On 01/11/2012 02:33 AM, M. David Peterson wrote:
On Tue, Jan 10, 2012 at 1:14 PM, Alvaro Lopez Ortega
<[email protected] <mailto:[email protected]>> wrote:
It's basically the same code, packed in a new tarball with a
regular version string.
Perfect! Builds and installs as expected, and all scripts and binaries
(cherokee > ~-worker, cherokee-admin (with desired options), and
cherokee-admin-launcher (without any options specified)) launch
without issue and thus far run exactly as expected on both OS X Lion
and Ubuntu Oneiric x64.
Question: Given the nature of what appears to be a massive
architectural redesign of the underlying cherokee core, should we
spend some time pulling together an ad-hoc build/test/deploy test
suite that can be easily cloned/used/contributed to by the community
to ensure we are able to find/report/squash any regressions that may
be introduced along the way on as many hardware and software platforms
as the community-at-large represents?
That would be a great thing to do. Actually, it is the sort of thing
that should have been scheduled as a very high-priority task if we had
enough resources.
I'm going to spend the next few months working on architectural
improvements. The target is to all fix the issues that have popped up
during the last couple of years, and to get the server architecture
ready for implementing the new technologies that seem to be gaining
acceptance among the Web-tech community.
All in all, I'm afraid I won't have enough bandwidth to work on QA while
I'm doing all the architecture changes, so it'd be awesome if you could
take over the QA effort.
Our current QA bench is composed by almost 300 functional tests. It has
proven to be a really useful tool worth investing work on. Of course
there is room for improvement there, but it is a fairly acceptable
functional test bench.
However, we do miss all kinds of Integration and User Acceptance tests.
Problems like the "A1" trailing bug in the first alpha of the 1.3.0
series should have been detected by those tests. They could cover:
compilation, cross-compilation, system integration (rrdtool support, log
files, etc) and even the proper function of cherokee-admin.
It is neither a trivial task, nor something that can be implemented over
night. However, if done properly, it could have a HUGE positive impact
on the project. I'm well aware of this, so if you finally take over the
QA effort, I'll do my best to be involved with it and help as much as
possible.
That's actually a project I'd be happy to head-up and run with in the
background while you continue to kick-a$$ coding up all the backend
goodness that will undoubtedly result of your redesign work. Let me
know if you like the idea and I'll start pulling things together and
pushing it forward if you do. :)
It'd be absolutely fantastic :-)
All the best,
Alvaro
_______________________________________________
Cherokee mailing list
[email protected]
http://lists.octality.com/listinfo/cherokee
