ui. So you would clearly have to spend a lot of time in updating those. But for chrome tests it should be easier…

By: Robert Kaiser

Robert Kaiser — Sat, 30 Jan 2010 23:44:23 +0000

I somehow wonder if I would buy a mini myself, what the possibilities would be to get it set up the same way and run the same stuff but for SeaMonkey.
One problem is that I don’t have licenses for any of those other OSes, another is getting everything set up – and even one more to get all the tests written/modified to be run.
A way to test our updates would be really cool though…

By: Henrik Skupin

Henrik Skupin — Fri, 29 Jan 2010 13:45:55 +0000

For now we would have to manually scan those reports. There is no automatism behind to inform us about failures. But it’s something we wanna check in the future. It’s a great idea in any way. So one of the first things I have to check is what’s possible with the Python view server. Then we can have a rich web interface.

By: Aaron Train

Aaron Train — Fri, 29 Jan 2010 04:26:30 +0000

Thanks for posting this. I was curious as to how it was all coming together the past few months, now I have a better understanding.

What currently is the procedure if a test run encounters some failed tests, are you notified/is it recorded, or is this on a glance check only?

Yes, in the soon future it looks like the web interface will need the ability to query up previous test runs as well as sort by failures.

- AaronMT