Switch branches/tags
Nothing to show
Find file
Fetching contributors…
Cannot retrieve contributors at this time
58 lines (41 sloc) 2.26 KB
The deployment world changes quickly, and I'm trying to keep these
examples up-to-date with recent changes. To that end, I'll explain
here all the changes I make as I update the tutorial.
March 2011 (PyCon)
A pretty major re-work based on experience doing this class over the last
year. The outline's different, and a few of the technologies themselves have
changed/improved. There's more of a focus on the web servers themselves (as
oposed to database servers and such) and more expository material (as opposed
to demos). The big changes are:
* Switched from Fumblerooski to Mingus as the example application. The need
for a data file to go with Fumblerooski and the additional overhead of the
needed Buildout config makes this a bit too much.
* Because of the above switch, I've dropped Buildout in favor of
* Added coverage of Gunicorn, though there's no config because Gunicorn's just
that easy.
* Memcached and Pgpool get much briefer references, though I've kept the
configs around for those who want to see 'em.
* Added some very basic Chef/Puppet coverage as well as an example Chef config
for the mini-cluster built in the example. I don't yet have the equivalent
Puppet setup because I'm a few years out of date in my Puppet knowledge. I'd
appreciate it if someone could contribute an equivalent to contrast/compare.
* Added Vagrant config for using local VMs.
July 2010 (OSCON)
* Updated to Ubuntu 10.04 LTS.
Whenever possible, I prefer to deploy against Ubuntu's LTS (long-term
support) releases. However, when I originally wrote this tutorial (Feb
2010) the previous LTS was exceedingly long in the tooth and missing most
of the modern Python kit.
Now that 10.04 is out along with all the modern Python goodness it'll
probably remain my preferred server OS for a couple-three years.
* Dropped the database replication example.
PostgreSQL 9.0 ships soon, and will radically change PostgreSQL's
replication/clustering story. Also, in previous classes this was the
least "interesting" part according to attendees. So, it's been dropped so
that I can spend more time on the other parts. Instead, I'll cover
replication as part of the "what's next" roundup at the end.