Bug#416611: mailing list discussion

2007-03-30 Thread Carl Johnstone
On the mod_perl list it has been considered a DOS attack, and not an exploit. It's also only in Apache::PerlRun - so doesn't affect users using the more popular Apache::Registry (was fixed mid-2000). Nor does it affect users using pure-handlers. I'd also point out that the release also fixes

Bug#272686: include mod_deflate with apache 1.3

2004-09-21 Thread Carl Johnstone
Package: apache Version: 1.3.31-5 Severity: wishlist Could this be included with the debian apache package? http://sysoev.ru/en/ It's an alternative to mod_gzip, which seems to be much more mature - it also links with zlib rather than including it's own version. Carl

RE: Bug#242543: More information

2004-09-03 Thread Carl Johnstone
> Can you give us the patch directly since you have it working? I'm actually a little concerned that it might break something else! I've been working with a clean (non Debian) apache source, so it's really an upstream problem. What is their policy regarding regular bugs in 1.3? Are they s

Bug#242543: More information

2004-09-02 Thread Carl Johnstone
I've managed to track down why QUERY_STRING works and PATH_INFO doesn't. In src/modules/standard/mod_include.c around line 2181 mod_include fixes up the QUERY_STRING. If I add similar code to fixup PATH_INFO ie: if (r->path_info) { ap_table_setn(r->subprocess_env, "PATH_INFO", r->path_i

RE: Apache2 VirtualHosts

2004-06-02 Thread Carl Johnstone
Hi, The "debian way" is the standard Apache way - check out the documentation on the Apache web site - specifically here's the URL for the virtual host docs. http://httpd.apache.org/docs-2.0/vhosts/ Carl PS This is the apache-development mailing list for discussing apache development. Y

RE: Bug#243918: apache adds webmaster alias but forget newaliases

2004-04-15 Thread Carl Johnstone
> I wonder if we should reassign this bug to all the MTA since > some of the > ship newaliases in /usr/bin and others in /usr/sbin (that > apache uses). You don't need to run newaliases with exim - which is Debian's recommended MTA. How do other packages handle adding email aliases? Might

Bug#242543: apache: PATH_INFO not set for

2004-04-07 Thread Carl Johnstone
> No, they shouldn't! PATH_INFO '... holds the additional path > information > that remains after the URI has been translated into a file path.' [1] > So, if your URI is 'http://site/test2.html' > > 1) URI to file translation: > > http://site/test2.html => {DOCUMENT ROOT}/test2.html >

Bug#242543: apache: PATH_INFO not set for

2004-04-07 Thread Carl Johnstone
Package: apache Version: 1.3.29.0.2-4 Severity: normal test.html --- test2.html - - Requesting http://site/test.html/path?query gives me: path query Requesting http://site/test2.html gives me: (none) query They should ma

RE: Virtual Hosts

2004-02-26 Thread Carl Johnstone
Hi Joe, That's the correct documented behaviour for Apache - once you define VirtualHosts, the first VirtualHost becomes the default site. Just add a blank VirtualHost before the others you've designed, by default it will inherit all it's settings from your "main" Apache config. Carl PS

RE: Bug#233538: apache: dotfiles not skipped, scanning /etc/apache/conf.d

2004-02-18 Thread Carl Johnstone
> While you're at it ;-) please also skip files matching '*.dpkg-*'. That a good point - dpkg/apt etc will be dropping conffiles into /etc/apache/conf.d and -dpkg- files may be left behind due to this. Carl GMG Regional Digital is

Bug#231134: Wish: Dummy LoadModule directive in apache2.conf to ease module installations through apxs (e.g. PHP5)

2004-02-05 Thread Carl Johnstone
Similar to my apache 1.3 bug - wouldn't it be better to fix apxs to generate the .load file needed to make it work the new "Debian" way? Carl GMG Regional Digital is part of the Guardian Media Group plc. CONFIDENTIALITY NOTICE. The

RE: [SPAM] - apache 1.3.29.0.1-4 and -5 - Domain does not exist

2004-01-27 Thread Carl Johnstone
> If -4 will go in without any problem, -5 will contain a major > change from > the actual old config layout to the apache2 config layout, > that will make > easier for everyone to transit from one package to another > and vice versa > without too much headache. is -5 for sarge+1 then? Carl-

Bug#229117: apache2-common: multiple instances of apache

2004-01-23 Thread Carl Johnstone
> The security benefits of the experimental perchild module can be > approximated by running multiple instances of apache2, you can't run two instances of apache on the same server using the same port as it's always been possible to run multiple copies of apache as different users, being able t

Bug#227997: mod_usertrack causes segfault

2004-01-19 Thread Carl Johnstone
severity 227997 important tags 227997 upstream tags 227997 pending stop quit Fabio: > it seems to me that this option can be used only when mod_usertrack is > statically compiled into apache (and in our case it is not > since we make use of DSO). "Compiled in" is poor choice of wording in the d

Bug#228077: apache-dev: apxs doesn't work well with modules-config

2004-01-16 Thread Carl Johnstone
Package: apache-dev Version: 1.3.29.0.1-3 Severity: minor when installing 3rd party modules using apxs: * no .info file is created in /usr/lib/apache breaking module-config * a configuration line is added directly to the apache config, rather than using modules-config to enable the module

Bug#227997: mod_usertrack causes segfault

2004-01-16 Thread Carl Johnstone
> > Two separate apache installs both exhibit the same problem - > one of which is a fairly clean re-install I did recently for > testing purposes. I assumed it would be easy to reproduce :-) > Just re-installed apache from scrach again, and apache exhibits the same problem - wierd! Carl

Bug#227997: mod_usertrack causes segfault

2004-01-15 Thread Carl Johnstone
Package: apache-common Version: 1.3.29.0.1-3 Severity: grave Enabling mod_usertrack causes apache to Segfault. Carl

RE: sarge package transition

2004-01-05 Thread Carl Johnstone
> Recently sarge has received a new version of perl but apache > and mod-perl > packages in sarge are old. Until the sid packages will not enter sarge > there is no way for us to "fix" this problem since it is already fixed > since a while in sid. We have no control over the package > flow from