I have to use 2.4. Our servers are running RedHat Enterprise Linux 5.4
which ships py2.4 by default.
Yeah, RedHat Enterprise, our beloved museum...
I used to compile python2.6 which is better, but having to install
compilation tools on these servers is considered a security risk.
Have you
--
--
| Marcin Kasperski |(...) the only completion criterion for
| http://mekk.waw.pl | the Analysis and Design phases is - the date
Jonathan Vanasco jonat...@findmeon.com
writes:
What version of Routes do you have installed?
1.11
It definitely does not requires the leading slash in Routes-1.10
OK, I was not sure, someone on IRC suggested that the change was
introduced in 0.9.7. Still, it deserves some warning, I just
Is there a way to make Mako aware of beaker.cache.type setting inside Pylons?
Let me quote recent Mako 0.2.5 announcement:
-
Mako 0.2.5 is now released and available at
http://www.makotemplates.org/download.html .
The new feature in here (which I've been using for many months now,
In logs of my pylons app I spotted recently a few backtraces like the
one below. Not sure whether this is harmful, but maybe it is worth
a look?
2009-09-10_06:08:37.07570 08:08:37,075 WARNI [paste.httpserver.ThreadPool]
Thread 47284430936400 hung (working on task for 9610 seconds)
I was thinking about signing up with a web host that supports Pylons
(among many other things) and one of the differences between the
various plans is application memory for long-running processes. The
plan I'd like to sign up for has 80MB. Does anyone know if this is
enough for basic Pylons
master page here or there.
PS And it would be truly beautiful if there existed single pretty
PDF file which would allow one to print 'everything he needs...'.
--
--
| Marcin Kasperski | Communication takes place between people
WSGIDaemonProcess let you set the number
of dedicated processess...
--
--
| Marcin Kasperski | A process that is too complex will fail.
| http://mekk.waw.pl | (Booch
Attempt to easy_install pylons 1.9.7beta on windows ends during
simplejson 1.8.1 install with:
C:\Program Files\Microsoft Visual Studio .NET 2003\Vc7\bin\cl.exe /c /nologo /Ox
/MD /W3 /GX /DNDEBUG -Ic:\python25\include
-IC:\users\path-to-my-virtual-env\PC /Tcsimplejson/_speedups.c
Are there any chances to have this patched, in any - even temporary
and dirty - way, so easy_install-ed weberror does not crash? At the
moment I must manually patch it on every environment, that's really
troublesome.
--~--~-~--~~~---~--~~
You received this
and smoother.
PS I roughly did the aforementioned split, can post it if somebody is
interested. Not much testing so far though.
--
--
| Marcin Kasperski | Users typically do not know exactly what
| http://mekk.waw.pl | they want
PylonsApp seems to do two separate things:
- construct and register objects like request, response, session, g,
h, ...
- dispatch the request to correct controller (and wrap it with
testing support and error handling)
Couldn't those two be separated?
Why? Well, I would fairly like to
I have comments and docstrings in Polish in my (utf-8 encoded) sources.
--
--
| Marcin Kasperski | In any large change, 1/3 will think it is
| http://mekk.waw.pl | great, 1/3 will think it is stupid, and 1/3
to be adapted...
--
--
| Marcin Kasperski | Software is not released,
| http://mekk.waw.pl | it is allowed to escape
id=blahblah
...
/body
- in CSS
#header {
background: ...default-one...;
}
#blahblah #header {
background: ...specific-one...;
}
Of course you can also rewrite it to classes instead of ids.
--
--
| Marcin Kasperski
Damjan [EMAIL PROTECTED] writes:
AFAIK
you'll have to start httpd with
env - ORACLE_HOME=... /usr/sbin/httpd
... and add PassEnv ORACLE_HOME (or sth like that, haven't used it for
some time) to the apache configuration. Apache removes environment
variables for some reason...
http://wiki.pylonshq.com/display/pylonsmisc/Pylons+Cheatsheet
Minor remark: in mako templates I prefer
${ next.body() }
not
${ self.body() }
If you use only one-level inheritance then there is no difference, but
next is useful if you have sequence of templates (say main layout,
And I don't setup db conn on every requests. At least I think so.
sqlalchemy.default.echo_pool = True
--
--
| Marcin Kasperski | You have the right to produce quality work
| http://mekk.waw.pl | at all the times
application startup (and
save them in public directory just then).
Using caching reverse-proxy (and setting caching headers) is another
possibility.
--
--
| Marcin Kasperski | Working overtime sucks the spirit and
| http
.
--
--
| Marcin Kasperski | In any large change, 1/3 will think it is
| http://mekk.waw.pl | great, 1/3 will think it is stupid, and 1/3
|| will wait (Reed
...
--
--
| Marcin Kasperski | If we are to be successful, we must still have
| http://mekk.waw.pl |the courage to put our faith in people as
|| opposed to a process. (Booch,Martin,Newkirk
?
--
--
| Marcin Kasperski | You have the right to produce quality work
| http://mekk.waw.pl | at all the times. (Beck
the destination object using the
redirected-to route without generating any redirect to the browser)
as one of the methods.
--
--
| Marcin Kasperski | People resist change, regardless of what
| http://mekk.waw.pl
before .url)
--
--
| Marcin Kasperski | Software is not released,
| http://mekk.waw.pl | it is allowed to escape
for
pylons request need not be that good.
I am fairly happy using nginx as reverse proxy (and static content
handler) for pylons served by paste...
--
--
| Marcin Kasperski | The cost of a few uncorrected non-critical
| http
Actually, Ian clarified that the threadpool can be turned off. Add the
following to the [server] section:
use_threadpool = false
On my linux system, that cut the Virtual from 100Mb to 18Mb.
But this is then single-process mode, I guess?
* threadpool_workers = 10
virtual 108104, resident 18900
(...)
So it looks like it is all about thread-local storage. Maybe just
the thread stack size...
What figures do you get for the above if you run:
ulimit -s 512
*Far* smaller, for example with workers=10 I got virtual about
Any ideas how could I reduce pylons app RAM usage? At the moment
pylons process takes above 100MB (almost static site serving some
templates) - both when run with paste, and when run under mod_wsgi.
Quite a lot, considering that for example whole moinmoin manages to
work on ~25MB.
Short
Alberto Valverde [EMAIL PROTECTED] writes:
Thanks for the summary Martin!
Good old habit from times when I participated in tru64-unix-managers.
On that list the initial poster was always expected to create
'solution' post with the summary info, at least in case reply was not
trivial. I must
Just a (post-mortem?) note. After having very good experience from
migration (on some non-pylons site) from moinmoin hosted on apache
with fastcgi to moinmoin on twisted proxied behind nginx (I did not
make any formal measures, but it just feels that the app runs faster)
I opted for similar path
[server:main]
threadpool_nworkers = 5
did not work for me.
Resolved. One should use
threadpool_workers = 5
(without 'n'). Then it works. Some observations in the next post.
--~--~-~--~~~---~--~~
You received this message because you are
Ben Bangert [EMAIL PROTECTED] writes:
On Oct 31, 2007, at 8:39 AM, Marcin Kasperski wrote:
I am still not very happy with paster (~100MB virtual, 22MB resident),
and hoping to search for some improvements here. But at least it is
just one such process...
Can you ask your ISP
you should see if the Apache configuration you are using
defines a value for the ThreadStackSize directive.
Thanks for the hint.
I am still to make my experiments, but here is initial summary for
this thread from my point of view.
It is highly suspected, that significant part of the RAM
I've run a few tests on my OSX machine, can you run them on your box?
Could you share the [EMAIL PROTECTED] of test apps? I wouldn't like to
experiment too much with creating them
Cheap VPS's (like serveraxis.com) start around $30/mth with half
a gig of ram, and a gig of swap...
Well, I
Are the Zope and Moin numbers from the same environment your Pylons
is running on?
Yes. [EMAIL PROTECTED] in both cases.
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
pylons-discuss group.
To post to this
Any ideas how could I reduce pylons app RAM usage? At the moment
pylons process takes above 100MB (almost static site serving some
templates) - both when run with paste, and when run under mod_wsgi.
Quite a lot, considering that for example whole moinmoin manages to
work
Well. This works nice:
static_app = StaticURLParser(config['pylons.paths']['static_files'],
cache_max_age = 3600)
But ... here one can not pass such a param:
javascripts_app = StaticJavascripts()
After noticing that:
class
Any ideas how could I reduce pylons app RAM usage? At the moment
pylons process takes above 100MB (almost static site serving some
templates) - both when run with paste, and when run under mod_wsgi.
Quite a lot, considering that for example whole moinmoin manages to
work on ~25MB.
Well, I am just making my first experiments with pylons deployment,
and while trying to 'run egg' I immediately got an error like:
Sat Oct 13 22:25:56 2007] [error] [client 127.0.0.1] OSError: [Errno 20] Not a
directory:
a id='MyLink' title=Blah blah href=...Some link/a
shows fairly nice floating text when the mouse is over the link.
Is it possible to (using reasonably small javascript code and
utilizing pylons helpers) load the title AJAXically in onMouseOver?
I tried using sth like
${h.link_to(uBlah blah,
plus Twisted's Deferreds and callbacks are enough to
make many other programmers run screaming from the room.
Heya, they are beautiful ;-) Especially with the @deferredGenerator
and @inlineCallback syntaxes. This is pretty:
@deferredGenerator
def thingummy():
thing =
How would you tune pylons application to reduce its RAM
usage as much as possible? I mean the case of rather-small-application
with rather-small-trafic, which is expected to run on machine
with 64 or 96MB RAM (sideways some other apps).
(*) In case somebody is curious: not, not embedded system.
Has anybody tried merging pylons techniques with twisted?
In general it should be possible as twisted contains WSGI
container...
Purpose? Well, I have some twisted app (some works over the network)
which would benefit from additional web interface. And would
be nice if this could be colocated
Indeed, in Routes 2.0 the behavior you expect will be the default, I
blogged more about other Routes 2.0 and 1.X features:
http://groovie.org/articles/2007/08/29/routes-planning-and-the-road-to-routes-2-0
Cool!
Apart from clearing name route confusion, I noticed you plan to solve
another
(after additional reconsideration)
I think that the - with respect to the current behaviour - whole
term 'named route' and mapper syntax is misleading.
Let's once more take a look at the original example:
from routes import *
m = Mapper()
m.connect('gallery_thumb',
Ben Bangert [EMAIL PROTECTED] writes:
On Aug 24, 2007, at 2:56 AM, Dan Korostelev wrote:
Yeah, I also worked around the problem this way, but I don't like
it. Looks like there's a bug in Routes. Thanks for information.
There's no bug in routes, this is exactly how it works, and how the
m.connect('gallery_thumb',
'images/gallery/:(image_id)_:(image_size).jpg', image_size='thumbnail')
m.connect('gallery', 'images/gallery/:(image_id).jpg')
Well, that's generally the same trick as discussed earlier as a
workaround - forcing parameter difference (here - extra param). I do
from routes import *
m = Mapper()
m.connect('gallery_thumb', 'images/gallery/:(image_id)_thumbnail.jpg')
m.connect('gallery', 'images/gallery/:(image_id).jpg')
url_for('gallery_thumb', image_id=1)
'/images/gallery/1_thumbnail.jpg'
url_for('gallery', image_id=1)
The problem with pyLucene in my eyes is that it uses Jython (...)
And the memory usage of my apache process just blasted up so
much,
Thanks for the warning. For me it more or less kills this path.
You may take also a look into Xapian,
I will take a look. IIRC moinmoin is to be (or maybe
1) h.url_for is a great idiom to generate urls to dynamic
pages. But what should be used to generate links to
*static* resources? (...)
(...)
map.connect(attachments, /attachments/:rec_id/:(filename).jpg
h.url_for(attachments, rec_id=5, filename=foo)
= /attachments/5/foo.jpg
The context: I have some textual pages, rendered by template
controller.
Apart from the actual text, those pages have some 'metadata-level'
content. For example - page title, date of the last change,
short summary. This information is used within page - in standard
mako inheritance way (base
Just a few questions about pylons programming style.
1) h.url_for is a great idiom to generate urls to dynamic
pages. But what should be used to generate links to
*static* resources? There is h.stylesheet_link_tag,
but what should I use to generate links to static HTML,
images,
A lot of answers already, so just two points:
a) There are plenty of cheap VPS hosting options (you get
your own virtual machine where you have root and can do whatever
you like). See for example http://linode.com/ (but maybe
some Xen-based provider can be better as it offers better
53 matches
Mail list logo
