Title: Anybody made a memcached SessionStore?
We're thinking about creating a memcached SessionStore backend, but (being quite lazy and not as smart as most other people) I'd much rather copy or adapt somebody else's code. Has anyone done this already? Any advice before I start in on this
: Ben Parker
[mailto:[EMAIL PROTECTED] Sent: Tuesday, October 25, 2005 6:22
PMTo: Hancock, David (DHANCOCK)Cc:
webware-discuss@lists.sourceforge.netSubject: Re: [Webware-discuss]
Anybody made a memcached SessionStore?Hancock, David
(DHANCOCK) wrote:
We're thinking about creating
Title: Question about FormEncode
Much much progress today with ZPTs and FormEncode. Here's the latest set of questions from our team--suggestions and corrections gratefully accepted:
I am having a problem with formencode:
When I feed html to into an htmlfill.FillingParser instance, this
Title: FEComponent and FormEncode problem
First, thanks for all the assistance so far with ZPTKit and FEComponent and FormEncode. We're getting close to understanding things.
A developer in our group (not a list member) asked me to post the following questions:
The FormEncode
Title: Using more than one Component at a time?
We're investigating the marriage of ZPT (via ZPTKit) and FormEncode (via a component to be written or FEComponent found on the list).
A couple initial questions:
1. Is ComponentKit designed to handle more than one component for a
samples, etc. would be gratefully
accepted.
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Ian Bicking [mailto:[EMAIL PROTECTED]
Sent: Saturday, May 21, 2005 1:52 AM
To: Hancock, David (DHANCOCK)
Cc: webware-discuss@lists.sourceforge.net
Subject: Re
: Saturday, May 21, 2005 12:58 PM
To: Hancock, David (DHANCOCK)
Cc: webware-discuss@lists.sourceforge.net
Subject: Re: [Webware-discuss] ZPTKit and standard_template question
Hancock, David (DHANCOCK) wrote:
Thanks for the ideas; I'll see if I can implement one or more (or get
somebody smarter than me
Title: ZPTKit and standard_template question
I'm working with ZPTKit so I can use Zope Page Templates with Webware, and functionally things are going well. I'm seeing a performance problem, and hope that someone can point me in the right direction from here.
My setup is Webware 0.8.1 plus
Title: FormEncode and ZPTKit
One of the things we're working on is getting our form code a bit more abstracted, and FormEncode is looking pretty good to us. Another thing we're looking at is templating with ZPT. Has anyone got a recipe for using both at once? It seems like a good use of
to the old question: Are there some semantic differences in how in-memory sessions are handled vs. file-based sessions?
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Hancock, David (DHANCOCK)
Sent: Tuesday, March 15, 2005 9:08 PM
To: webware
it's entirely possible that we've been causing them problems with
File sessions.
Again, thanks.
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Geoffrey Talvola [mailto:[EMAIL PROTECTED]
Sent: Wednesday, March 16, 2005 10:42 AM
To: Hancock, David
Title: Expiration problem after switching to Dynamic sessions
We've been seeing the occasional KeyError from the SessionStore code on our production website. We use 'File'-based sessions. We originally started using File because we kept session data on an NFS-mounted directory so we could
Title: Message
Hmm. I
sent this yesterday and received it yesterday, but got it again this morning.
Thank you, SourceForge. I'm still interested in hearing of people's
experiences with WebKit + TAL, but I won'tannoy the list with this
questionagain. Thanks.
=
We have
Title: Recipes for using TAL with WebKit
We have been using WebKit + Cheetah for some pages, and WebKit + FunFormKit for some other pages, and we're now looking for something a little more JSP-like, where we can encapsulate logic in a few custom tags and make the page templates HTML. PSP (as
Title: More information about double-retrievals with Firefox
I see in my logs that there are a lot of in-the-same-second retrievals of the same issue, and I think I saw it mentioned on the list recently. I hope a few things I just noticed helps point to a fix.
I only see the
I haven't got good answers for your other two questions, but what I use
for determining servlet execution times is the built-in logging of
Webware.
Under appworkdir/Config are the following items:
[EMAIL PROTECTED] Configs]$ grep Activity *.config
Application.config: 'LogActivity':
While it's not specifically for web applications, we've had good results
with Jim Trockij's 'mon' package. I looked at Big Brother a while ago,
but mon seemed easier to extend. It's basically a configurable scheduler
and alerter, with some monitors in the box. What we like is that we
can write a
I benchmarked File sessions vs. Dynamic and we chose File. At the time,
we were using a hardware load balancer across two webservers, so storing
the sessions on an NFS-mounted directory was a convenient
synchronization mechanism.
With other factors held constant, File vs. Dynamic resulted in a
:[EMAIL PROTECTED]]
Sent: Wed 10/20/2004 6:12 PM
To: Hancock, David (DHANCOCK)
Cc: [EMAIL PROTECTED]
Subject: Re: [Webware-discuss] session strangeness
Thanks for that.
I guess that it's not way :-)
But it does depend on what gets pickled; if it stays in memory, there's
no pickling
My results on Windows 2000:
OK for 2 threads - 20 threads running in parallel 1000
Results on Linux (Red Hat 7.3):
OK for 2 threads - 20 threads running in parallel 1000
Interestingly, on Windows (in a DOS window) if the window has focus, it
runs very slowly. Bring another window
For what it's worth, here's the output from our server just now--with the
exception of one thread (maybe it's the session-sweeper or TaskKit or some
such), we're balancing nicely. Thanks (Tim Roberts) for ps axf -- I
didn't know about that either.
22147 ?SN 0:00 /bin/sh
My understanding of the CacheContent directive in the config is that it lets
WebKit cache the content and serve it from the cache rather than from the
disk. There's some performance increase there, but not from the client side.
From the perspective of the browser it's still dynamic content. (That
We're running the latest released Webware, Linux (RH 7.3), Python 2.3.2,
Apache 1.3.x, mod_webkit adapter, and we're using the File session store.
We're seeing an increased number of KeyError tracebacks, where the key is
the session key. Interestingly, we just got two tracebacks within a couple
I'd like to run a script from the command line to look at sessions (we use
File as the storage for sessions), but I'm a tiny bit stuck. I know how to
do it from a servlet, because the servlet already has enough context to do
something like:
self.application().sessions().keys()
But I am
can use
there.
Thanks, and
Cheers!
-Original Message-
From: Scott Russell
To: Hancock, David (DHANCOCK); '[EMAIL PROTECTED]'
Sent: 4/26/04 5:41 PM
Subject: Re: [Webware-discuss] Standalone scripting
I've never done this, but I suspect you want to dig around in
/WebKit/SessionStore.py. I
Thanks, Ian. This entire activity (getting domain, getting hosting,
migrating content, building a Wiki) has been amazingly fast.
One thing I just noticed: Clicking the Wiki link on webwareforpython.org
still takes you to the old Wiki (at colorstudy.net). Is this on purpose?
Seems like a Bad Idea
I'm not running Red Hat 9.0, but it's modern enough that maybe it runs
Apache 2.0 by default. If so, you probably want to compile mod_webkit2
instead of mod_webkit.
Caveat: This advice could also be completely wrong--I use Red Hat 8.x and
Apache 1.3.x.
Cheers!
--
David Hancock | [EMAIL
Kurze Fragen...quick questions...
1. Does anyone have an Oracle connector for SQLObject already built?
2. How do people (in general) and how do object-relational mappers (in
particular) deal with changing data models?
Sorry #1: Both of these are NOT Webware-specific.
Sorry #2: The second
I hope the following is helpful. It's a function daemon that exists on
Linux (Redhat, at least, maybe other SysV init systems, too). It's used to
make a backgrounded daemon out of a foreground process, and it looks like it
deals with getting the *right* PID. Another possibility is to switch to
What database are you using? If it's one that supports auto columns or
sequences (automatically incremented counters, specifically for primary
keys), consider using such a function when you insert. If your database
doesn't support that, then I'm not sure what you can do.
What I just described is
I think that Apache does NOT pass certain environment variables through to
CGI programs (and by extension, to application servers such as Webware. You
can override this behavior by including lines like the following in
httpd.conf:
PassEnv HTTP_HOST REMOTE_ADDR HTTP_REFERER
This should pass the
I have heard that RH 9.0 has a new improved threading model, so it's likely
that you've really got the right number of threads running, but they're
running in a single process (as they should). Earlier OS releases use
lightweight processes for the threads. Those processes are sharing
resources
I won't profess to understanding exactly how this works, but one of our
developers did exactly what you're trying to do. We wanted to save and email
the full HTML red-bar traceback for oncall engineers, but we only wanted
users to see a short message with an identifying number so we could
Title: Webware/WebKit and load-balancing
Has anyone in Webware-land been successful implementing a load-balancer between Apache and one or more WebKit instances? I've been trying to do this for many weeks without success. I wrote about my problems a while ago, but I still haven't had any
FWIW, we've got an application with about 60 servlets, and we start 30
threads, and we pool Oracle connections. Right now, after a week, we're at
190MB. It doesn't seem to be leaking, though; it grows to about 200MB very
quickly and stays there. We've got lots of RAM and this isn't a problem for
: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Sent: Friday, August 29, 2003 8:56 PM
To: Hancock, David (DHANCOCK); [EMAIL PROTECTED]
Subject: Re: [Webware-discuss] Problem using WebKit through a load-balancer
-- Friday, August 29, 2003, 7:35:39 PM, [EMAIL PROTECTED] wrote:
I'm experimenting
Title: Problem using WebKit through a load-balancer
I'm experimenting with using pythondirector for load-balancing several WebKit servers, but I get the following error messages in the Apache log:
[Fri Aug 29 23:59:34 2003] [error] [client 144.243.111.8] Premature end of script headers:
I'm a big fan of Webware--we use it where I work.
But for what you're trying to do, you should consider simply running a
recent build of Mozilla. I had been using an HTTP proxy called Proxomitron
for sniffing the HTTP interactions, but Mozilla has it built in as an
option: Tools - Web
Tomi: I wonder if this is related to my experience sending email to you when
I copy the list. Every such message (three so far) has been permanently
bounced by comcast.net because it says your mailbox is over its size limit.
I'll bet that the sourceforge lists are also getting occasional
I'm not sure what you mean by slow. Do you mean less traffic or do you
mean the speed with which a message to the list gets rebroadcast to the
subscribers?
Neither of the interpretations of speed seem like problems on
webware-discuss to me, though.
Cheers!
--
David Hancock | [EMAIL PROTECTED] |
t sure about that. When we see
the processes looking like they do below, they really ARE wedged, though,
and require manual termination.
Cheers! -- David Hancock |
[EMAIL PROTECTED] | 410-266-4384
-Original
Message- From:
Hancock, David (DHANCOCK)
-Original Message-
From: Hancock, David (DHANCOCK)
Sent: Friday, August 01, 2003 4:57 PM
To: [EMAIL PROTECTED]
Subject: Anyone seen WebKit processes going into a weird state?
Several times a week on our production systems, we're seeing our WebKit processes (normally entitled python Launch.py
Title: Anyone seen WebKit processes going into a weird state?
Several times a week on our production systems, we're seeing our WebKit processes (normally entitled python Launch.py ThreadedAppServer) lose their command lines in the output from ps. They're also well wedged, and the processes
I think that the offending line is the SetHandler--it tells Apache to send
everything to WebKit. Try commenting it out, restarting Apache, and see if
that fixes it for you.
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Ian [mailto:[EMAIL
Title: Signals and Webware/WebKit
We occasionally spawn long-running processes from WebKit servlets, and sometimes they hang up. We'd like a way to detect this and abort cleanly. In process-land, we would have done this with os.signal, SIGALRM, and a signal handler. This isn't working out for
The output from running the AppServer goes to standard out, which is why
you see it on the screen until you log out.
We start the server in the following way:
nohup ./AppServer webkit.log
The nohup command ensures that the process isn't killed when you log out,
and the webkit.log
-thread issue.
Cheers!
-- David Hancock | [EMAIL PROTECTED] |
410-266-4384
-Original Message-From: Hancock, David
(DHANCOCK) [mailto:[EMAIL PROTECTED] Sent: Tuesday, March 11,
2003 6:57 PMTo:
[EMAIL PROTECTED]Subject: [Webware-discuss]
Oracle/DCOracle/Python problem while
[mailto:[EMAIL PROTECTED]
Sent: Monday, March 10, 2003 3:07 PM
To: 'Hancock, David (DHANCOCK)'; [EMAIL PROTECTED]
Subject: RE: Wedged threads (was RE: [Webware-discuss] help evaluating We
bWar e)
Which extension modules are you using (for database access or other
purposes)? Being pure Python
Title: Oracle/DCOracle/Python problem while using Webware
First, sorry for the off-topic posting, but I'm trying several venues with this question because it's becoming a larger problem. This is not a Webware problem, nor probably a Python or even DCOracle2 problem, but it's biting us in the
-Original Message-
From: Geoffrey Talvola [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 06, 2003 9:41 AM
To: 'Hancock, David (DHANCOCK)'; [EMAIL PROTECTED]
Subject: RE: [Webware-discuss] help evaluating WebWare
Hancock, David (DHANCOCK) [mailto:[EMAIL PROTECTED] wrote:
We have had some
There is an aspect of Zope that drove our team into using Webware--if there
were two simultaneous long-running requests, Zope seemed to block or queue
additional web requests until one of the two completed. Our application has
a need to shell out to processes that may take 30-45 seconds to
I hope someone chimes in with a better idea, because my ideas involve some
work...
If you have sufficient rights to have cron jobs, consider writing a little
monitor that runs via cron. It just needs to look for the appserver
process, or retrieve a page using wget or lynx -dump, and if it can't
8192KB of stack is a lot, and that seems to be the default on Linux. We
haven't had any stack-limit problems running very large apps in Webware on
Linux.
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Ian Bicking [mailto:[EMAIL PROTECTED]]
Sent:
Matt Feifarek notes below that Dynamic sessions haven't been working for a
while. So far as I can tell in 0.7, they seem to work OK. Am I missing
something?
I've been thinking about getting a little more current with CVS, but I need
the Dynamic sessions. Is there anything else that worked in
We've got up to 12 people working on local copies of the 'make app workdir'
application directories, with the master copy in CVS (exactly as suggested
below). We're also using mod_webkit with 12 different port numbers, and it
handles them nicely. We just call the locations /WK1, /WK2, etc. We
Nick: I have heard of something called the global interpreter lock in
Python that sounds like it means only one process can get hold of some
critical part of the interpreter at one time. Maybe that's related, maybe
I'm uninformed... Probably the latter.
As nice as httplib is for benchmarking,
I've never done this in Webware, but you should be able to do what you want
with some combination of:
- something in wmail.sendMessage to write to the browser
- something after each call to wmail.sendMessage
- flushing stdout (or whatever the response.write method is writing to)
I doubt
Please forgive a naïve question, but how do class docs differ from the
pydoc-generated HTML files in Webware (or what you can see by running pydoc
-p 8080 and then browsing on port 8080)?
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: F. GEIGER
There is also Comanche (www.comanche.org), which is a Tcl/Tk-based GUI for
Apache configuration.
I think, though, that the question was asking about programmatic
configuration (as is, building or modifying httpd.conf on the fly) rather
than graphical front-ends for editing httpd.conf.
Comanche
I can't speak for the MySQL module, but when using the DCOracle2 module,
database connections are automatically and reliably closed when the app
server is stopped or restarted.
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Costas Malamas
PROTECTED] | 410-266-4384
-Original Message-
From: Geoffrey Talvola [mailto:[EMAIL PROTECTED]]
Sent: Monday, May 06, 2002 10:02 AM
To: 'Hancock, David (DHANCOCK)'; '[EMAIL PROTECTED]'
Subject: RE: [Webware-discuss] WebKit Error traceback pages don't close th
eir connection?
Hancock, David wrote
tool, and if we could get all pages (even ones with mistakes) to
load programmatically (via urllib), we'd have a leg up on automated testing.
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Hancock, David (DHANCOCK) [mailto:[EMAIL PROTECTED]]
Sent
a MiddleKit setting) with MySQL
at 20 connections.
-Chuck
On Thursday 02 May 2002 03:11 pm, Hancock, David \(DHANCOCK\) wrote:
I only recollect that it was reported that there was no speed boost
using MySQL, and nobody else had reported any results.
My testing has shown about 10:1 performance
I only recollect that it was reported that there was no speed boost using
MySQL, and nobody else had reported any results.
My testing has shown about 10:1 performance boost using DBPool.py to pool
connections to Oracle. After looking at the database machine I'm connecting
to, I can tell you for
Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Geoffrey Talvola [mailto:[EMAIL PROTECTED]]
Sent: Sunday, April 28, 2002 9:34 PM
To: Hancock, David (DHANCOCK); 'Geoffrey Talvola';
'[EMAIL PROTECTED]'
Subject: Re: [Webware-discuss] WebKit Error traceback pages don't close
127.0.0.1:9034
FIN_WAIT2
tcp0 0 127.0.0.1:9034 0.0.0.0:* LISTEN
[dhancock@dhancock dhancock]$
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Hancock, David (DHANCOCK) [mailto:[EMAIL PROTECTED]]
Sent
box). If there's more information being collected, perhaps that's where
the hang is.
Cheers!
-- David Hancock | [EMAIL PROTECTED] |
410-266-4384
-Original Message-From: Geoffrey Talvola
[mailto:[EMAIL PROTECTED]] Sent: Friday, April 26, 2002
1:02 PMTo: 'Hancock, David (DHANC
-4384
-Original Message-From: Hancock, David
(DHANCOCK) [mailto:[EMAIL PROTECTED]] Sent: Friday, April 26, 2002
2:23 PMTo: 'Geoffrey Talvola';
'[EMAIL PROTECTED]'Subject: [Webware-discuss]
(no subject)
I
first noticed this problem using IE-on-Windows. I also see
-From: Geoffrey Talvola
[mailto:[EMAIL PROTECTED]] Sent: Friday, April 26, 2002
2:42 PMTo: 'Hancock, David (DHANCOCK)';
'[EMAIL PROTECTED]'Subject: RE:
The Ids section is
the last part of the exception report unless you have IncludeFancyTracebacks
set to true, in which case
Title: WebKit Error traceback pages don't close their connection?
When I get an error traceback page (it happens a lot, I make a lot of mistakes), it seems like the HTTP connection doesn't get closed. The symptoms I see are that the traceback page loads, down to the end (I think), where the
Title: Prior incarnation of WebKit?
One of the people in my group found the following link with a 1996-vintage product called WebKit.
Has anybody heard of this? (It purports to be part of WebWorx, which also sounds dead.) Is anyone worried about underflap.com getting after Webware to change
On zope.org I recall reading about such a thing. I think it had a piece that
ran like a proxy, collecting actions into a script, and then could run the
script. That's all I recall; I hope this helps.
Cheers!
--
David Hancock | [EMAIL PROTECTED]
-Original Message-
From: Ian Bicking
To:
I've got two books, both of which I can recommend. The first is from my
pre-Python days: _Network Programming with Perl_, but Lincoln Stein,
Addison-Wesley.
But because I'm using Python now, I can also recommend _Web Programming with
Python_ by Steve Holden, New Riders. It isn't specifically
and reopening the
connection in the try/finally block as you suggested.
I appreciate the help.
Cheers!
--
David Hancock | [EMAIL PROTECTED] | 410-266-4384
-Original Message-
From: Tripp Lilley [mailto:[EMAIL PROTECTED]]
Sent: Saturday, March 02, 2002 3:15 PM
To: Hancock, David (DHANCOCK
Title: WebKit seg-faulting when connecting to Oracle
I'm using a recent (within the past two weeks) download of Webware 0.6. I've gotten WebKit working via CGI, via mod_python, and now via mod_webkit. But now I'm trying to get sophisticated and connect to Oracle to see what the performance
75 matches
Mail list logo
