8,376 karma · joined October 9, 2008
You, and all of Canada are welcome!
Again: I'm sorry
> Why does Rackspace's API use "Keystone" auth instead of just API tokens or Oauth like everyone else? We would have to create a special Rackspace auth system just to support Rackspace with our service. Everyone else in the world uses Oauth or API tokens, or passwords which works a lot like API tokens, but Rackspace uses this crazy system nobody else in the world uses that requires refreshing tokens.
Actually, Token-based auth is quite common, as well as token refreshes. This is pretty standard for cloud provider APIs for compute, storage, etc. Amazon, Rackspace, Google and others all use the account -> API Key model. This actually uses API tokens which is the industry and OpenStack standard.
For OAuth - we've discussed this, there is a cloud partner in the cloud tools market place which does this. This isn't the first time I've gotten this request, so I'm going to pass it to the product teams as well.
We (rackspace) have had great success with Fog, jclouds, pkgcloud, libcloud, and other multi-cloud toolkits. For openstack / Rackspace cloud we also have php-opencloud, openstack.net and pyrax (which uses the python-XXXX clients from openstack core).
It definitely is a fine line to walk; a good abstraction though means that you can use the most common features easily and float through providers, but you have access to the advanced features if you need or want them.
Commonly those advanced features (on a provider level) mean a greater level of vendor lock-in which is why they may not make sense to expose in the higher level abstraction.
The LCD issue is being resolve in libcloud - we have multiple FT devs working on it right now, and does not come up as an issue with Fog, jclouds, or pkgcloud. It's possible to make an abstraction that is simple enough to start but powerful enough to access the advanced features of each individual cloud host.
I hope users and consumers will keep us honest - as we said in the post, our interest is pretty much no lock-in on an API level (hence: openstack) and no lock-in on the code level. We really want to make tools that allow our users and others to fluidly move between hosts.
Yes, it is currently focused on openstack clouds, but the goal is to reach parity with Fog (ruby), jclouds (Java), Libcloud (Python), pkgcloud (node.js), and so on.
The focus is a series of clean abstractions for users to be able to build tools and programs which can talk across any cloud provider easily using a common set of names / abstractions.
We have it on github already: https://github.com/rackspace/gophercloud
Sorry I can't give you a specific date, but I don't want to set unreasonable expectations.
Sooner or later in any programmers life you're going to reach a point where it seems / feels easier to start with less. You start reaching for tools that feel more like the unix way - they do one thing very well, and allow you greater compose-ability and flexibility.
On the other hand, I have to fight this urge: because much like Python - the language that comes with community included, you sort of don't want to be the one person using something, or understanding what you did.
Your best bet is to find something that provides the simplicity you find yourself looking for, but that still has the greater community support you and others can lean on, and existing domain knowledge.
For example - let's say you want to go with a micro framework: look at bottle, web.py and my favorite: Flask. The nice thing with Flask for example is it stays out of my way, still has a ton of plugins (so, for example, I don't have to write my own damned Oauth system) and a vibrant community.
That ache in your gut is looking for things that get rid of boilerplate and get out of your way to just write awesome stuff. If you don't feel like the bigger frameworks like django (which I love for larger projects) get out of your way enough: don't drop down to raw WSGI, go spend some time with something smaller!
As for the rest - yes. The communications and deprecation of python-cloudfiles was poorly handled. The crux of what happened is that python-cloudfiles was a part time project and ultimately superseded by the python-swiftclient work in OpenStack.
We have requirements that for an SDK / tool to be officially supported, we have dedicated full time headcount to support and maintain the package(s) - the decision was made to proceed with Pyrax which leverages python-swiftclient and other openstack tools under the covers as the official Python SDK for Rackspace.
This was not made clear to the users, or the community by any parties involved, and for that I apologize. We're continually working on better upstream communication/collaboration and this series of events (you're not the first to give us this feedback) has made us take a very hard look internally on how this is all managed and communicated.
The good news is that Ed - our current Pyrax SDK maintainer is full time and dedicated to working with you and others on pyrax - this means the Python 2.6 support will be taken care of.
Tangentially: Yes - python-novaclient and python-swiftclient from OpenStack will work on Rackspace Cloud - you can actually look at the Pyrax source to see that it uses these under the covers. Rackspace runs OpenStack with as few differences as possible so OpenStack tools should work great on Rackspace.
So, in short: I apologize. The communications and management of this to upstream (the community) were poor, and we are renewing all of our efforts to focus on upstream first.
Jesse