Allan Saddi's projects blog Comments

James: flup 1.0.2 released

nospam@example.com (James) — Wed, 10 Jun 2009 10:02:29 -0700

Hi,

I'm forced to use the prefork method in my django-fastcgi configuration as when using the threaded method, the django fastcgi process seg-faults - again, it works fine in prefork.

When I say it seg-faults, let me explain:

python manage.py runfcgi host=127.0.0.1 port=8001 method=threaded daemonize=false debug=true

is the django fastcgi startup comment I'm testing with. If I then run a benchmark test to load the server like so:

ab -n1000 -c15 -H 'Host: www.example.com' http://127.0.0.1/

With nginx running on port 80 in a fastcgi configuration to port 8001, a few requests will go through successfully, but then eventually the fastcgi process segfaults. No stack trace.

I'm hoping to use the threaded model in the future as it should use less memory than the prefork. Anyhow, the only reason I can think of as to why this is bugging out is to do with threading issues to do with the Python GIL (http://docs.python.org/c-api/init.html#thread-state-and-the-global-interpreter-lock)

Thoughts? Can I provide further information to you? Feel free to email me. Thank you!

James

amittiwari: flup 1.0.2 released

nospam@example.com (amittiwari) — Tue, 09 Jun 2009 00:36:50 -0700

marketing

Jon Nelson: flup 1.0.1 released

nospam@example.com (Jon Nelson) — Thu, 21 May 2009 11:38:04 -0700

I just saw in mercurial that you have done this.
Fantastic!

I'll try to give this a test soon.

I'm really looking forward to the next release!

Allan Saddi: flup 1.0.1 released

nospam@example.com (Allan Saddi) — Mon, 18 May 2009 15:35:31 -0700

I don't really remember the rationale for deprecating the factories in Paste. Though I guess it makes more sense if they were packaged with flup.

A problem is that flup's paste_factory is a one-size-fits-all deal. (And it doesn't really fit all...)

I guess I might look into pulling Paste's factories into flup. This may cause some issues, mainly with regards to defaults.

Jon: flup 1.0.1 released

nospam@example.com (Jon) — Mon, 18 May 2009 14:56:16 -0700

Another problem is that PasteScript is deprecating the use of this style of factory:

use = egg:PasteScript#flup_fcgi_thread

However, if one does this:

use = egg:flup#fcgi
(or egg:flup#fcgi_thread, they are the same)

PasteScript#flup_fcgi_thread accepts any of

host, port, and socket
(and a few others that aren't relevant here).

If you don't specify them, then bindAddress=None is used.
There is no way to replicate bindAddress=None (that I can see) using flup#fcgi, a problem for any app that relies on that behavior.

Can you think of a reasonable fix?
Also, when might the next flup release be?

Timothy Farrell: Dabbling in Python 3.0

nospam@example.com (Timothy Farrell) — Tue, 10 Mar 2009 14:23:44 -0700

Just so you know, I'm working on a native Python 3 framework (yet to see the public eye). I've tested it with your FCGI adapter and have had success with both Apache on Linux (via pipes) and LightTPd on Windows (via FastCGI server).

I don't have file upload support in yet and there is a bit of wonkiness in cgi.FieldStorage.

Anyway, thanks for the fine adapter!

v+python: flup 1.0.1 released

nospam@example.com (v+python) — Sun, 30 Nov 2008 23:54:51 -0800

flup seems to depend on setuptools at http://cheeseshop.python.org/packages/2.6/s/setuptools/setuptools-0.6c6-py2.6.egg

Now it is at
http://cheeseshop.python.org/packages/2.6/s/setuptools/setuptools-0.6c9-py2.6.egg

the c6 version doesn't exist, the c9 version is the only one there. Renaming the c9 version to c6 and placing it in the flup directory seems to work, but could confuse or delay people.

Dan Roberts: flup 1.0.1 released

nospam@example.com (Dan Roberts) — Thu, 24 Jul 2008 02:28:24 -0700

Oh, I didn't realize there was a factory available with paste... Well at least there's now a version in trunk that works

Thanks for the speedy response

Allan Saddi: flup 1.0.1 released

nospam@example.com (Allan Saddi) — Wed, 23 Jul 2008 22:57:00 -0700

Thanks for the suggestion. It looks like the original flup factories (bundled with Paste) supported the 'socket' option. I didn't realize that feature was lost.

I went ahead and committed your change, but I have no idea when I'll be making the next release (I knew I was a bit hasty with yesterday's release). But the latest is always available at http://hg.saddi.com/flup-server/

And as for trac, the username/password is trac/trac1234. I should make it more prominent.

Thanks again.

Dan Roberts: flup 1.0.1 released

nospam@example.com (Dan Roberts) — Wed, 23 Jul 2008 21:45:57 -0700

Sorry, the comment system ate my leading spaces, but i'm pretty sure the meaning is obvious

Dan Roberts: flup 1.0.1 released

nospam@example.com (Dan Roberts) — Wed, 23 Jul 2008 20:37:42 -0700

First I'd like to say that flup is awesome! However, I'm having a little bit of trouble...

I can't seem to get flup.server.fcgi to work with a unix domain socket. After investigating the problem a bit, i found this file:

flup/server/paste_factory.py

which (for me, as a turbogears user) was actually controlling the creation of flup.server.fcgi I noticed that no matter what, bindAddress is set to a tuple (therefore eliminating the ability to run a server on a unix domain socket)

The simple fix that I employed was to add a conditional to line 6

if not 'bindAddress' in local_config:

which would make it so that the tuple is only created if bindAddress is empty.

Alternatively, you could provide a config value "socket"

and do something like

if 'socket' in local_config:
local_config['bindAddress'] = local_config['socket']
del local_config['socket'] #no sense storing it twice, plus socket could cause a name collision
else:
local_conf['bindAddress'] = (host, int(port))

**** I prefer this second method as it provides a more consistent configuration value to the user

Also sorry, I would have posted a trac ticket, but I don't see a way to sign up to do so...

Anyways great work! There's my two cents.

cheers
-Dan

owen: EPIPE no longer ignored

nospam@example.com (owen) — Wed, 11 Jun 2008 12:56:54 -0700

perdon por el idioma.
como consigues el .scgi?

Erwin: flup 1.0 released

nospam@example.com (Erwin) — Tue, 25 Mar 2008 04:16:35 -0700

Thank's for the response. I did upgrade from Beaker-0.5 to Beaker-0.9.1 and now it works. Thank's again for helping me.

Allan Saddi: flup 1.0 released

nospam@example.com (Allan Saddi) — Mon, 24 Mar 2008 14:14:37 -0700

I use Beaker and ajp-wsgi pretty extensively and I haven't run into any problems like you've described. The frameworks I use with the two are Pylons, Flannel, or sometimes no framework at all (e.g. an authentication middleware component that just requires sessions).

The issue could be with web.py or with mod_jk, as these are the 2 major differences between our configurations. (I use Apache 2.2's mod_proxy_ajp.) Though I don't see how mod_jk could really affect this.

Have you tried web.py + Beaker on another WSGI server? I'm a bit suspicious about the way session.save() is being called in web.py. And like you mentioned, the non-web.py example worked fine.

Erwin: flup 1.0 released

nospam@example.com (Erwin) — Mon, 24 Mar 2008 02:25:08 -0700

Beaker didn't work for me in conjunction with ajp-wsgi. I got the same session-id for different browsers. Can you give me an example for Beaker Session in conjunction with ajp-wsgi.

I'have documented my experiences at http://www.xhostplus.at/static/webpy-ajp-wsgi.html.