Great thanks for your quick answer. 

If I understand well setting LimitRequestLine 10240 in name based virtualhost ? 
So what is your recommendation. 

Anyway, thanks for the time already spent on this. 

Regards 



Florent THOMAS 

☎ +33 972 457 755 
✉ [ mailto:florent.tho...@mind-and-go.com | florent.tho...@mind-and-go.com ] 


Mind & Go 
14, Rue Pierre Cartelet | 66000 PERPIGNAN [ https://mind-and-go.com/ |   ] [ 
https://www.facebook.com/Mind-And-Go-296462057466508/ ] [ 
https://twitter.com/MindAndGo |   ] [ 
https://www.linkedin.com/company/mind-and-go |   ] 


De: "Eric Covener" <cove...@gmail.com> 
À: "users" <users@httpd.apache.org> 
Envoyé: Dimanche 21 Janvier 2024 14:48:03 
Objet: Re: [users@httpd] [httpd2.4.57]Possible conflicts between VHosts with 
SSL and LimitRequestLine 

The hostname, normally in the Host header, is not read until after the request 
line. So it cannot be effectively set in name based vhosts. 

The manual already warns about it 

On Sun, Jan 21, 2024, 9:26 AM Florent Thomas 
<florent.tho...@mind-and-go.com.invalid> wrote: 



Hi everyone, 

I'm running : 
Server version: Apache/2.4.57 (Debian) 
Server built: 2023-04-13T03:26:51 
Server's Module Magic Number: 20120211:127 
Server loaded: APR 1.7.2, APR-UTIL 1.6.3, PCRE 10.42 2022-12-11 
Compiled using: APR 1.7.2, APR-UTIL 1.6.3, PCRE 10.42 2022-12-11 
Architecture: 64-bit 
Server MPM: event 
threaded: yes (fixed thread count) 
forked: yes (variable process count) 
Server compiled with.... 
-D APR_HAS_SENDFILE 
-D APR_HAS_MMAP 
-D APR_HAVE_IPV6 (IPv4-mapped addresses enabled) 
-D APR_USE_PROC_PTHREAD_SERIALIZE 
-D APR_USE_PTHREAD_SERIALIZE 
-D SINGLE_LISTEN_UNSERIALIZED_ACCEPT 
-D APR_HAS_OTHER_CHILD 
-D AP_HAVE_RELIABLE_PIPED_LOGS 
-D DYNAMIC_MODULE_LIMIT=256 
-D HTTPD_ROOT="/etc/apache2" 
-D SUEXEC_BIN="/usr/lib/apache2/suexec" 
-D DEFAULT_PIDLOG="/var/run/apache2.pid" 
-D DEFAULT_SCOREBOARD="logs/apache_runtime_status" 
-D DEFAULT_ERRORLOG="logs/error_log" 
-D AP_TYPES_CONFIG_FILE="mime.types" 
-D SERVER_CONFIG_FILE="apache2.conf" 

I've been struggling for almost 2 weeks on this weird issue. 
I have multiple vhosts working perfectly either in http and httpS with let's 
encrypt. 

Yet for one vhost we need to set limitrequestline to a higher value. I'm 
familiar with the procedure and I changed the value in the vhost. Sadly it 
didn't worked as expected. 
After searching in any place, enabling/disabling securities, modules etc... I 
found another conf conflicting with mine. 

The ssl.conf vhost placed in conf-enabled is causing troubles. However its 
content is not really complicated : 

LoadModule ssl_module /usr/lib/apache2/modules/mod_ssl.so 
Listen 443 
<VirtualHost *:443> 
ServerName publicdomainname.tld 
SSLEngine on 
SSLCertificateFile "/etc/ssl/certs/apache-httpd.crt" 
SSLCertificateKeyFile "/etc/ssl/private/apache-httpd.key" 
</VirtualHost> 

My other vhost is as follows : 

<VirtualHost *:443> 
ServerName otherdomain.tld 
.../... 
<IfModule mod_ssl.c> 
RequestHeader set X-Forwarded-Proto "https" 
</IfModule> 
.../.... 
LimitRequestLine 10240 
.../... 
Include /etc/letsencrypt/options-ssl-apache.conf 
</VirtualHost> 

When both are enabled, the LimitRequestLine is not considered when reaching 
htps://otherdomain.tld . When I comment the Vhost in ssl.conf everything works 
fine. 

Is there any "permeability" between vhosts ? is there any precedence that could 
cause this vhost to be considered as the "master" of some options? Could anyone 
lead me to wha I am doing wrong? 

Sorry if I'm not clear, any question and/or feedback would be appreciated. 

Regards, 

Florent THOMAS 




Reply via email to