Howdy folks,
Seems like we're having another flare-up of frustration on the list...
I left AOL about a year ago and haven't had much time to contribute
since. I probably wrote (re-wrote and wrote again) 90% of the code
and had several teams hacking Tcl code for dozens of AOL web sites
ove
Hi,
After reading through all the responses to my "aolserver focus" post,
it seems to me Thorpe's comments below are the most realistic and
actionable, i.e., it's the documentation / getting started stuff
that's insufficient.
Otherwise, technically there are a few things that could be fi
This reminds me of the Smarty for PHP. Has anyone looked at porting
Smarty to ADP? It's pretty PHP-specific but the syntax, like Tom's
stuff below, is pretty convenient.
http://smarty.php.net/
-Jim
On Sep 25, 2007, at 10:50 AM, Tom Jackson wrote:
Jeff,
I developed a templating syst
Howdy,
I wrote the core SOB code years ago -- at the time I think our goals
included:
-- Security: Server connects out to known clients.
-- Performance: State-full connections, minimal protocol overhead,
lots of caching, no encoding overhead
-- Simple: String keys, string data, integrati
Agreed :) Most of AOLserver core is ok being loaded into core Tcl
although the server config and module/Tcl init framework
incompatible. It can be made to work but old-style AOLserver modules
would need to be updated.
-Jim
On Feb 28, 2008, at 6:53 PM, Jeff Hobbs wrote:
Matthew M. Bur
There's an "ns_internalredirect" Tcl command which calls the
Tcl_ConnRedirect C function -- maybe that does the right thing. I
suppose you could also use "ns_register_adp" to map the URL to an ADP
file instead of using "ns_register_proc" -- the ADP could then do the
login and include g
Hi,
I don't think you lose anything -- Tcl_Finalize carefully tears
everything down which is good when the code is used in an embedded
system but in AOLserver as a normal process on Unix or Windows the
resulting _exit doesn't care about all the work Tcl_Finalize may have
done.
-Jim
Howdy,
I'm the one that made most of "assumed good" changes to AOLserver 4.5
-- my fault. Seemed like a good idea at the time. Is there a
complete list of changes to right that wrong? I may be able to hack
in the fixes quickly -- I'm just not sure the scope because I'm way
behind on th
ns_share, static vars (config
structure)
atomic variables: nsv arrays
synchronizers: ns_mutex, ns_cond, etc.
Incidentally, those aren't so much language features, they're more
features of the AOLserver/Tcl API and environment. "Tcl" didn't have
nsv_* or ANY of the cool st
been to use the Tcl event loop in a 2nd dedicated
thread: Requests would be read with the current AOLserver I/O event
loop and then passed to a Tcl event loop if needed before being queued
for execution by a dedicated thread. Hmm
-Jim
On Apr 14, 2008, at 8:15 PM, Tom Jackson wrote:
S
Hi,
Brent Welch wrote some ns_share code which worked with Tcl variable
traces to emulate the original code. It works except for a few edge
cases but is generally considered less efficient and flexible than the
new nsv_* commands. So, when I say we gave up on ns_share I mean we
stopped
Hi folks,
I wrote the code. The explanation below is correct -- I chose inode/
dev combination to cache the same file even with multiple names which
was the case at AOL -- hundreds of symlinks and hard links to the same
file. The same strategy is used for ADP templates. I think the code
Hi,
I haven't looked at a "directory change notification" type scheme in a
long time but that could be very clever. Aside from addressing issues
discussed here, the key benefit would be to avoid the repeated "stat"
syscalls. Those stat calls always bothered me conceptually but the
perfo
Your right, the code snippet below could trip over a race condition as
you've described. But, that's not reason enough to change the
fastpath, it's reason to better document the behavior so folks don't
write code that uses ns_returnfile for temporary, dynamic content.
Although fastpath t
though I still think a dynamic app using
files should use open/cached fd's).
-Jim
On Aug 19, 2008, at 9:25 PM, John Caruso wrote:
On Tuesday 05:39 PM 8/19/2008, Jim Davidson wrote:
Your right, the code snippet below could trip over a race condition
as you've described.
It&#x
rnfp, but
I'm thinking that may break more than it fixes if folks are assuming
the behavior it has now (e.g, an ns_returnfile of an image where they
expect the not-modified responses).
-Jim
On Aug 20, 2008, at 9:23 AM, Titi Alailima wrote:
-Original Message-
From: AO
ng was
plugged years ago, e.g., when fsirand was introduced for scrambling
NFS vnodes.
-Jim
On Aug 20, 2008, at 12:54 PM, John Caruso wrote:
On Wednesday 08:45 AM 8/20/2008, Jim Davidson wrote:
Overall, it seems one thing to do would be to switch to filename-
based
cache keys by default,
Hi Folks,
I agree with Eric, even though I wrote the original code and was one
of the first to suggest is wasn't a bug. This thread has surprised me
in a few ways:
-- The bug was indeed subtle and curious
-- The debate on dynamic vs. static and underlying assumptions and
performance was
cialized "artblaster" (architecturally an AOLserver opposite) and/
or some commercial, geographically distributed, and generally smart
CDN -- leaving AOLserver to mostly ADP tasks.
-Jim
On Aug 20, 2008, at 7:59 PM, russell muetzelfeldt wrote:
On 21/08/2008, at 1:45 AM, Jim Davidson
Hi,
Yes -- the original reason for dev/inode on Unix instead of filename
was to reduce memory consumed in the case of a large # of symlinks or
hardlinks to the same file. This was the case for AOL's
digitalcity.com back in 1999. For better or worse, the "AOL" in
AOLserver means AOL was
8, at 11:27 PM, John Caruso wrote:
On Thursday 02:34 PM 8/21/2008, Jim Davidson wrote:
To clarify one point: There is no technical solution to creating
temp
files with the same name and avoiding the race condition without
additional synchronization.
To clarify as well: the original code d
HI,
Sorry -- I wrote the original and goofy code leading to this
confusion...
Based on this discussion I think it's NOT working as intended. If
there was a time I thought ns_adp_abort should mean "return NS_ERROR
for connection status and thus no-traces..." I've long since
forgotten t
Hi,
A bit old but let me try to be helpful here...
On Apr 3, 2009, at 11:45 PM, Tom Jackson wrote:
On Fri, 2009-04-03 at 16:20 -0600, Jim Davidson wrote:
HI,
Sorry -- I wrote the original and goofy code leading to this
confusion...
Thanks for monitoring our head scratching!
I have a
resource), it would make sense but for an old-
school common log format it may be odd (or at least suspect, but
perhaps no less suspect than what's going on now).
-Jim
On Apr 13, 2009, at 2:48 PM, Tom Jackson wrote:
On Mon, 2009-04-13 at 13:49 -0400, Jim Davidson wrote:
Hi,
A bit old
Hi,
I haven't been able to test myself but the changes make sense to me
too. Thanks Tom!
-Jim
On Apr 15, 2009, at 6:52 AM, Gustaf Neumann wrote:
Dear Tom,
your rewrite of the two functions below look fine to me. The structure
is much clearer now, results of the authorization handling are
I'm wondering if there needs to be some more specific exception flags
stored in the connection structure to handle all these cases and some
C and Tcl api's to access/modify the same. It seems you're doing the
good work of rationalizing all the error conditions that had been
pretty confused
Hi,
Do you have some sort of background job that calls "ns_server
active" (or similar) regularly? That could lead to random crashes.
The description in http://dev.aolserver.com/trac/ticket/152 is
accurate: The code, by design, is not strictly safe as it's assumed
to only be used intera
t, is prohibited.
On Thu, May 14, 2009 at 10:19 AM, Jim Davidson
wrote:
Hi,
Do you have some sort of background job that calls "ns_server
active" (or similar) regularly? That could lead to random crashes.
The description in http://dev.aolserver.com/trac/ticket/152 is
accurate
t 12:33 PM, Jim Davidson wrote:
Yup -- should really have been documented better -- sorry about
that.
Anyway, what is the monitoring attempting to dig up? There may some
other safe ways to get the same.
-Jim
On May 14, 2009, at 2:04 PM, Jade Rubick wrote:
Ironically, we have some monit
Hi,
I'm looking at the head code and it appears it's now safe -- the
connection list is walked with a pool lock held and the request data
(method, url) seem to be copied with a global "reqlock" mutex held.
Jade: What version of AOLserver are you using?
-Jim
On May 15, 2009, at 10:06 A
:
I guess if it's thread safe now, the ticket should be closed with a
reference to aolserver 4.5.
2009/5/16 Jim Davidson
Hi,
I'm looking at the head code and it appears it's now safe -- the
connection list is walked with a pool lock held and the request data
(method, url)
Hi,
The code for "ns_info threads" looks about the same in 4.0 and 4.5 and
it doesn't look super safe. It does have a lock around walking the
list of threads but there are place where it copies data from strings
which may be changing. And, like "ns_server active", it's really
designed
The code should work ok. I remember it was a bit messy to map these
Aolserver tcl commands which pre-dated the Tcl channel stuff so it was
compatible years ago and there were examples of memory leaks from long
running detached tcl threads but with care it should be ok
Jim
Sent from my iPh
With fancy switches and/or proxies like varnish can you effectively
blend aolserver with other app servers (lamp, ruby etc) now without
actual code changes ? I'm wondering if folks have done that
successfully
Jim
Sent from my iPhone
On Oct 27, 2009, at 6:55 PM, Dossy Shiobara wrote:
needs to run
in a forking environment so it can play nice with PHP)
No, this won't be as great and efficient as AOLserver is, but it
will be a whole lot easier for people to use and sell to fellow
staff, management, clients, etc.
Just my thoughts on the subject! (And I do realise this i
Howdy,
Looking back at the code and trying to remember what I was thinking at the
time, I ran across the header comment to "NsConnContent" which mentions the
possibility of a mapping failure (see below). This reminded me of what was
going on...
Originally there was no "spool to file" option
Right -- the pre-queue thing operates within the driver thread only, after all
content is read, before it's dispatched to a connection. The idea is that you
may want to use the API to fetch using event-style network I/O some other bit
of context to attach to the connection using the "connection
On Dec 1, 2009, at 4:45 PM, Jeff Rogers wrote:
> Jim Davidson wrote:
>> Right -- the pre-queue thing operates within the driver thread only,
>> after all content is read, before it's dispatched to a connection.
>> The idea is that you may want to use the API to fetch us
your connection gets running.
-Jim
On Dec 1, 2009, at 4:45 PM, Jeff Rogers wrote:
> Jim Davidson wrote:
>> Right -- the pre-queue thing operates within the driver thread only,
>> after all content is read, before it's dispatched to a connection.
>> The idea is that you m
Hi,
I think we were talking about this about a month ago. I updated the source to
enable upload-progress checking with a combination of ns_register_filter and
nsv -- there's an example at the latest ns_register_filter man page (pasted
below). This may work for you although it would require co
etc. While all that stuff was a bit messy and confusing on
AOLserver, it worked -- performance of interesting LAMP apps like Drupal seem
to suffer for lack of such lower level design principles we had in AOLserver.
Interesting to see how this has evolved over 15 years (some of the first code
for multith
The method of checking progress on a separate URL similar to the
example I sent does result in repeated requests during upload. But,
they're trivial by comparison - easily in the 100's of req/sec range
of response time and throughput. A bit goofy, but over a single keep-
alive socket for
Howdy,
I still watch the list but haven't had time to dig into the code in a long time
-- maybe later this summer :)
Anyway, these connection filters are a bit confusing and not well documented --
my fault. Here I think it's ok to not call the filters again but it could go
either way. If yo
You can use normal Tcl or aolserver socket functions but the core server
listening and http processing is too low level to expose via Tcl. Was always a
good idea, eg a Tcl command that creates a server instance, but would require a
lot of refactoring the core.
Jim
Sent from a phone
On Jun
Hi,
The short answer is no, there's no access log entry although there may be a
server log message buried in the chatter.
The reason is the access log is a "trace" that fires at the end of an HTTP
connection and the request isn't a connection until all the content has been
read and the data
>
> From: AOLserver Discussion [AOLSERVER@LISTSERV.AOL.COM] On Behalf Of Jim
> Davidson [jgdavid...@mac.com]
> Sent: 23 June 2011 13:47
> To: AOLSERVER@LISTSERV.AOL.COM
> Subject: Re: [AOLSERVER] Problem with file uploads larger than maxinput
>
ad. Then, the
> connection thread can decide how to handle the request -- but, the driver
> thread still makes sure that the hard limits are enforced on the request that
> will be passed on to the conn thread ... ?
>
>
> On 6/23/11 8:47 AM, Jim Davidson wrote:
>> The reason
rror: conn[44]: max
> content exceeded
>
> Brian
>
>
> From: AOLserver Discussion [AOLSERVER@LISTSERV.AOL.COM] On Behalf Of Jim
> Davidson [jgdavid...@mac.com]
> Sent: 23 June 2011 14:44
> To: AOLSERVER@LISTSERV.AOL.COM
> Subject: Re
Agreed. Should go on the bug/feature list. Is there a list ? :)
Sent from a phone
On Jun 23, 2011, at 9:58 AM, Dossy Shiobara wrote:
> Yes, on E_CRANGE, the driver thread should return HTTP 413 Request too large,
> IMHO. And, we should be able to configure a custom response for status code
I think this one case can be handled this way because you could truncate the
request (some content may have been read ahead) and then flag it for special
handling in the normal connection processing code/threads. The reason is the
server has a full request, you're just choosing to not read all
Cool. The driver thread is technically a single threaded thing which shouldn't
block and the connection return functions are blocking but as the response is
small, it probably just works. Also, the connection code hasn't been verified
to run in the driver (filters, traces, etc.) but again it l
ully cared for the Windows support in AOLserver --
> once upon a time, I had built the Windows binaries that some folks were
> using. Through discussions I had with Jim Davidson, the new build mechanism
> for AOLserver 4.5.x was meant to make building AOLserver on Windows easier
> th
Hi
It's a socket so it can be monitored by select and poll. It should be SOCKET, I
think it was in the past.
On windows "lib-c" file handles returned by _open aren't the same as sockets.
You can see this in the libc source Microsoft provides. They can't be
monitored with select. The ns_so
Hi,
I'm looking at the code now -- definitely needs to be SOCKET in nsd.h. The
reason can be seen in ns_sockpair in fd/sock.c where the code for a socket pair
is done. It's just a wrapper around Unix socketpair() but has a bunch of extra
code to do the loopback-connect thing on Windows. The
Sounds good to me.
-Jim
On Aug 4, 2011, at 4:21 PM, Jeff Rogers wrote:
> Hi all,
>
> I'm implementing a minor enhancement to the adp parser to make null end tags
> (aka empty elements or minimized tags) work. Or more simply, if you have a
> registered adp tag where the end tag matches the o
----Original Message-
> From: AOLserver Discussion [mailto:AOLSERVER@LISTSERV.AOL.COM] On Behalf Of
> Jim Davidson
> Sent: 03 August 2011 21:21
> To: AOLSERVER@LISTSERV.AOL.COM
> Subject: Re: [AOLSERVER] Progress in Aolserver
>
> Hi,
>
> As Dossy mentioned, we spent so
I vaguely remember never figuring this out either and deciding to ifdef it out.
In practice it doesn't do much -- I've never come across a on-exit handler
that really needed fire. Curious if anyone has.
Jim
Sent from a phone
On Aug 6, 2011, at 3:29 PM, Maurizio Martignano
wrote:
> Hello
Howdy,
Looking at the code, ns_returnfile passes the filename through to the core
Ns_ConnReturnFile without any of the care that core Tcl does handling
filenames. You may be able to replace ns_returnfile with ns_returnfp, passing
a file handle returned from Tcl's open command which should be
Agreed -- this has always been one of the must unique and/or goofy aspects of
AOLserver and Tcl. I gave the keynote at the 7th (yes, 7th!) tcl/tk conference
years ago:
http://www.aolserver.com/docs/intro/tcl2k/
You could mention how this work has continued on and off over the years.
Hi,
Cool! Nice updates :)
On the version # question a few days back, I agree this is 4.x update. For a
5.x release, in addition to what you listed below, I'd add:
-- integrate SSL support directly (comm driver, api's)
-- integrate the comm drivers
-- figure out some better build environment f
Good catch -- trying to get IP6 working makes sense as well.
-Jim
On Oct 25, 2011, at 3:54 PM, Daniël Mantione wrote:
>
>
> Op Tue, 25 Oct 2011, schreef Jim Davidson:
>
>> Hi,
>>
>> Cool! Nice updates :)
>>
>> On the version # question a few d
Dossy: Thanks for handling this!
-JIm
On Oct 26, 2011, at 9:43 AM, Dossy Shiobara wrote:
> Hi,
>
> There has been little discussion or response to this matter, and the few
> responses have all been in favor of moving the lists to SourceForge.
>
> Today is the 26th, so I'd like to do the list
In a message dated 4/3/2001 12:00:12 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
>
> Date:Mon, 2 Apr 2001 15:38:13 -0500
> From:Rob Mayoff <[EMAIL PROTECTED]>
> Subject: test case for 3.3.1 crash?
>
> Recently there was discussion of 3.3.1 crashing due to improper adoption
> of Z
In a message dated 4/6/2001 12:04:09 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> > I believe, on my 4th attempt, I may have correctly implemented Zoran's
> > patches to the 3.3.1 code to avoid the crash.
> >
> > Hopefully all is well now. This stuff has to work now - Tcl 7.6 has been
In a message dated 3/31/2001 1:03:13 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> I've been trying to tune AOLServer for performance with static pages.
> The things I've done is to turn of logging,set threadmax to 20 and 1
> connection per thread
> Is it possible to get AOLServer to out
In a message dated 4/7/2001 12:04:07 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> Check http://www.acme.com/software/thttpd/benchmarks.html and look
> at the graph at the bottom. There is a discussion of sending http
> headers out separate from the data, which is (I believe) what AOLSe
In a message dated 4/14/2001 12:04:43 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> What do people think? I'd be willing to take the diff and apply
> it to 4.0 if the community thinks its worthwhile/safe.
>
> - Dossy
>
>
> On 2001.04.13, Mr. Source Forge <[EMAIL PROTECTED]> wrote:
> > S
In a message dated 4/19/2001 12:03:25 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> From:Peter Harper <[EMAIL PROTECTED]>
> Subject: ns_cond, threads and file descriptors.
>
> I've been trying to find a solution to the following problem...
>
> I have two message queues in shared memo
In a message dated 4/21/2001 12:02:34 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
>
> gmake[3]: Entering directory
> `/usr/local/aolserver3_3_1/tcl8.3.2/unix'
> gmake[3]: *** No rule to make target `libtcl83g.a'.
> Stop.
> gmake[3]: Leaving directory
> `/usr/local/aolserver3_3_1/tcl8.3.2
In a message dated 4/28/2001 6:12:27 PM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> On 2001.04.28, Rob Mayoff <[EMAIL PROTECTED]> wrote:
> > > I'd like to conduct a quick informal poll. What are the
> > > top 10 most commonly used ns_* API calls that you use?
> >
> > Do you mean at autho
there are no plans to either discontinue support for Tcl nor engage in
major architectural changes to support these technologies directly within
AOLserver. However, others have extended AOLserver in such ways (e.g.,
support for Python and in-process Java VM's) using the existing C API.
-Jim
Ji
In a message dated 4/30/2001 8:58:36 PM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> On 2001.04.30, Jim Davidson <[EMAIL PROTECTED]> wrote:
> [ ... ns_stats, command traces, nsd/tclstats.c, etc. ... ]
> >
> > I removed this from 4.x because it was being
In a message dated 6/26/2001 12:03:21 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> Date:Mon, 25 Jun 2001 19:02:54 -0500
> From:"Michael C. Urban" <[EMAIL PROTECTED]>
> Subject: Re: Segfaults when running on port 80
>
> On Mon, Jun 25, 2001 at 05:38:57PM -0500, Rob Mayoff wrote:
In a message dated 12/8/2001 1:17:17 AM Eastern Standard Time,
[EMAIL PROTECTED] writes:
> When calling Ns_TclEnterSet, I'm confused about how to choose between
> NS_TCL_SET_TEMPORARY (0) and NS_TCL_SET_DYNAMIC (1).
>
> When I look at NsTclDbCmd (in dbtcl.c, AOLserver 3.3+ad13), I see that
> it
In a message dated 12/8/2001 1:17:17 AM Eastern Standard Time,
[EMAIL PROTECTED] writes:
> In nsd/return.c we have modified the function
> Ns_ConnReturnRedirect. Please note the additional block of
> code marked with a "ZZZ BNA" comment. What this does is
> return the user specified host name
In a message dated 12/28/01 12:02:35 AM, [EMAIL PROTECTED] writes:
>For C modules, simply create the C functions you want exposed from your
>module as externs and call them Ns_SomeName. Just make sure you don't
>conflict with a C function name that already exists in the core or in other
>modules.
In a message dated 1/12/2002 12:02:57 AM Eastern Standard Time,
[EMAIL PROTECTED] writes:
> Date:Fri, 11 Jan 2002 18:49:55 -0800
> From:Jim Wilcoxson <[EMAIL PROTECTED]>
> Subject: Access log format bug w/400 (Bad Request) errors
>
> Saw this today in my access log:
>
> XX.XX.XX.XX - - [
In a message dated 2/18/2002 12:01:25 AM Eastern Standard Time,
[EMAIL PROTECTED] writes:
> I'm writing some C code to handle HTTP header parsing inside of AOLserver
> (don't ask -- it's for CGI emulation inside of parsed Python scripts) and
> I'm running across a problem where a redirect is cau
In a message dated 2/20/2002 12:00:16 AM Eastern Standard Time,
[EMAIL PROTECTED] writes:
> > Before you do and spend a lot of time coding form.tcl in C though,
> > I would recommend seeing what the AOLServer head is doing. I didn't
> > examine it closely, but it's only about 20 lines of Tcl no
In a message dated 3/7/02 12:01:29 AM, [EMAIL PROTECTED] writes:
>> That explains a few things.&-96; If the allocator modification is made to
>> the Tcl core, will the AOLServer distribution be changed to link to the
>> system Tcl lib's in a conventional manner?&-96; (eg, instead of including
>>
In a message dated 8/29/02 11:48:19 PM, [EMAIL PROTECTED] writes:
> Date: Thu, 29 Aug 2002 15:46:57 +0100
> From: Harry Moreau <[EMAIL PROTECTED]>
> Subject: How to crash AOLserver 3.4 in one line of tcl...
>
> The line is
>
> ns_rwlock destroy rid0x83daf50
>
> providing that rid0x83daf
In a message dated 8/28/02 11:48:47 PM, [EMAIL PROTECTED] writes:
> Subject: ns_param SystemScope obsolete?
>
> Is the ns_param SystemScope obsolete?
>
> Grepping through both the 3.3+ad13 and 4.x (cvs head) sources, I don't
> see it anywhere. It looks to me as if the thread scope is determin
In a message dated 9/23/2002 11:42:19 PM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> What I would really like to see is a better wakeup mechanism so that
> all of the threads don't wake up at once when a new request comes in.
> I think that's how it works now in many cases, though don't kn
In a message dated 9/29/2002 11:39:47 PM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> We now have three modules (that I know of) which use OpenSSL. I think
> it's time to break OpenSSL into its own openssl.so module, and have it
> loaded prior to loading nsencrypt.so, nsopenssl.so and nsima
In a message dated 10/30/02 9:03:22 AM, [EMAIL PROTECTED] writes:
The article mentions that they stayed away from Java because of the thread
implementation on FreeBSD (presumably 4.x). Given that AOLserver uses
threads heavily, does anyone have experience running it under FreeBSD? Is
it OK? O
101 - 185 of 185 matches
Mail list logo
