> Hello,
> 
> On Fri, Jun 30, 2006 at 07:55:35PM +0200, Florian Wagner wrote:
>> attached to this mail is a patch for Pound 2.0.9 which adds code for
>> rewriting the "Destination" http header,
> 
> I'm afraid that this patch got ripped away by the pound mailing list. 
> Could you resend it inline, please?  It fixes a bug that was annoying
> me for quite a bit of time.

You're right. But at least the archives don't lose attachments. So you
can download it at

http://www.apsis.ch/pound/pound_list/archive/2006/2006-06/1151690135000


Regards,
Florian

> Aargh!  It looks like the list software you guys use rips away any
> attachments.  Here's the patch again:

Attachments can be downloaded from the list archives.

Simon

On Sat, 2006-07-01 at 02:35 +0200, Adam Borowski wrote:
> Hello!
> 
> It would be really a shame to have 2.1 force people to recompile just
> to get working WebDAV.  While it's not a real hoop for people
> installing pound from source, those who use distro-provided packages
> will have to look around.
> 
> Also, another problem is that --enable-msdav enables not just WebDAV
> but a bunch of proprietary M$ extensions as well.  This is not what
> an average user wants.
> 
> 
> An obvious fix would be to change the #define to a config-file
> option.  It modifies just a single line in the code anyway.  However,
> I went further -- this patch distinguishes between 4 levels:
> 
> ## WebDAV
> ##      0       disabled
> ##      1       fake (nothing but LOCK and UNLOCK)
> ##      2       enabled (partial RFC compliance, implies xHTTP)
> ##      3       MSDAV (Microsoft extensions)
> 
> I can't really find out what enabling just LOCK and UNLOCK would be
> good for; it doesn't hurt to preserve this option, though.  Setting
> WebDAV to 3 will work the same as --enable-msdav does right now, and
> setting it to 2 will turn all non-MS-only methods that pound knows.
> 
> I believe that having "2" separate from "3" is quite a good idea --
> at least grepping the logs for "CONNECT" can be telling :p
> 
> 
> Cheers and schtuff,

Thanks for the suggestion - I'll put it high on the wish list. Right now
we're in feature freeze mode until 2.1, but we'll get to it right
afterwards.
-- 
Robert Segall
Apsis GmbH
Postfach, Uetikon am See, CH-8707
Tel: +41-44-920 4904

On Fri, 2006-06-30 at 19:55 +0200, Florian Wagner wrote:
> Hi,
> 
> attached to this mail is a patch for Pound 2.0.9 which adds code for
> rewriting the "Destination" http header, which is used during svn copy
> and svn move actions (those translate into a http COPY request).
> Rewriting is done for https://... URIs to a simple http://... This
> allows Pound to be used for providing https access for a http SVN
> repository, without losing svn copy and move functionality.
> 
> The patch includes an additional config file instruction (named
> "FixSvnCopy") and additions to the man file, which document the new
> instruction.
> 
> My C skills aren't that great, so the patch may well be suboptimal, but
> at least I can confirm that it has been working on my server (hosting
> perhaps 10 repositories with low traffic) for the last six months or so.

Nice, but I have a few questions:

- why limit this to COPY? At least MOVE uses the Destination header as
well, so why not handle all such headers?

- why just replace https with https? Shouldn't the host/port parts be
replaced as well with the back-end info as well?

I'd be happy to hear what you think. In any case this is for after 2.1 -
we're in feature freeze right now.
-- 
Robert Segall
Apsis GmbH
Postfach, Uetikon am See, CH-8707
Tel: +41-44-920 4904

> - why limit this to COPY? At least MOVE uses the Destination header as
> well, so why not handle all such headers?

During writing the patch I only came across COPY requests. If I remember
correctly a svn move translates into a DELETE and COPY. I could, of
course, be wrong...

> - why just replace https with https? Shouldn't the host/port parts be
> replaced as well with the back-end info as well?

Because I didn't think that far... ;)


Regards,
Florian

On Wed, 2006-07-05 at 10:28 +1000, Mike Withers wrote:

> and when I do:
> root(at)black:~# /usr/local/sbin/pound -c
> it reports:
> 
> Segmentation fault
> 
> Is it my config file at fault or something else?

Usually you pass your config file as the argument to -c.  So it's not
your config file but rather the lack of one ;-)

/usr/local/sbin/pound -c /path/to/my/config.file

Still, it's not correct for pound to segfault under any circumstances,
so that's a bug.

Regards,
Cliff

On Thu, 2006-07-13 at 21:30 -0500, Omar El-Domeiri wrote:
> I would like to setup a pound service entry that matchs on the root
> URL, that is a GET / or a  GET /. ... but I can not seem to write such
> a rule.  Any ideas?
> 
> -Omar

URL "^$" or URL "^\.$" would probably do the trick. In any case, telling
us what you tried and what doesn't work would be helpful...
-- 
Robert Segall
Apsis GmbH
Postfach, Uetikon am See, CH-8707
Tel: +41-44-920 4904

 Hello,
 
I am very pleased to discover your product : pound. And as I understand
right it's only working on a Unix/Linux like OS software.
Maybe I may ask you an advice.
 
I have 2 Web production platform's running on Windows 2000 server and IIS
and need the option to  redirect: requests distributing the URL  among the
two servers according to the requested URL. On this moment we only have one
WAN IP-address and all websites running on one webserver.
To migrate to a second webserver or to make a separation of the websites, a
second webserver is neccesary. 
I wonder how to manage that with your wonderfull product: pound.
 
As example: 
Directly behind the SDSL-router a Linux-platform is needed running POUND.
Pound has programmed to work as followes:
URL: www.abc.com is redirected to Webserver-1 and URL: www.xyz.com is
redirected to webserver-2 
 
Is that right?
 
Will there be small "stand-alone" platforms running Linux, like a lot of
Firewalls as NetAsq are working?
 
Thanks for your patiance reading and answering this email; I will appreciate
that very much.
 
For now, with regards,
 
Dirk Dekker

-----Oorspronkelijk bericht-----
Van: pound(at)apsis.ch [mailto:pound(at)apsis.ch] 
Verzonden: vrijdag 14 juli 2006 22:23
Aan: dirk dekker
Onderwerp: Welcome to [Pound Mailing List]

Hello "dirk dekker" <hf.dekker(at)chello.nl>,

you are now subscribed to Pound Mailing List.

Send your mails to: pound(at)apsis.ch

If you want to unsubscribe from Pound Mailing List, send a mail with subject
(not body!):

unsubscribe 

to: pound(at)apsis.ch

Yours, roseg(at)apsis.ch

Return Receipt
                                                                           
   Your       RE: Welcome to [Pound Mailing List]                          
   document:                                                               
                                                                           
   was        jbrown(at)camsys.com                                            
   received                                                                
   by:                                                                     
                                                                           
   at:        07/14/2006 04:53:46 PM                                       
                                                                           

On Fri, 2006-07-14 at 22:28 +0200, Hennie Dekker wrote:
>  Hello,
>  
> I am very pleased to discover your product : pound. And as I understand
> right it's only working on a Unix/Linux like OS software.

No, it works on any "sufficiently" POSIX-like system, including cygwin.

> Maybe I may ask you an advice.
>  
> I have 2 Web production platform's running on Windows 2000 server and IIS
> and need the option to  redirect: requests distributing the URL  among the
> two servers according to the requested URL. On this moment we only have one
> WAN IP-address and all websites running on one webserver.
> To migrate to a second webserver or to make a separation of the websites, a
> second webserver is neccesary. 
> I wonder how to manage that with your wonderfull product: pound.
>  
> As example: 
> Directly behind the SDSL-router a Linux-platform is needed running POUND.
> Pound has programmed to work as followes:
> URL: www.abc.com is redirected to Webserver-1 and URL: www.xyz.com is
> redirected to webserver-2 
>  
> Is that right?

Almost. Pound does NOT redirect, but passes the requests.

> Will there be small "stand-alone" platforms running Linux, like a lot of
> Firewalls as NetAsq are working?

If interested we could probably do something like that, as is probably
the case for quite a few members of this list...
-- 
Robert Segall
Apsis GmbH
Postfach, Uetikon am See, CH-8707
Tel: +41-44-920 4904

On Thu, 2006-07-20 at 16:36 +0200, Michael Stegemann wrote:
> Hello List.
> 
> I just get startet with pound 1.9 and everything works fine.
> But I have big problems with sessions from clients, that came from AOL.
> 
> This is my config-file:
> 
> UrlGroup  
> ".*"                                                                     
>                                                                          
>                                                                          
>                                                                          
>                                                                          
>                                                                          
>                                                                          
>                                        HeadRequire Host ".*(spion| 
> single).*"                                                               
>                                                                          
>                                                                          
>                                                                          
>                                                                          
>                                         BackEnd 127.0.0.1,8888,2
> BackEnd 192.168.1.103,8888,3
> Session COOKIE GAST 1400
> EndGroup
> 
> It seems that AOL can't handle with Session-Cookies. I also tried to  
> handle the Session with the IP in the config-file.
> I found out that the reason for this problem is a proxie-farm that  
> aol uses to connect to websites.
> 
> Has anyone an idea or a workaround for this AOL-Problem?

I have no idea what it means that "AOL can't handle with
Session-Cookies". A cookie is a function of the browser, so the network
has nothing to do with it.

In any case I would suggest you go to the latest 2.x release, as there
are some problems with the 1.x session code.
-- 
Robert Segall
Apsis GmbH
Postfach, Uetikon am See, CH-8707
Tel: +41-44-920 4904

Hi All

Cliff, thanks for your reply. I missed it first time round - I was letting 
pound assume the default config file path ie /usr/local/etc/pound.cfg which 
it seemed to do. I'll be more careful in future.

The version of pound I was having trouble with was 2.0.4 on a slackware 
10.2 distro. I have switched to to pound version 2.0 and have been able to 
install OK.

Given that I have had trouble with version 2.0.4 and 2.0.9 does anyone have 
a suggestion as to what tools and/or procedures I might use to try to track 
down the bugs in the later versions. Hopefully I could then give some 
meaningful bug reports.

Thanks Mike

At 04:15 PM 7/07/2006 -0700, you wrote:
>On Wed, 2006-07-05 at 10:28 +1000, Mike Withers wrote:
>
> > and when I do:
> > root(at)black:~# /usr/local/sbin/pound -c
> > it reports:
> >
> > Segmentation fault
> >
> > Is it my config file at fault or something else?
>
>Usually you pass your config file as the argument to -c.  So it's not
>your config file but rather the lack of one ;-)
>
>/usr/local/sbin/pound -c /path/to/my/config.file
>
>Still, it's not correct for pound to segfault under any circumstances,
>so that's a bug.
>
>Regards,
>Cliff
>
>
>--
>To unsubscribe send an email with subject 'unsubscribe' to pound(at)apsis.ch.
>Please contact roseg(at)apsis.ch for questions.
>http://www.apsis.ch/pound/pound_list/archive/2006/2006-07/1151752508000/1152314123000

>

I'm sorry... I forgot to attach patch :/ So here it is

Kepi

Po, čec 31, 2006 ve 03:32:41 +0200, Ondrej Kudlik napsal:
> Hi,
> 
> I need to have fully customized error pages from pound but standard
> ErrXXX doesn't provide this functionality... It prepend and append
> some HTML code to choosen file.
> 
> So I made small patch to add this future to pound; I sent similar 
> patch to this list for 1.8 version, now there is some modification 
> for 2.0.9 version.
> 
> Setting of HTML pages is made in same manner as standard err, but
> there are new config options. HTMLErr414, HTMLErr500, HTMLErr501 and
> HTMLErr503 (man page is also modified to involve this future), i.e.
> 
> HTMLErr503 "/var/www/errors/503.html"
> 
> I will be very happy to see this functionality in some future
> versions and if it is possible, note that I'm really not good C 
> programmer and even if this is only small change in code I think it
> can be written better and cleaner. This is "just work" patch for me.
> 
> There is limitation in size of included HTML file, so it can be
> shorten if exceeds. But I'm not able to correct this with my
> current knowledge.
> 
> Have a nice day

On Mon, 2006-07-31 at 03:36 +0200, Ondra Kudlik wrote:
> I'm sorry... I forgot to attach patch :/ So here it is

I could not find any patch in this email too... :(

Regards
-- 
Johannes Findeisen
 http://hanez.org

attached files are only accessible through web archiv, but I
can't still see my message here...

So, for now... you can download it on
http://dev.igloonet.cz/os/pound/pound_2.0.9_htmlerr.patch

Kepi

Po, čec 31, 2006 ve 01:14:05 +0200, Johannes Findeisen napsal:
> On Mon, 2006-07-31 at 03:36 +0200, Ondra Kudlik wrote:
> > I'm sorry... I forgot to attach patch :/ So here it is
> 
> I could not find any patch in this email too... :(
> 
> Regards

Everyone,

I really do apologise for this - I have been absolutely slammed with work 
both at work and at home. I must say that I did not realise that the new 
version was waiting for me.

Please accept my apologies, I will send the information as soon as I can so 
that it can be included in the version after 2.1.

Regards
Richard



----- Original Message ----- 
From: "Robert Segall" <roseg(at)apsis.ch>
To: <pound(at)apsis.ch>
Sent: Monday, July 31, 2006 12:24 PM
Subject: [Pound Mailing List] Release 2.1


> We would very much like to release an "official" 2.1 by the end of the
> week. This is basically 2.0.9, with a fix for the wrong sizeof() problem
> and a replacement for inet_ntoa().
>
> We have delayed this due to a supposed problem in the new Location
> rewrite code (see the original report a month ago by Richard Hamilton).
> Since that time we have had no further information, neither confirming
> nor denying the problem.
>
> In the absence of additional information we shall release 2.1 on Friday.
> If you have come across this problem and can confirm it, or if you can
> confirm it works as it is supposed to, please let us know ASAP. We'll
> delay 2.1 only if we have positive proof of a problem that needs to be
> fixed.
>
> Many thanks for any info.
> -- 
> Robert Segall
> Apsis GmbH
> Postfach, Uetikon am See, CH-8707
> Tel: +41-44-920 4904
>
>
> -- 
> To unsubscribe send an email with subject 'unsubscribe' to pound(at)apsis.ch.
> Please contact roseg(at)apsis.ch for questions.
> http://www.apsis.ch/pound/pound_list/archive/2006/2006-07/1154345047000
> 

Po, čec 31, 2006 ve 01:24:07 +0200, Robert Segall napsal:
> We would very much like to release an "official" 2.1 by the end of the
> week. This is basically 2.0.9, with a fix for the wrong sizeof() problem
> and a replacement for inet_ntoa().
> 
> We have delayed this due to a supposed problem in the new Location
> rewrite code (see the original report a month ago by Richard Hamilton).
> Since that time we have had no further information, neither confirming
> nor denying the problem.
> 
> In the absence of additional information we shall release 2.1 on Friday.
> If you have come across this problem and can confirm it, or if you can
> confirm it works as it is supposed to, please let us know ASAP. We'll
> delay 2.1 only if we have positive proof of a problem that needs to be
> fixed.

I had simmilar problems as Richard Hamilton report but with 2.0.9
version and RewriteLocation set to 0 everything works fine!

Thanks for you great work

-- 
 .''`. Ondra 'Kepi' Kudlik
: :' : Debian GNU/Linux User
`. `'
  `-   http://www.nosoftwarepatents.com/cz/m/intro/index.html 

All reports loaded, btw

--------------------------
Sent from my BlackBerry Wireless Handheld


-----Original Message-----
From: Richard Hamilton <ricky.hamilton(at)btopenworld.com>
To: pound(at)apsis.ch <pound(at)apsis.ch>
Sent: Mon Jul 31 08:03:40 2006
Subject: Re: [Pound Mailing List] Release 2.1

Everyone,

I really do apologise for this - I have been absolutely slammed with work 
both at work and at home. I must say that I did not realise that the new 
version was waiting for me.

Please accept my apologies, I will send the information as soon as I can so 
that it can be included in the version after 2.1.

Regards
Richard



----- Original Message ----- 
From: "Robert Segall" <roseg(at)apsis.ch>
To: <pound(at)apsis.ch>
Sent: Monday, July 31, 2006 12:24 PM
Subject: [Pound Mailing List] Release 2.1


> We would very much like to release an "official" 2.1 by the end of the
> week. This is basically 2.0.9, with a fix for the wrong sizeof() problem
> and a replacement for inet_ntoa().
>
> We have delayed this due to a supposed problem in the new Location
> rewrite code (see the original report a month ago by Richard Hamilton).
> Since that time we have had no further information, neither confirming
> nor denying the problem.
>
> In the absence of additional information we shall release 2.1 on Friday.
> If you have come across this problem and can confirm it, or if you can
> confirm it works as it is supposed to, please let us know ASAP. We'll
> delay 2.1 only if we have positive proof of a problem that needs to be
> fixed.
>
> Many thanks for any info.
> -- 
> Robert Segall
> Apsis GmbH
> Postfach, Uetikon am See, CH-8707
> Tel: +41-44-920 4904
>
>
> -- 
> To unsubscribe send an email with subject 'unsubscribe' to pound(at)apsis.ch.
> Please contact roseg(at)apsis.ch for questions.
> http://www.apsis.ch/pound/pound_list/archive/2006/2006-07/1154345047000
> 


-- 
To unsubscribe send an email with subject 'unsubscribe' to pound(at)apsis.ch.
Please contact roseg(at)apsis.ch for questions.
http://www.apsis.ch/pound/pound_list/archive/2006/2006-07/1151752508000/1154
347420000

This email may contain privileged information, and is intended only for the
named
recipient and may be privileged. Distribution, disclosure or copying of this
email
by anyone other than the named recipient is prohibited. If you are not the
named
recipient, please notify us immediately, by return email, and permanently
destroy
this email and all copies of it. Internet email is not private, secure, or
reliable.
Trading instructions received from clients or staff by email will not be
accepted.
Neither Blackmont Capital Inc., its affiliates nor employees or contractors are
liable for any errors or omissions in the content or transmission of this
email.
Any opinions contained in this email are solely those of the author and, unless
clearly indicated otherwise in writing are not endorsed by Blackmont Capital
Inc.
or its affiliates.

Hi all,

Is this patch included in 2.1 release or/and can I use it with Pound 2.1 
release ?

Ondra Kudlik wrote:
> attached files are only accessible through web archiv, but I
> can't still see my message here...
>
> So, for now... you can download it on
> http://dev.igloonet.cz/os/pound/pound_2.0.9_htmlerr.patch
>
> Kepi
>
> Po, čec 31, 2006 ve 01:14:05 +0200, Johannes Findeisen napsal:
>   
>> On Mon, 2006-07-31 at 03:36 +0200, Ondra Kudlik wrote:
>>     
>>> I'm sorry... I forgot to attach patch :/ So here it is
>>>       
>> I could not find any patch in this email too... :(
>>
>> Regards
>>     
>
>   

-- 
Best Wishes,
Stefan Lambrev
ICQ# 24134177

Hi Robert,

Did you look into the issue of other non-thread-safe functions? As I 
pointed out in:

http://www.apsis.ch/pound/pound_list/archive/2006/2006-07/1151804232000#1151962139000

there are issues at least in the internal functions log_time and 
log_bytes which use a static buffer, and depending on the platform, 
there might be issues in localtime and gethostbyname.

Also, I haven't confirmed it with 2.0.9, but 2.0.4 still has at least 
one memory leak somewhere, our pound process grows pretty quickly to 
several hundred megabytes, though I have to say I have no idea 
where/why. Did something get fixed in that respect since then?

Thanks,

Jacques.

At 13:24 31/07/2006, Robert Segall wrote:
>We would very much like to release an "official" 2.1 by the end of the
>week. This is basically 2.0.9, with a fix for the wrong sizeof() problem
>and a replacement for inet_ntoa().
>
>We have delayed this due to a supposed problem in the new Location
>rewrite code (see the original report a month ago by Richard Hamilton).
>Since that time we have had no further information, neither confirming
>nor denying the problem.
>
>In the absence of additional information we shall release 2.1 on Friday.
>If you have come across this problem and can confirm it, or if you can
>confirm it works as it is supposed to, please let us know ASAP. We'll
>delay 2.1 only if we have positive proof of a problem that needs to be
>fixed.
>
>Many thanks for any info.
>--
>Robert Segall
>Apsis GmbH
>Postfach, Uetikon am See, CH-8707
>Tel: +41-44-920 4904
>
>
>--
>To unsubscribe send an email with subject 'unsubscribe' to pound(at)apsis.ch.
>Please contact roseg(at)apsis.ch for questions.
>http://www.apsis.ch/pound/pound_list/archive/2006/2006-07/1154345047000

Po, čec 31, 2006 ve 02:20:04 +0200, Ondra Kudlik napsal:
> Po, čec 31, 2006 ve 01:24:07 +0200, Robert Segall napsal:
> > We would very much like to release an "official" 2.1 by the end of the
> > week. This is basically 2.0.9, with a fix for the wrong sizeof() problem
> > and a replacement for inet_ntoa().
> > 
> > We have delayed this due to a supposed problem in the new Location
> > rewrite code (see the original report a month ago by Richard Hamilton).
> > Since that time we have had no further information, neither confirming
> > nor denying the problem.
> > 
> > In the absence of additional information we shall release 2.1 on Friday.
> > If you have come across this problem and can confirm it, or if you can
> > confirm it works as it is supposed to, please let us know ASAP. We'll
> > delay 2.1 only if we have positive proof of a problem that needs to be
> > fixed.
> 
> I had simmilar problems as Richard Hamilton report but with 2.0.9
> version and RewriteLocation set to 0 everything works fine!

I found one important problem. When I have RewriteLocation set to 0
i cannot go to directories withou ending slash.

So if i have directory data and I go to
http://www.something.com/data it timeouts. I know that server
redirect this to http://www.something.com/data/ 

I'm not sure on global impact but it is very serious bug for us :/

Thanks for any idea

-- 
 .''`. Ondra 'Kepi' Kudlik
: :' : Debian GNU/Linux User
`. `'
  `-   http://www.nosoftwarepatents.com/cz/m/intro/index.html 

On Mon, 2006-07-31 at 16:50 +0200, Jacques Caron wrote:
> Hi Robert,
> 
> Did you look into the issue of other non-thread-safe functions? As I 
> pointed out in:
> 
>
http://www.apsis.ch/pound/pound_list/archive/2006/2006-07/1151804232000#1151962139000
> 
> there are issues at least in the internal functions log_time and 
> log_bytes which use a static buffer, and depending on the platform, 
> there might be issues in localtime and gethostbyname.

Yes, these are fixed as well.

> Also, I haven't confirmed it with 2.0.9, but 2.0.4 still has at least 
> one memory leak somewhere, our pound process grows pretty quickly to 
> several hundred megabytes, though I have to say I have no idea 
> where/why. Did something get fixed in that respect since then?

We have fixed a potential memory leak in the authorisation decoding. I
don't really know if that covers it all.
-- 
Robert Segall
Apsis GmbH
Postfach, Uetikon am See, CH-8707
Tel: +41-44-920 4904

On Mon, 2006-07-31 at 17:41 +0200, Ondra Kudlik wrote:
> I found one important problem. When I have RewriteLocation set to 0
> i cannot go to directories withou ending slash.
> 
> So if i have directory data and I go to
> http://www.something.com/data it timeouts. I know that server
> redirect this to http://www.something.com/data/ 
> 
> I'm not sure on global impact but it is very serious bug for us :/
> 
> Thanks for any idea

Most web servers reply with a redirect when the client requests a
directory. The sequence is:

- client requests /data
- server replies with a redirect to /data/
- client requests /data/
- server replies with the default for the directory (index.html,
listing, etc)

That means that without RewriteRedirect you won't be able to get it
working the way you expect.

What exactly happens with RewriteRedirect? Where is the client
redirected to? Is rewriting working (never/sometimes/always) as
expected?
-- 
Robert Segall
Apsis GmbH
Postfach, Uetikon am See, CH-8707
Tel: +41-44-920 4904

Po, čec 31, 2006 ve 06:26:20 +0200, Robert Segall napsal:
> On Mon, 2006-07-31 at 17:41 +0200, Ondra Kudlik wrote:
> > I found one important problem. When I have RewriteLocation set to 0
> > i cannot go to directories withou ending slash.
> > 
> > So if i have directory data and I go to
> > http://www.something.com/data it timeouts. I know that server
> > redirect this to http://www.something.com/data/ 
> > 
> > I'm not sure on global impact but it is very serious bug for us :/
> > 
> > Thanks for any idea
> 
> Most web servers reply with a redirect when the client requests a
> directory. The sequence is:
> 
> - client requests /data
> - server replies with a redirect to /data/
> - client requests /data/
> - server replies with the default for the directory (index.html,
> listing, etc)

yes, I know this, but...
> 
> That means that without RewriteRedirect you won't be able to get it
> working the way you expect.
> 
> What exactly happens with RewriteRedirect? Where is the client
You meen RewriteLocation or RewriteRedirect is still in Pound 2.X ?
> redirected to? Is rewriting working (never/sometimes/always) as
> expected?

So, to be clear, I have this config:

-----------------------
ListenHTTP
	Address	X.X.X.X
	Port	80

	RewriteLocation 1 # default
	
	Client 20
End

Service
	HeadRequire	"Host: .*somehost.cz.*"
	BackEnd
		Address	127.0.0.1
		Port	8000
	End
End
----------------------

1/ With this config, I can go to www.somehost.cz/data and it
correctly redirect to www.somehost.cz/data/

but I have also set RewriteRules to redirect hosts without www to
www variant and when i go to somehost.cz it gives me redirect loop.

2/ When I change RewriteLocation to 0 redirect from nonwww variant
to www variant work great, so i can go to somehost.cz and it
redirects me to www.somehost.cz

but when i go to www.somehost.cz/data it is loading and loading and
after long time it timeouts.

In pound 1.x a was happy with parametr RewriteRedirect 2 but if I'm
right it was replaced by Change3x (or something like) and then with
RewriteLocation

If you need more details or I'm doing something in bad way, let me
know.

Have a nice day

-- 
 .''`. Ondra 'Kepi' Kudlik
: :' : Debian GNU/Linux User
`. `'
  `-   http://www.nosoftwarepatents.com/cz/m/intro/index.html