Re: [AOLSERVER] is it necessary

[Bernd Eidenschink]

> This why all of the Apache supported languages either require the pre-fork
> model where they keep one interpreter per process (like PHP) or party like
> it's 1993 in a FastCGI environment like Ruby (on Rails) does. Yuk.

Don't name the FastCGI-party-people, they are mean!

Albeit there applications have momentum, FAQ's, HowTos, packages in nearly 
every linux distribution, books to buy, thousands of developers; albeit they 
power all the world-famous well-known apps everyone uses, they use this lame 
technology. Even worse, if there apps stall, they place nginx or lighttpd with 
memcached in front of it and survice slashdot, just mad!

Hm.

Even using and relying on several AOL- and Naviservers for years now I cannot 
get used to the technical superiority discussions...

"So what? TCL is thread safe (tm)!"

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to 
 with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] nsopenssl dropping connections

[Bernd Eidenschink]

Hi Scott,

that's really very interesting!

> Be aware that using keepalive holds a connection open for longer
> periods of time; I haven't looked at the code to determine whether
> those connections take up a slot in the 'maxconnections' bucket, but
> I suspect that they do as AOLserver needs to maintain the connection
> information.

I assume default settings of 
listenbacklog = 32
maxconnections = 100
maxthreads = 10

Would it be correct to say:

(A)
If keepalive is turned off and it is known or very likely that your threads 
are long running and computing expensive tasks, "listenbacklog" may become a 
limiting factor when "maxthreads" (default 10) are reached and waste away in 
parallel. The 23rd+ visitor/request would be dropped (listenbacklog minus 
maxthreads).
"maxconnections" is not really relevant here.

(B)
If keepalive is turned on at least "maxconnections" people are served for a 
keepalive period of time. Not every request is served immediately of course 
that depends on "maxthreads".
Let's assume 10 threads serve 10 connections each ("maxconnections" reached) 
then there is a "buffer" of "listenbacklog" for new requests.
The 133rd request is dropped.

Scenario (A) and (B) always assume "maxthreads" are reached.
In (A) it would make sense to increase the "listenbacklog".
In (B) it would make sense to increase "maxconnections".
Of course, increasing "maxthreads" would also work, depending on the available 
memory resources.

Ha, and the admin of (B) could trick and set "connsperthread" = 1 ... ;-)

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] nsopenssl dropping connections

[Bernd Eidenschink]

Hi!

> AOLserver 3.5.6 limits the listen backlog to 32 new connections via
> the BACKLOG define in nsconf.h. If your server is getting new
> connections at a rate faster than AOLserver can accept them, and you
> reach the limit of 32 in the listen queue, connections will be
> dropped until the backlog drops below 32.

It would be interesting to know in what way the maxconnections and 
listenbacklog settings affect each other, as: is the maxconnections default 
setting of "100" always overruled by "32" of listenbacklog?

[...]

ns_section ns/parameters
# Number of sockets to use in listen call (default 32)
ns_param listenbacklog  32

ns_section ns/server/$server
# Max connections to put on queue
ns_parammaxconnections  100

[...]

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] Aolserver and ttrace

[Bernd Eidenschink]

Hi Agnieszka,

> I used ttrace version 0.9 and followed the instructions.
> So I copied init.tcl from ttrace-0.9/aolserver directory to
> aolserver 4.x bin directory.

did you

configure
make install

to actually also install "ttrace.tcl" to where your TCL libs are located?

-Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] Aolserver and ttrace

[Bernd Eidenschink]

Ups...

> Hi Agnieszka,

... To: AOLSERVER@LISTSERV.AOL.COM

Sorry for sending this over the list! It was not intended :(

-Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] adp files are not parsed

[Bernd Eidenschink]

Hi Razvan,

does the default configuration work for you (instead of fancy)?

Is the adp extension mapped for the server? (should be default)

ns_section "ns/server/${servername}/adp"
ns_param   map "/*.adp"  ;# Extensions to parse as ADP's.

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] port 80 cannot listen on

[Bernd Eidenschink]

Hi,

> the command nsd -t nsd.tcl -b mydomain:80 -u www -g www from root and i

instead of "mydomain:80" try "your-ip-address:80"

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] ns_adp_compress usage - again

[Bernd Eidenschink]

> Apache docs on it's approach to compression at
> http://httpd.apache.org/docs-2.0/mod/mod_deflate.html
> seem to say they do additional vetting of browsers before deciding to send
> something compressed.

hm, the rl_returnz module for AOLserver checks against IE5.5 and adds about
2000 extra whitespace chars on top of the content:

  /* IE 5 has an inexcusably buggy implementation
 on a refresh, it will (sometimes) discard about 2K of text! */

did you ever try rl_returnz to compare the results against ns_adp_compress?


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] Filename Problem, AS 4.0.9.

[Bernd Eidenschink]

> The installation crashed on one particular RPM,
> libstdc++-3.4.3-9.EL4.i386.rpm. If you wget
> http://www.lamarowen.net/files/unpacked/CentOS/RPMS/libstdc++-3.4.3-9.EL4.i
>386.rpm you will see why; it returns the text 'File Not Found' as the actual
> file instead of the file (that is actually THERE on the server).  AOLserver
> does NOT return a 404, either.  Here is the result of a test:

try:

wget
http://www.lamarowen.net/files/unpacked/CentOS/RPMS/libstdc%2B%2B-3.4.3-9.EL4.i386.rpm

I changed the plus char to "%2B".

Looks exactly like bug number 972994,
http://sourceforge.net/tracker/index.php?func=detail&aid=972994&group_id=3152&atid=103152

Quote: "To reproduce, try to download an RPM package such as:
libstdc++-3.3.2-1.i386.rpm." :-)

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] AOLserver facelift.

[Bernd Eidenschink]

>> This is my wish-list. Do whatever you consider appropriate.
>> It is *far* from complete, but I guess I have to start with something.
>> (I have no reason of hiding it, I'm not secret service):

* A debug/development mode that reveals all running filters,
  triggered filters,  better insight in memory consumption(s),
  encoding/charset stuff

* Filters/Procedures: Have them as a list, delete them,
  reorder them

* Don't having to restart the server only to re-read
  some config values. Maybe only as an option: For security
  reasons the current way is ok.

* Having a 'real' config file with _all_ default values.
  Maybe the current grep-thru-code way could somehow be simplified.

* Altering pageroot depending on HTTP host header plus
  dis-allowing any adp/tcl funcionality for these pageroots
  on a file level (no interpretation) as an option

* Server-wide shared variables

* Up-to-date patches for PHP (...) integration

* Some kind of watchdog functionality

...

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] AOLserver facelift.

[Bernd Eidenschink]

Hi Dossy,

> I think the reason behind both issues is because the proposed
> contributions weren't that great.  I'm sure this statement tweaks a lot
> of people, but I think it's the truth.

I'm not the expert to verify that on the C level, but the situation was: There
was a solution done by aD from Rob Mayoff and at least one version that
patched the aD Part into the main 3.x line. The aD version was unmaintained
with the fall of the company. And if your own company relies on AOLserver you
would always like to see things incorporated into the main line rather in
patches you don't know how many people use them (missing a lot of
collaborative testing effort) and if they would work with the next server
release.
I guess the encoding/charset thing was one of the main stoppers for a lot of
people to join the AOLserver community. So at that time it was not easy to
understand why existing _and_ working code was not incorporated.
Optimizations and code refactorings are always possible later on. JMHO.

> The upside is that in 4.1.x, expect to see a better way of handling the
> whole encoding/charset mess, along with better support for non-HTTP
> protocols in the core.  Done right.  (Yes, it's another major
> contribution from Jim Davidson.)

Thats great. The current discussion, your RoadMaps, Feature Lists etc. reveal
a lot of information. It would be great to have developments like these from
the Core developers being summarized early on the Wiki!

> > No matter if the changes are led by a Core Team or a Steering
> > Committee, the important step would be decisions at all.
>
> For better or worse, I think I've been good about making decisions and
> following through as best I can.

Of course! I like the activities that are going on! With "decisions at all" I
specifically meant the result of all of what is being talked about in the
moment. Everything is a process (tm).

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] AOLserver facelift.

[Bernd Eidenschink]

> On Monday 07 February 2005 16:51, Vlad Seryakov wrote:
> > I just gave up participating in SF tracker thing.
>
> Sad.
>
> Well, this is what I said... People go away or do nothing...
> We need the "vision" statement.

Exactly. As you wrote before: "acceptance or rejection of future ideas".

It was very very sad, e.g. that it took years to incorporate one of the
different approaches for encodings/charsets that have been around (from aD
and others), or that lengthy discussions (e.g. the "Support for non-HTTP
protocols" back in August 2004) are left open without a result.

No matter if the changes are led by a Core Team or a Steering Committee, the
important step would be decisions at all. Stability can become a fetish ;-)

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] AOLserver facelift.

[Bernd Eidenschink]

Hi Dossy,

> It didn't even get as far as doing a plain CVS import of AOLserver
> sources.  Go look at the SourceForge project for yourself:
>
> http://sf.net/projects/opennsd
>
> What's changed in the last four years that suggests another fork attempt
> will be any different?

This fork was no success because even the people behind this attempt did not
like that idea for the usual good reasons. If I remember correctly it was
mainly an attempt of developers that totally rely on AOLserver, the OpenACS
toolkit developers. The ArsDigita patches crucial for the toolkit did not
make it (fast enough?) into the core, so the attempt was understandable. Even
other attempts have been made that avoid the "fork", like including some of
AOLservers virtues into Apache or TCL. This seems still a subject:
http://openacs.org/forums/message-view?message%5fid=263944

And around that fork attempt AOLserver 4 was announced, early enough to slow
down the commitment to fork
(http://openacs.org/forums/message-view?message_id=51660 as one example).

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] The 'weird 0 length problems' revisited

[Bernd Eidenschink]

Hi Dossy, Jim,

thanks for your answers.

Dossy: We did not count HEAD-requests, only the GET's and do not stream.
Jim: Indeed we use rl_returnz on this server, so we already tried it
some time without it, but it made no difference.

It's similar the same like in this original thread:
http://www.mail-archive.com/aolserver@listserv.aol.com/msg05617.html
Every piece of content goes (finally) through ns_returnfile or ns_return
(or rl_returnz).

On one bad day we had about 3.5% served 0-byte-requests.

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

[AOLSERVER] The 'weird 0 length problems' revisited

[Bernd Eidenschink]

Hi,

last year there was a thread concerning a 'weird 0 length problem':
Wojciech Kocjan discovered requests served with a Content-Length
of "0". As I see it there was neither a solution to that problem nor a
bug to be found at that time.

I found this behaviour on one of our production machines, exactly like
depicted in the old thread (reload returns the page), and very hard to
debug, too. It's a 3.5.x. (Also seen, but less frequent on a 4.x due
to minor load)

My question is: Do you see (no, just a jew, dozens...) lines with
Content Length 0 in your accesslogs?
(Smth. like "grep '200 0' *.log"...)

I'm just curious,

Regards,

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

[AOLSERVER] ns_returnfile

[Bernd Eidenschink]

Hi all,

I'm using ns_returnfile to allow users downloads with filesizes of about
20 megabyte. Based on its return value "1" I nsv a successful download
and block repeated attempts to download the same file.

How reliable is ns_returnfile's information of a successful download?
Can you think of situations where a download stops/breaks and
ns_returnfile reports a success?

Regards,

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] off-topic: AOLserver development costs $531,245

[Bernd Eidenschink]

Hi Nathan,

> I bet the numbers would look even better if you looked at only the core
> Web Server code, and didn't count all of the associated modules that are
> checked in under $TOP/aolserver.

Thats true.
And a new-economy-salary-equalizer-parameter is missing, too :-)


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

[AOLSERVER] off-topic: AOLserver development costs $531,245

[Bernd Eidenschink]

Hi,

a little bit off-topic, but I wanted to point you to
www.koders.com

It's something like Google(tm) for sourcode, so you
can search for ns_gethostbyaddr, if you like ;-)

Funny how they fetched all the content of repositories,
e.g., if you really look for ns_gethostbyaddr it offers
a link "Project info: AOLserver" with nice stats,
e.g. the "Labor Cost to Develop:"  $531,245.
and of course you can recalculate that.

Enjoy,

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> 
with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: 
field of your email blank.

Re: [AOLSERVER] Max file size upload

[Bernd Eidenschink]

Hi,

in the context of Khy's question: Is it safe to say:
Always set both "maxpost" and "maxinput" to the same values?
Or else there are any side effects...?

Is "maxpost" still relevant? It's still in the code, but in
what way does it affect "postings"?

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] Max file size upload

[Bernd Eidenschink]

Hi,

try:

---

ns_section "ns/server/${servername}"
ns_param   maxpost  1024*1024*10  ;# 10 MB  (default: 256k)

---

Regards,

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] Error logs not rolling?

[Bernd Eidenschink]

Hi Janine,

> Actually I think this should do as the access log code does and roll
> the log like one would expect.  But as long as it is the way it is, I
> guess my script is just fine.

if you have "logrotate" on your machine, you might also give that
a try. I place a file with content like this in /etc/logrotate.d/:

/opt/chroot/nsd/log/server.log {
rotate 30
compress
daily
dateext
maxage 36
missingok
copytruncate
size +4M
}

You can configure it in many different ways, like compressing files, you
can add your own post- or pre-processing scripts and more.
I use a script that watches the server.log for different regular
expressions and mails me if they match; as the script stops when the
watched file gets removed ("is rotated"), the copytruncate argument
helps me in this case.

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] Fix to (known) nsopenssl issue

[Bernd Eidenschink]

Hi,

just FYI: "It goes on forever!"

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=29964

Not the same, but similar... :-)

cu
BE


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] nsopenssl debugging

[Bernd Eidenschink]

Hi,

> My symptoms are on something like Aolserver 4.01 + nsopenssl 3 beta 17
> or so. I only have about 15 users on my production box, and about once
> a day, the connections get gradually taken up, and the load gradually
> rises, until it's unreachable.

I see this with a AOLserver 4.0.5, nsopenssl 3 beta 17, Linux Kernel
2.6.5-x (SuSE).

netstat shows about 25 connections on the https port in CLOSE_WAIT
state, right now e.g. all owned by googlebot-crawlers (...)
The connections stay and stay open and load increases very slow
(fortunately). I restart the server manually every 3 or 4 days.

I'll try the latest nsopenssl beta soon, as I wasn't able to check it
out the last days due to cvs problems at sourceforge.

Regards,
Bernd.



tcp 0 0 linux.site:https  crawl34.googlebot:39207 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:39973 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:58923 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:63281 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:58125 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:55057 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:40729 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:51552 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:47722 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:36714 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:52548 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:64590 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:37710 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:63317 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl35.googlebot:46458 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:51551 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:41890 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:65184 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:58283 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:56465 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:56289 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:53224 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:40686 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:54767 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:54768 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:34299 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:41976 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:46335 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:62147 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:46538 CLOSE_WAIT
tcp 0 0 linux.site:https  crawl34.googlebot:49112 CLOSE_WAIT


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] Support for non-HTTP protocols

[Bernd Eidenschink]

>   a. YES, I like AS to be more powerful multiprotocol
>  server instead of being a powerful http-server only.
>  And, I like proposal X better than Y because of Z.

I vote for (a).

I read both discussions from the sourceforge-links you provided and I
see no "jealousness" or "vanity" between the two patch-providers, so I'm
optimistic they will provide good information/arguments for a discussion
with a final vote.

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] file upload corrupts jpg images

[Bernd Eidenschink]

> The convert command works fine on the Aolserver 3.4.2 server.
> >> I have two servers running: one is aolserver 3.4.2; the other is
> >> aolserver
> >> 3.5.0.

just a try:

please go into the modules/tcl directory of each of your installations
and do a

grep translation *.tcl

if one of them lacks the line

form.tcl: fconfigure $fp -translation binary

thats the problem.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] file upload corrupts jpg images

[Bernd Eidenschink]

> I am running RH 9.0 on a linux box.
>
> I do not know if the file is corrupted before or after it is moved from
> the temporary location. My understanding is that the temporary file is
> deleted when the process stops. I am not sure how I would find out.

You might try to run a checksum tool over the file, e.g. "md5sum" (a
tool found on most linux systems) - on the correct and corrupted version
of the file (call it with 'exec').

Or you utilize the standard TCL library (http://tcllib.sourceforge.net)
There are packages and commands like:

::jpeg::debug file
 Prints everything we know about the given file in a nice format.

::md5::md5 ? -hex ? [ -channel channel | -file filename | string ]
(...) If the -filename argument is given then the file is opened, the
data read and hashed and the file is closed. (...)

With the md5-hashes you can look for the step where the image gets
corrupted.

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] ANN: New project leader for AOLserver

[Bernd Eidenschink]

Hi Talli,

> bah. i'm sick of the "we need php/python/brainfuck in AOLserver to be
> successful and gain mindshare."

hm. it was a discussion about strategy ("the focus"). Of course it's easier to
abort that discussion and directly start with graphic design and IA.
And when it comes to the marketing documentation you simply write your point
of view:

> i'm really quite sick of this argument that AOLserver is hard to pitch.
> i personally have no problem doing so whatsoever. in fact, i usually
> win projects on the sole basis that we are wizards in AOLserver and the
> vast majority of these projects are with clients who had never heard of
> the system before I presented it to them.

that's great, really. Every won pitch and success is a success for AOLserver.
But a more known AOLserver would ease even your business. It's not true that
being a whizard in some technology opens all doors, or is it that you want to
tell? Or that all others are lame ducks if they loose the pitch?

> so i am happy to volunteer to design and build the new AOLserver.com on
> OpenACS (using AOLserver of course) with the latest OpenACS bugtracker
> (which is a lot better than Sourceforge's crap, and bugzilla for that
> matter) and write up the marketing documentation appropriately so that
> it's attractive and interesting to newbies. i am not volunteering to be
> the webmaster nor to maintain any of the technical docs, but I will be
> happy to do the graphic design and the IA.

IA = Information Architecture.
A profession many clients don't understand and therefore don't want to pay
much money for. In the famouse IA book (O'Reilly) the authors write, that
there is a point in client meetings where suddenly the importance of their
work is understood: One person stands up and shouts "Bah. I'm sick of the
discussion. Let's start, where's the problem?!"

Reminds me of that, sorry.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] ANN: New project leader for AOLserver

[Bernd Eidenschink]

> Ok, fine with me! But, if you want to tell this to your potential users,
> you must also give them a reason to switch. If the reason to switch
> is not to use TCL and use the AOLserver API, what is it? Users won't
> switch because they can leverage the vast knowledge they already have,
> for the simple reason they can already do that on Apache.

Oh, there are a lot of people out there that will give AOLserver a try
if only they are able to use it with something they need anyway, let's
assume PHP for the sake of that thread. I talk about curiousity, giving
something a try, learning and understanding something, 'being
different', -- if it works as stable as with Apache, why use Apache?
Please don't get me wrong: I don't want to insist on the PHP-train, but
I think this big-budget-decision-thinking and rational-arguing-thing is
strategically wrong.

There are more reasons than superiority, best-of-breed, premium,
world-class, amazing - bla bla bla bla - features for people to prefer
one thing over the other. Things evolve. Remember how unthinkable BAD
PHP was in it's old days? Today you can talk about it and and win the
competition, because the little son of your potential customer runs the
family homepage with it.
In the meantime we
* pay more for webspace, because our server is unknown
  (not good for people who want to give it a try)
* are confronted with 'Tcl - what?' questions
  (not sexy for people who want to give it a try)
* have _two_ arguments: '...backbone of the largest and busiest...'
  and '...for large scale, dynamic...'
  (not enough for people that google hours for finding out how
  to do virtual hosting)

I'm not arguing for implementing PHP to change that. I simply hope that
the Community plays a bigger role in the future of AOLserver and it's
features.

> So, what advantage does it then have to run PHP (or language x) on
> AOLserver instead of Apache? This a critical in the communication to potential
> users, since, likely they already use Apache.

One advantage is that an AOLserver administrator, confronted with the
requirement to offer PHP (for WHATEVER reason), would simply add PHP to
AOLserver with no need to install Apache. Because it's AOLserver he/she
is competent in, not Apache. This scenario is the opposite of yours,
where the person already likes and runs AOLserver and does not have to
be convinced.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] ANN: New project leader for AOLserver

[Bernd Eidenschink]

Hi Daniel, Bas,

I agree with most of your arguments. But I don't think we should think
of AOLServer as a "pathetic wannabe" should it allow to combine it with
PERL, PHP, HypeXY: If you don't need it, don't use it. People will
notice that use of TCL is superior for many things but they are not
bound to it. And there are people using AOLserver simply as a proxy to
do something completely different than serving web pages, that's
exciting.

There are successful combinations of Apache, PHP, Java, Perl - this
flexibility is part of the Hype. Too much people have to spend too much
time everyday to convince people of AOLserver, TCL, or AOLserver _and_
TCL - the reasons for it are widely known.
Hypothetically: why relax and wait until Apache2 is equally fast and AOL
itself drops use of AOLserver? Why not lay groundwork for its openness
now? Personally I don't mind using a tool that's part of a hype - if
it's AOLserver, the better!

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] ANN: New project leader for AOLserver

[Bernd Eidenschink]

> - It's not L.A.M.P. People are sceptical to changes of the way of doing
>   things and often act like sheep.

and

> Clearly AOLservers biggest competitor is PHP. Were the current website
> falls short is to explain why people need to choose AOLserver over
> Apache/PHP. I.e.:
>

PHP works with AOLserver, at least some folks use it. I don't know how
well maintained the bridge ist, nor if PHP can be compiled with 4.0, but
supporting PHP would be absolutely no harm. I would think of it as the
honeypot. It would allow people to use and maintain their (needed) PHP
applications and let them eventually get used to TCL.

> - Perhaps ridiculous, but it happened several times that I had to explain
>   to some people that AOLserver was open-source. They had made some quick
>   conclusion from the .com address.

;-) I would turn peoples attention to

'mysql.com' or 'zend.com' (which did major contributions to the
php-core).


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

[AOLSERVER] Install Script ('nsd-installer')

[Bernd Eidenschink]

Hi,

I wrote a small bash install script for a chrooted AOLserver 4 to
automate this task. You can fetch it here:
http://www.kinetiqa.de/aolserver/

Right now it is some kind of blueprint, but it works and allows others
to extend it.

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] nsdb module

[Bernd Eidenschink]

> > > It might help to post your entire config .tcl and more of the server
> > > log at startup.
> >
> > no, the line for the driver was there, it was another mistake on my side.
> > I was blind, so blind. Sorry. It works now...
>
> Don't keep us in suspense!  What was it?  :-)  It's always good to know
> what the symptom of mistakes are (and how to possibly correct them).

ok, ok :-) but it's embarrassing...
I had two config files in my directory with similar names. I tested
other things with one of them and did not really focus on the db driver.
Then added it hastily, and, ... wrote 'nsdb.so' as the driver (and the
module) instead of nspostgres.so. When I started the server and read the
error messages, I became aware of the mistake, but 'corrected' always
the other of the two config files I did not use for the server
startup... (...) (...) ;-)


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] nsdb module

[Bernd Eidenschink]

> Where's the line that looks like:
>
> ns_param postgres ${bindir}/postgres.so
>
> ???
>
> It might help to post your entire config .tcl and more of the server
> log at startup.

no, the line for the driver was there, it was another mistake on my side.
I was blind, so blind. Sorry. It works now...

___
... and the winner is... WEB.DE FreeMail! - Deutschlands beste E-Mail
ist zum 39. Mal Testsieger (PC Praxis 03/04) http://f.web.de/?mc=021191


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

[AOLSERVER] nsdb module

[Bernd Eidenschink]

Hi,

it must be something simple, but I don't see it:

I placed the nsdb module under the modules-section of the config
file, like:

ns_section "ns/server/${servername}/modules"
ns_param nssock ${bindir}/nssock.so
ns_param nslog ${bindir}/nslog.so
ns_param nsdb ${bindir}/nsdb.so

When starting AOLServer 4.0 I get:

Notice: modload: loading '/nsdbin/nssock.so'
Notice: modload: loading '/nsdbin/nslog.so'
Notice: nslog: opened '/servers/server1/modules/nslog/access.log'
Notice: modload: loading '/nsdbin/nsdb.so'
Warning: modload: could not find Ns_DbDriverInit in /nsdbin/nsdb.so
Error: dbdrv: failed to load driver 'postgres'

Shouldn't it call Ns_ModuleInit instead of Ns_DbDriverInit?
Or what else is going wrong?

Maybe it's a compile thing? (gcc 3.3.3, gmake 3.80)


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] report OpenSSL library version?

[Bernd Eidenschink]

Hi Andrew,

> Under Linux, is there some more general way to answer the question,
> "Exactly what shared libraries is this running process actually
> using?"  Some command line tool, or a particular thing to check under
> gdb?  Or some C function which I could call from inside AOLserver
> itself?

maybe:

on the binary:
ldd /the/path/nsopenssl.so

during runtime:
lsof /the/path/nsopenssl.so
lsof | grep nsd
lsof -p 

and the huge answer:
strace ./nsd -u ... -g ... -f -t ...


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] not recognizing extensions aolserver-4.01

[Bernd Eidenschink]

> I meant the return of index.html when I request http://server/index
>
> It does not return the index.html (which exists) and returns
>
> The requested URL was not found on this server.

Yes, i meant the same: it's correct that it does not show up, as 'index'
alone is

a) no directory (where the server could look for a $directoryfile)
b) no existing file (you request 'index' without an extension)

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] not recognizing extensions aolserver-4.01

[Bernd Eidenschink]

Hi,

> So no page shows for http://server/index
> While there is an http://server/index.html

should it serve the 'index.html'-file here? '//server/index' could also
be a directory, a 404 will be served if neither a directory 'index' nor
a file 'index' exists.

and '//server/' simply returns $directoryfile, like expected.

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

[AOLSERVER] FYI: "AOLserver meeting" in Regensburg/Germany

[Bernd Eidenschink]

Hi,

I just want to announce "The Regensburg AOLserver meeting" (this is our
working title so far...) on Thursday, the 8th of April.

A few people will come together and talk about the way they use
AOLserver - for projects, for business, for fun. This is a chance to
discuss, learn and get to know each other.

It takes place in Regensburg (Bavaria), Germany, where our company is
located; see our current Agenda below.
If you want to join, please register (http://www.r2010.de) so that we
know how much coffee we have to provide :-)

Regards,
Bernd.

*
Current Agenda
*

(1) Who is who :-)

* Mapping e-mail addresses to faces

(2) Presentations of Projects, code, etc.

* Zoran:
  - The ThreadTrace project
  - AOLserver as Application Server
"PresStore: Scalable Data Management Solution"
* Bernd:
  - compiling_and_chrooting_aolserver.sh
* Dirk:
  AOLserver in the context of
  - .LRN as professional web-application or
  - OpenACS as Rapid Development Platform

(3) "AOLserver to the masses"

* "Don't be afraid of AOLserver!"
  Pros and cons for SysAdmins that
  never used AOLserver
* Top-10-wishes of the AOLserver community
* Libraries or "How to avoid reinventing the wheel"
* Bernd/Dirk about their AOLserver article

(4) Open Discussion

* How do the attendees use AOLserver?
* AOLserver as Application- and Webserver
* How do the attendees use TCL in/with AOLserver?


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] Threads and lifetime

[Bernd Eidenschink]

Hi Elizabeth,

thanks for your reply. What I don't know right now is how to interpret
(1) against (2) (see below).
(1) started Feb05 and the TIME column reports 00:00:00, whereas (2)
started Feb15 and reports 00:38:00.

Does this mean that (1) is a thread waiting for some ns_schedule_proc
(e.g.) job whereas (2) is a connection thread doing heavy work?

On a development machine hitting AOLserver with concurrent users (using
http_perf) the server started finally (and temporarily) 20 threads with
about 4-6 doing nothing (0% CPU).

Regards,

Bernd.



# ps -afx -o user,pid,ppid,start_time,time,command

  PID  PPID START TIME COMMAND
(1)
22185 22184 Feb05 00:00:00  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]

(2)
32514 22184 Feb15 00:38:00  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

[AOLSERVER] Threads and lifetime

[Bernd Eidenschink]

Hi,

below is what I get on one of my servers, running ps (I shortened
"COMMAND" to make it readable).
This is from AOLserver series 3. I don't know exactly how to interpret
"TIME", especially when looking at "START".

Sure, after a (configurable) while (or when using ns_eval), one or more
threads are re-started (explaining the wide PID-differences), but,
especially as I use ns_eval very often for some API's, it does not
explain the gaps of the "START" column.

How to read this? Thanks!

Bernd.

--

# ps -afx -o user,pid,ppid,start_time,time,command

  PID  PPID START TIME COMMAND
22181 1 Feb05 00:01:28 [...]/nsd -izt [...]/c.cfg [-u / -g]
22184 22181 Feb05 00:00:00  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
22185 22184 Feb05 00:00:00  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
22186 22184 Feb05 00:03:45  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
22191 22184 Feb05 00:00:00  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
22192 22184 Feb05 00:08:43  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
22195 22184 Feb05 00:03:15  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
16197 22184 Feb14 01:03:42  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
20454 22184 Feb14 00:53:41  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
24594 22184 Feb14 00:47:37  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
29495 22184 Feb15 00:43:17  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
30635 22184 Feb15 00:54:57  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
32514 22184 Feb15 00:38:00  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
 9750 22184 Feb16 00:23:25  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
10646 22184 Feb16 00:21:26  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
10682 22184 Feb16 00:22:09  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
10977 22184 Feb16 00:21:06  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
12620 22184 Feb16 00:19:40  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
12621 22184 Feb16 00:20:12  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]
30210 22184 Feb17 00:05:05  \_ [...]/nsd -izt [...]/c.cfg [-u / -g]


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] ns_db gethandle -timeout -- anyone notice this is kinda busted?

[Bernd Eidenschink]

Hi Dossy,

we are specifying it within our db api for every call (-timout 20) and
never use or used -1. So we won't be affected by your change (which
makes sense to me, nevertheless).

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

[AOLSERVER] It's not AOLserver :-)

[Bernd Eidenschink]

Hi,

just a quick note on an oddity that can make you sweat and waste time:
On different sites running AOLserver we had the problem that many users
experienced missing images (jpegs) and after a while the browsers
stalled - even more -- the operating system refused to work properly.
(The JPEG-Of-Death... ;-)

Unfortunately this problem is related to the most used browser (IE)
on the still used W95/98 series. So it was urgent to fix it.
Could it be related to AOLserver? Unlikely. Could it be related to
some underlying tool like Image Magick? But why only IE on Win?

It seems like the problem is related to metadata, comments,
EXIF, ICM, IPTC... Data stored by Photoshop, digi cams, and so on.
Removing that data from images uploaded by users not only
greatly reduced the image sizes but also showed up as the solution
to this problem.
If you always can remove these metadata may depend on your special
use case, but I guess in most cases it's safe to do so.

Two users mentioning the problem and two different approaches to strip
extended headers out:
http://blog.rot13.org/index.cgi/id_11
http://toadstool.se/journal/2003/01/02/site_maintenance_and_kermits_too.php

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] aolserver 3.4.2 configuration questions

[Bernd Eidenschink]

Hi,

> I also have a question regarding database access with
> the postgres driver. I have installed postgres 7.3.4
> and it works fine. However, I wasn't able to find no
> postgres.so file in any directory on my system. I
> thought aolserver would install it somewhere and I
> could use that location to find it. What I did find
> was a /usr/local/lib/postgresql/plpgsql.so file, and I
> added it to db section, but I don't think this will do
> the trick:-)

You are right, you have to compile the postgres driver!

Take a look at
http://sourceforge.net/project/shownotes.php?release_id=129749

Compile it and the result will be the correct nsxyz.so file.


Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] aolserver 3.4.2 configuration questions

[Bernd Eidenschink]

Hi Dino,

> It is very strange but I don't see no 'Notice: tcl:
> enabling .tcl pages'...
> And it also seems that my enabbletcl true statement in
> my config file has no influence!!
> I can change that string to everything I want..the
> server just starts and doesn't complain about even
> wrong things I enter there...
>
> Still not a clue as to what is happening. My log file
> contains nothing about it.

Try this:

1. global debug parameter

> >  > # Global server parameters
> >  > #
> >  > ns_section "ns/parameters"
> >  > ns_param   home$homedir
> >  > ns_param   debug   false

set "debug" to true, you will get more output into
the logfiles.

2. tcl configs
point to the aolserver-tcl-module-files
(change path to where they reside, see below)

#
# Tcl interpreter
#
ns_section "ns/server/${servername}/tcl"
ns_param   autoclose   true  ;# Close files the Tcl interp
opened
ns_param   debug   true ;# Names of files sourced is logged
ns_param   nsvbuckets  8 ;# No. of buckets to hold nsv's
ns_param   statlevel   0 ;# How many levels deep to run
tclstats
ns_param   statmaxbuf  100   ;# Maximum entries in tclstats
table

# this is an example library path.
ns_param  library  "/www/servers/${servername}/tcl"



(the library-path includes files like form.tcl, util.tcl etc.)

one is called "file.tcl" and contains the lines that produce
the output you are looking for:

-
set path ns/server/[ns_info server]
set on [ns_config -bool $path enabletclpages off]
ns_log notice "conf: \[$path\]enabletclpages = $on"
-


if these libraries are not sourced at startup, you will
not see any output.

Hope this helps,

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] nsd and memory leaks

[Bernd Eidenschink]

> On 2004.01.09, Andrew Piskorski <[EMAIL PROTECTED]> wrote:
> > On Fri, Jan 09, 2004 at 12:31:39PM -0800, Jim Wilcoxson wrote:
> >
> > > fragmentation problems - not necessarily leaks.  Just restart your
> > > server once a day/week.
> >
> > Which is still quite frequently, and thus very conservative.  From
> > what I've heard many people don't restart their AOLservers for months.
> > I tend to schedule mine to restart once a week just to be safe.
>
> $ ps -A -olstart,vsz,args | grep dossy.org
> Sun Nov 16 00:14:54 2003 32512 /home/aolserver/bin/nsd [...]
>
> A month and a half without being restarted and it's only 32MB large, AND
> it's using the supposedly leaky (or unpatched) version of nscgi.

# ps -A -olstart,vsz,args | grep nsd
Thu Jan 30 18:19:55 2003 134064 /opt/chroot_ext/nsd [...]

No need to restart it since then, as we use some ns_eval construct to
reload our tcl-api-files should they change (and no nscgi, but lot's of
nsv-stuff and postgres).

The first few weeks I was concerned about memory leaks, as after 47 days
the size of nsd was about 62 MB, two days later 76 MB, but, finally it
tended to that 130MB shown above (I just found a file where I did some
statistics with top).

# uptime
11:11am  up 388 days, 15:05,  1 user,  load average: 0.00, 0.00, 0.00

ok, this particular webserver is serving only about 5500 visits/month,
but I'm really very happy that AS can be so "carefree"...

The only moment you can never foresee is when AS grabs twice as much
memory because of its high water mark principle. But until it's not
running the OS into swap space it also doesn't matter.

cu
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] Generating Dynamic PDF's from AOLs

[Bernd Eidenschink]

> I want to create a PDF version of a database report for downloading.
> Does anyone know of a way to dynamically generate PDF's within
> AOLserver?

Hi Steve,

I use htmldoc (http://www.easysw.com/htmldoc) which should be
sufficient for most purposes. You are able to specify page
sizes (or use standard sizes like letter, legal, a4 and universal),
margins, password, dis/allows generation of links and even permissions
for the PDF: copy, no-copy, modify, print, no-print and others.

You simply include images (gif/jpeg) with img-tag and are able to
specify a compression level for the jpegs. It also allows some
iso-charsets (8859-1/-15). You call it interactively from the
command line or - without parameters - use it with a GUI.

Regards,
Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] aolserver 4.0 shutdown hangs

[Bernd Eidenschink]

Same on Linux 2.4.4 if I try to stop the Server from the foreground
mode with Ctrl-C. Needs to be 9ed.

Regards,
Bernd.

> Hi,
>
> I installed AS 4.0 GM on my solaris 2.7 box yesterday and the shutdown is
> hanging as well.
> I didn't notice this problem with AS 4.0 beta 5 on the same machine.
>
> Regards.
>
> Selon Jamie Rasmussen <[EMAIL PROTECTED]>:
>
> > I ran into this with 4.0 on Win32.  It looked to me like the culprit was
> > the global static var "stopped" in driver.c.  That variable is
> > initialized to 1, but it doesn't look like it is set to 0 anywhere.
> > There are a few places in the code that look like:
> > if (!stopped && !shutdownPending)
> >
> > Jamie
> >
> >
> > Jim Wilcoxson wrote:
> >
> > > We're having this problem with 3.4 on Linux 2.4.20 too.  It didn't
> > > happen very often with 2.2.20, but with 2.4.20 AS frequently hangs
> > > after an ns_shutdown and has to be killed.
> > >
> > > Just another data point.
> > >
> > > Jim
> > >
> > >
> > >>Hello all,
> > >>
> > >>i was trying to upgrade to aolserver 4.0 but noticed that every time i
> > >>do ns_shutdown from nscp the server
> > >>just hangs. It appeares that the problem is call to
> > >>NsWaitDriversShutdown(&timeout); during server shutdown.
> > >>Last version i used without any problem beta10 did not have this call.
> > >>While debugging i found that it hangs in Ns_ThreadJoin call, looks like
> > >>DriverThread exited already and pthread_join just hangs.
> > >>
> > >>
> > >>--
> > >>AOLserver - http://www.aolserver.com/
> > >>
> > >>To Remove yourself from this list, simply send an email to
> > <[EMAIL PROTECTED]> with the
> > >>body of "SIGNOFF AOLSERVER" in the email message. You can leave the
> > Subject: field of your email blank.
> > >>
> > >
> > >
> > >
> > > --
> > > AOLserver - http://www.aolserver.com/
> > >
> > > To Remove yourself from this list, simply send an email to
> > <[EMAIL PROTECTED]> with the
> > > body of "SIGNOFF AOLSERVER" in the email message. You can leave the
> > Subject: field of your email blank.
> >
> >
> > --
> > AOLserver - http://www.aolserver.com/
> >
> > To Remove yourself from this list, simply send an email to
> > <[EMAIL PROTECTED]> with the
> > body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject:
> > field of your email blank.
> >
>
>
> --
> AOLserver - http://www.aolserver.com/
>
> To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with 
> the
> body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field 
> of your email blank.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

Re: [AOLSERVER] Jim's Sunday commits and AOLserver 4.1

[Bernd Eidenschink]

...ups, 4.1, 4.1?

Sorry, maybe I'm the only one that did not notice the 4.0 Release :-)

Did I miss the thundering announcement of 4.0 or was it really
released so silent?

Anyway - great news!

Regards,

Bernd.


--
AOLserver - http://www.aolserver.com/

To Remove yourself from this list, simply send an email to <[EMAIL PROTECTED]> with the
body of "SIGNOFF AOLSERVER" in the email message. You can leave the Subject: field of 
your email blank.

[AOLSERVER] Current 3.5.x version | Documentation PDF V.4

[Bernd Eidenschink]

Hi,

are there plans to release a 3.5.x (3.5.9) version
of AOLserver in the Sourceforge download area?
The latest one is 3.5.6 and somehow I'm just able
to checkout something that does not let me compile
it from scratch (what is my fault, of course...).
Maybe some CVS pro can give me a hint.

And, one other thing. Last year I found one or more
PDF Documents mentioned in a discussion board
that documented AOLserver version 2. Complete
tutorials describing AOLserver virtual servers,
SSI (if I remember right), a database frontend.
Do that PDF's still exist? The community, new
AOLserver users/admins could greatly benefit
from them! As Version 4 comes with features
Version 2 in a similar way offered (notably
virtual hosting) the update effort of the
documents may not be that big?

Thanks,

Bernd.


--
AOLserver - http://www.aolserver.com/
To Remove yourself from this list: http://www.aolserver.com/listserv.html
List information and options: http://listserv.aol.com/

[AOLSERVER] access logs

[Bernd Eidenschink]

Hi,

I just tried "analog", the log analyzer tool, on a dozen of access.log's
my webserver produced and it spit out various lines it refused to
analyze.

The first problem was that the german locale was used to produce
the month part of the date, so March became "Mär" instead of "Mar". I
"solved" this quickly with emacs... Is this a usual problematic side
effect with using locales, a problem the analyzing tool has to cope
with or smth. that should be changed in the logging code of the server?

And I also found lines containing HTML fragments like these:

--

132.199.x.x - - [10/Mar/2003:12:12:21 +0100] "GET /lebensart/" bgcolor=
HTTP/1.0" 200 238 "" "Mozilla/4.0 (compatible; BorderManager 3.0)"

132.199.x.x - - [10/Mar/2003:10:23:43 +0100] "GET /lebensart/">" 404 637
"" ""

132.199.x.x - - [10/Mar/2003:10:23:43 +0100] "GET /lebensart/" align=
HTTP/1.0" 200 238 "" "Mozilla/4.0 (compatible; BorderManager 3.0)"

--

Obviously, the request part seems to be wrong:
"GET /lebensart/" align= HTTP/1.0"
"GET /lebensart/">"

Did anyone of you detect something similar?

AOLServer-Version used is 3.5.1.

Thanks,
Bernd.


--
AOLserver - http://www.aolserver.com/
To Remove yourself from this list: http://www.aolserver.com/listserv.html
List information and options: http://listserv.aol.com/

[AOLSERVER] Size of AOLserver process

[Bernd Eidenschink]

Hi there,

with all that great ways of caching data - nscache, nsv... - in AOLserver and
a strategy of caching as much data as possible, at what point would you say
the size of the processes becomes a "problem" for the application running
AOLserver? I don't think of physical RAM and swap, I think of AOLserver and
starting new threads, using ns_eval to propagate changes from APIs to all
other processes, responding to requests and similar things.

What are sizes you usually see on your systems?

This is something on my machine, for example:

PID USER PRI NI   SIZE RSS  SHARESTAT CPU  MEM TIME COMMAND
26591 nsadmin   17   0 70944  68M  2616 S 0.0 13.6   0:50 nsd
26594 nsadmin   15   0 70944  68M  2616 S 0.0 13.6   0:00 nsd
26595 nsadmin   20   0 70944  68M  2616 S 0.0 13.6   0:00 nsd
26596 nsadmin   15   0 70944  68M  2616 S 0.0 13.6   0:21 nsd
26601 nsadmin   15   0 70944  68M  2616 S 0.0 13.6   0:00 nsd
26602 nsadmin   15   0 70944  68M  2616 S 0.0 13.6   0:06 nsd

(and it runs without problems so far).

Bernd.

Re: [AOLSERVER] gzip compression

[Bernd Eidenschink]

Hi Gabriel,

> I played around with the rlreturnz stuff and it's nice and it works,
> but my main problem with gzip encoded content is that the browser waits
> until it's downloaded the entire file before it decompresses the data
> and begins rendering the page. A side-effect is that it makes pages
> appear to take longer to load, even though they take less time to
> download. Of course it depends on the content that was compressed in
> the first place, but for HTML pages it's a noticeable difference, which
> is why I never adopted the use of the gzip content encoding.

I think you win more than you loose, I have some statistics
from a gzip-module currently in development, with some good numbers:

[08/Jan/2003:20:12:00 +0100]20598 Bytes 3819 Bytes  0.01sec
[08/Jan/2003:20:12:38 +0100]20598 Bytes 3819 Bytes  0.01sec
[08/Jan/2003:20:13:13 +0100]11336 Bytes 2479 Bytes  0.00sec
[08/Jan/2003:20:13:53 +0100]18025 Bytes 3131 Bytes  0.00sec
[08/Jan/2003:20:13:54 +0100]18017 Bytes 3133 Bytes  0.00sec
[08/Jan/2003:20:13:57 +0100]18642 Bytes 3503 Bytes  0.00sec
[08/Jan/2003:20:14:04 +0100]13407 Bytes 2748 Bytes  0.01sec

The difference between compression level 5 and 9, e.g, results only in
about 150 bytes; rlreturnz.c uses 3, if I remember right. Using the highest
level and then caching it similar like the fastpath stuff would eliminate
thinking about processor load, too.
Compression IMHO is most of the time a big win for the vast majority of
web sites. I really don't "feel" that it takes longer to wait for the page.
I have some pages with very long result sets, 300k and more, that
can be compressed very good and take a fraction of the time loaded
and rendered than uncompressed.

Regards,
Bernd.

Re: [AOLSERVER] How to remove unsafe commands?

[Bernd Eidenschink]

> > #catch { set fh [open init_interp_code.tcl w] ; puts $fh $init ;
> > close $fh }
> >
> > in namespace.tcl, proc _ns_getinit.
> > Somebody ever tried this?
>
> What do you mean?  What exactly are you trying to achieve?

Well, the line above, if uncommented, creates a (big) file which holds all
procs sourced in after startup.

looks like that (example):

namespace eval {} {
variable tcl_version 8.3
variable ...
proc do_something {
...
}
(dozens, hundreds of procs)
...
}

The namespace.tcl routines create that set of sourced-in procs every time
_ns_getinit is called:
#   AOLserver will invoke this script in the
#   initial thread's interp at startup and
#   after an ns_eval to generate a script for
#   initializing new interps.


Is it a bad idea, maybe before use on the production system, to let AOLserver
create the file, byte compile the procs, adding "load libtbcload..." and
sourcing this file instead of running through the process every time?

> What do you mean?  What exactly are you trying to achieve?

Would that help speed up things?
Would it "hide" all procs from the info vars/procs/body... command to make it
(more) difficult to inspect vars, procs and environment at runtime?

Bernd.

Re: [AOLSERVER] How to remove unsafe commands?

[Bernd Eidenschink]

Thanks for your answers, Jeff and Dossy.

> Just do 'rename unsafeCmd {}'.  There is no need to do it at the C level.

Yes, I already rename some commands to "" / {} nirvana :-)

Thinking about it I think it is reasonably secure, the only command that I
don't know how to deal with so far is "info", as it is called with "info
procs", "info body" (...) in "namespace.tcl" and I need the "info exists"
part by myself.

Btw: Would it be possible and/or make sense to byte-compile _everything_? Just
thought about it when examining the line

#catch { set fh [open init_interp_code.tcl w] ; puts $fh $init ; close $fh }

in namespace.tcl, proc _ns_getinit.
Somebody ever tried this?

Bernd.

[AOLSERVER] How to remove unsafe commands?

[Bernd Eidenschink]

Hi,

how can "unsafe" tcl commands be removed or disabled? I think this would be
done before compiling Tcl, so that they are disabled at that level...
Reason is to protect (better) against someone that is able to infiltrate code
fragments to the running webserver.

Or could/should  that best be done with some kind of module that disables the
commands  at startup?

Thanks,
Bernd.

[AOLSERVER] sha1 // ns_sha1

[Bernd Eidenschink]

Hi,

I wonder why

::sha1::sha1 "hashing is fun"
(http://expect.nist.gov/sha1pure/ from Don Libes,
and standard tcl library)

produces

b9ff9778b351918b01deebd508cae98a558bfe01

whereas

ns_sha1 "hashing is fun" produces
(ns_sha1 from sourceforge)

B9FF9778B351918B01DEEBD508CAE98A558BFE01

the hash in upper chars.

What's "correct"?

Kind regards,
Bernd.

Re: [AOLSERVER] ArsDigita 13 encoding patches for 3.4.2?

[Bernd Eidenschink]

Hi Eric, Eric,

take a look at:

http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/aolserver/aolserver/nsd/encoding.c

"encoding.c" is new in AOLserver 4.0. There seems to be no downloadable
4.0 (beta) version at the moment/anymore but you can browse the CVS
repository.

The patches from Rob Mayoff offer more than what 4.0 looks right
know. 3.5 lacks "encoding.c", don't know if the undocumented ns/encoding
parameter 3.4.2 has is still there or not.
The Arsdigita patches allowed to specify the encoding on a page per page
basis, whereas the current or upcoming version will do this based on the
mime-type.

> not suggest that this has been corrected.  We are fervently hoping that
> international character encodings (particularly UTF-8 encoding) will be
> better supported in v4.0 ,

This would be _so_ fantastic. Especially some simple iso8559-x encodings.
The missing encoding support is the number one reason people still use
the ad13-patch-versions. Almost any project that needs something
different than US-ASCII (and UTF-8 databases) can not make use of
latest AOLserver developments, what is very sad, as the current
list of features and modules is better than ever.

> but I can't recall any specific comments from
> the AOLserver development team on the topic of encoding in spite of
> multiple postings by a variety of users...which makes me very afraid that
> this is not on the v4.0 feature list.  I would, of course, be happy to
> receive clarification to the contrary.  :-)
>
> Eric
> 
> Eric Larkin
> Chief Technology Officer
> bom.com
> [EMAIL PROTECTED]
> 650 937-1032
>
> > -Original Message-
> > From: AOLserver Discussion
> > [mailto:AOLSERVER@;LISTSERV.AOL.COM]On Behalf
> > Of Eric Lee
> > Sent: Wednesday, October 16, 2002 9:33 PM
> > To: [EMAIL PROTECTED]
> > Subject: [AOLSERVER] ArsDigita 13 encoding patches for 3.4.2?
> >
> >
> > Has anyone ported the ArsDigita "13" patches that allow one to specify
> > encoding to be used for data written from TCL pages to
> > AOLserver 3.4.2? Or
> > does anyone know if this has or hasn't been incorporated into 3.5?
> >
> > Thanks,
> >
> > Eric Lee

[AOLSERVER] adp/mapping

[Bernd Eidenschink]

Hi,

I started to register a filter and a proc that I want to be called
for every request. But my ns_log's are never called if I request
an adp file. Seems like the parser/map settings in the config resp.
the adp mechanisms "are stronger" than ns_register_filter/proc.

If I uncomment the adp settings in the config, it works (of course...).
I then can parse the adp file for myself and can handle errors etc.
with my own tcl routines.

What are the disadvantages of not using the default way of
handling adp files? Is the caching of the adp pages lost?

Or am I doing something wrong? I'm not familiar with the OpenACS
package, but I think they don't touch the adp settings in the
config file (parsers/mappings) and are nevertheless able to handle
adp pages with their own procs.


Thanks,
Bernd.

[AOLSERVER] AOLserver 4 / default encoding

[Bernd Eidenschink]

Hi,

am I right - after a quick look at the cvs code - that the default
encoding in AOLserver 4 will be iso8859-1?

In what way is it supposed to be changed, e.g., from within
an adp page? Am I able to tell "What comes in is
(expected to be) in XY-Encoding" and
"What goes out should be in YX-Encoding"?

Thanks,
Bernd.


(
ps, other topic: maybe I did smth. wrong but is it possible the
current gmake process (A4) breaks if I use a location different from the
default location /usr/local/aolserver? (maybe it was to early in the
morning for my laptop :*)
)

Re: [AOLSERVER] AOLServer corrupting Form data

[Bernd Eidenschink]

Hi Pierre,

> > I wrote a document about this a while back when I tried to make it all
> > work at ArsDigita: . However,
> > the code examples in that document apply to the ArsDigita versions of
> > AOLserver (the last of which is 3.3+ad13). I hacked those versions to do
> > character set translations. Probably none of the example code will work
> > in the standard versions of AOLserver.
>
> I pulled the charset-related changes out of 3.3+ad13 and forward-ported
> them to 3.4.2, replacing the few charset tweaks introduced in AOLserver
> 3.4.  The patch includes your document.  Any takers?
>
> So far it seems to work for us.  We only have to support UTF-8, but
> the patch fixed problems with form submission.

So far, the 3.4.2 version features the undocumented parameters:

ns_section "ns/encodings"
ns_param adp iso8859-1

in the config.

I cite Kriston Rehberg from a mail in this list (March):

"[...] So for AOLserver 3.x the only real way to handle this kind of thing
is to use nsd76.  There is
something in 3.4.2 called "encodings" in the configuration to force the
nsd8x (Tcl 8.x) UTF-8 encoder to Do The Right Thing with the aforementioned
eighth-bit-set bytes but you'd have to ask someone else precisely how that
fixes this thing."

You say to took Robs code and ported it to 3.4.2 -- did you do that for your
own version of AOLserver or did you commit it to the Sourceforge version?
It would be great to know that Robs code or smth. similar is going to be
implemented in 3.4.2 or 4.0 of the standard AOLserver release. This would
solve
so much problems. So far the "ns_param adp iso8859-1" seems to work most
of the day but you can never be sure... And supporting ONLY UTF-8 - as
encoding both of the webserver AND the DB - is also not always the best
solution for many people.

Kind Regards,
Bernd.

[AOLSERVER] libtbcload question

[Bernd Eidenschink]

Hi,

I'm trying to find the correct place to load "libtbcload", the library to
load ".tbc" files along with three tbc files.

Should I place load/source in /modules/tcl/namespace.tcl e.g. in
_ns_getinit?
This works, but I don't know if that's a good idea or if it is efficient;
and it is called after my modules are already loaded.

If I write a file with "load libtbcload" and want all the other modules+tbc
files be sourced before _ns_getinit is called, the server starts but on the
first
request it breaks with the "called a copy of a compiled script" message.


Regards,
Bernd.

[AOLSERVER] nsv_incr can cause server restart (?)

[Bernd Eidenschink]

Hi,

I use the nsv_incr command to count requests for
different file types, including gif and jpg images.

It worked well so far until I started to use big
html pages with about 20-40 (small) images (helper gifs etc.)
The browser rendered only about 5-10 images, sometimes
the connection broke (I also tested with wget and retries up
to 200 (...)).

After commenting out all the nsv_incr commands, all works
well and really fast. What is interesting is that during
the page requests the server restarted very often.
(The same behaviour came up some time ago when I used
nsv_incr on a variable that did not exist (my fault))

Does this mean to increase parameters like "nsvbuckets"
in the config file?

Bernd.

Re: [AOLSERVER] Unicode, I18N, 3.4.2 / 4.0

[Bernd Eidenschink]

Hi,

> tells UTF-8 to use two bytes, which means you lose!).  So for AOLserver
3.x
> the only real way to handle this kind of thing is to use nsd76.  There is
> something in 3.4.2 called "encodings" in the configuration to force the
> nsd8x (Tcl 8.x) UTF-8 encoder to Do The Right Thing with the
aforementioned
> eighth-bit-set bytes but you'd have to ask someone else precisely how that
> fixes this thing.

what is TCL 7.6? :-)
Hm, as far as I can tell, using the "encoding" parameter does the job, so I
think it's better to use nsd8x and switch to the 4.x server later.

You mentioned the EURO character:

> I should also mention on a related topic that the EUR character is almost
> certainly not going to work unless you choose and stick to the proper
> character set... Windows has its own codepage and ISO has these fake
> ISO-8859-15 and ISO-8859-0 character sets to try to handle EUR.  People
> editting content in MS Windows are not going to use the same EUR code as
> people writing a Tcl script on Unix.  Watch for that.

There's a good page explaining alle the "EUR" char problems here:
http://www.cs.tut.fi/~jkorpela/html/euro.html

Using some kind of
"ISO-to-Unicode-"#"-regsub-numeric-char-representations-ns_adp_puts"-
approach ... I can submit a ALT-0128 (EUR) char typed in a Windows
environment (using
"accept-charset='ISO-8859-1'" within the FORM tag and HTTP-Header and
META-Header (...))
to a Linux AOLserver installation, store it in Postgres (funny enough,
created with
standard "SQL ASCII" - encoding, psql shows me a "200" (?) value for each
stored
EUR character) and adp_putsing it back as "#128" char, able to view it as
EUR
character (VERDANA font on Windows).
The "#128" character is mentioned in the document above as "undefined; if it
works, consider it as an illusion" ;-) so maybe only I can see it ...

The other representations are € and € (besides € or
€ ).

Thanks,
Bernd.

[AOLSERVER] Unicode, I18N, 3.4.2 / 4.0

[Bernd Eidenschink]

Hi,

I don't know how to set up a combination of the latest AOLserver,
using the nsd8x Interpreter, and a Postgres 7.2 database, that allows
me to safely work with a charset of iso8859-1. Please don't throw
stones, I know this has been discussed very often ;-)

How to do it?

The problems I run into during my own tests and the problems that
other people have (I read through some threads on various boards) are:

A.
Using the latest server "out of the box", working with most European
characters will almost always fail using the typical string and regexp
functions that internally use utf-8 ... _IF_ you are returning (in http-
header or meta-tags) a charset of e.g. iso8859-1 (you have to
know what comes from forms/submits; you try to return "umlauts"
or any language depending chars and tell the browser of it).

B.
I got the tip of using an undocumented parameter in the config
file, that maps the iso8859-1 charset to ".adp" files (last year on this
list).
But this does not guarantee that all characters that _leave_ an
adp will be in iso8859-1 encoding (e.g. if I use combinations of
"ns_adp_parse -file" (where strings from the DB are regexped and
stringed and whatever) and return the string with "ns_return").
At least if all what comes in goes iso-ed to the DB, you could do
a workaround and translate all outgoing chars to "{" html code.
(There was an example of a ns_adp_puts function that does this
given by Harray Moreau on the list)... If you "escape" all the characters
that way, I assume, you would not no longer have to return a
charset header or charset-meta-tag.

C.
The uncool way: Using "charset=utf-8" outgoing, then also
expecting it incoming. A special character should come in as unicode
and tcl should treat it this way. The database must be installed with
unicode-encoding. You will run into performace problems and/or, maybe,
some unresolved topics of Postgres unicode-implementation as well.
I have not tried this yet, I merely assume this would work. What's
your opinion on that?

D.
Using an AD-patched version of AOLserver. The Problem: It's an
older version of the server, will it be kept up to date in the future?
(Of course, there's a large user base running it)

E.
I did not try --enable-recode and setting up a charset table for
utf8 -> latin1 and latin1 -> utf8 for Postgres. Maybe this would
work if you can guarantee all charsets coming from AOLserver
are coming as utf8 or iso8859x. Maybe, don't know.
I tried putting "SET ENCODING TO 'UNICODE'" resp. "LATIN1" in front
of every SQL statement in my test api (and took into account that it may
make a
difference if you are SELECTing or INSERTing) for telling the Postgres
server that the client uses unicode or latin with and without the
undocumented feature of (B). This only lead to error messages
noticing me of failed encoding translations. Every insert _always_
was logged with german special characters (umlauts) and never
as unicode characters (don't know if this is correct).

Will AOLserver 4 come with I18N support that solves all the problems
and what to do if you need a solution here and now?
Solution B? C?


Thanks for reading through this one...

Bernd.

[AOLSERVER] Q concerning Encodings

[Bernd Eidenschink]

Hi,

can someone please tell me if I'm right with the following assumptions:


(Using AOLserver 3.4 from 2001-05-17 and nsd8x)

a)

Whenever using e.g. a "regsub" on a String that contains characters
in iso-8859-1 format, the command treats it as Unicode and the
characters are more or less lost.
("ü" becomes "ü", for example)

b)

This does not seem to work:

(foo is a ns_urldecoded form variable)


set foo [encoding convertfrom iso8859-1 $foo]

# just to test...
regsub -all -- {ä} $foo {\ä} bar

set bar [encoding convertto iso8859-1 $bar]


The characters are lost, or more exactly: Unicode.
(Tested: If I'm switching the browser to the Unicode
charset they look like they should)

Isn't it possible to do iso-8859-1 output with the standard
AOLserver distribution when using ADP's and nsd8x?


Thank you,
Bernd.