Apache errors

Which sections of httpd.conf should I look at to sort this out?


Code:
[Mon Apr 13 16:09:11.543425 2026] [ssl:warn] [pid 47840:tid 32754373140496] AH01873: Init: Session Cache is not configured [hint: SSLSessionCache]
[Mon Apr 13 16:09:11.548285 2026] [mpm_event:notice] [pid 47840:tid 32754373140496] AH00489: Apache/2.4.66 (FreeBSD) OpenSSL/3.5.4 configured -- resuming normal operations
[Mon Apr 13 16:09:11.548447 2026] [core:notice] [pid 47840:tid 32754373140496] AH00094: Command line: '/usr/local/sbin/httpd -D NOHTTPACCEPT'

I have made minimal changes to the sample config.

Code:
66,67c66,67
< LoadModule mpm_event_module libexec/apache24/mod_mpm_event.so
< #LoadModule mpm_prefork_module libexec/apache24/mod_mpm_prefork.so
---
> #LoadModule mpm_event_module libexec/apache24/mod_mpm_event.so
> LoadModule mpm_prefork_module libexec/apache24/mod_mpm_prefork.so
92c92
< LoadModule socache_shmcb_module libexec/apache24/mod_socache_shmcb.so
---
> #LoadModule socache_shmcb_module libexec/apache24/mod_socache_shmcb.so
129c129
< LoadModule proxy_module libexec/apache24/mod_proxy.so
---
> #LoadModule proxy_module libexec/apache24/mod_proxy.so
133c133
< LoadModule proxy_fcgi_module libexec/apache24/mod_proxy_fcgi.so
---
> #LoadModule proxy_fcgi_module libexec/apache24/mod_proxy_fcgi.so
148c148
< LoadModule ssl_module libexec/apache24/mod_ssl.so
---
> #LoadModule ssl_module libexec/apache24/mod_ssl.so
250,251c250,251
< DocumentRoot "/usr/local/www/nextcloud"
< <Directory "/usr/local/www/nextcloud">
---
> DocumentRoot "/usr/local/www/apache24/data"
> <Directory "/usr/local/www/apache24/data">

Maybe I need to add something.
 
According to AI:

The real problem is still that Apache isn’t listening, not this SSL warning.

Also, the things that need to be checked:

sockstat -4 -l | grep httpd
ps aux | grep httpd
cat /etc/rc.conf | grep apache

ps shows four instances of httpd but with an argument of -D NOHTTPACCEPT

Where does that come from?
 
but with an argument of -D NOHTTPACCEPT
It just means it's not using accf_http(9).

 
# kldload accf_http:

operation not permitted.

It's running this in a jail.

I only get this error when installing nextcloud. Apache works fine otherwise.
 
# kldload accf_http:

operation not permitted.

It's running this in a jail.

I only get this error when installing nextcloud. Apache works fine otherwise.
FreeBSD has a kernel module (called accf_http) that accepts the http/https requests and doesn't hand it off to apache until after enough of the initial data handshake happens and is clean. Either 1) let the jail run install the module (unlikely since that is done by your hosting provider), 2) turn that off. Try running without the -DNOHTTPACCEPT manually and see if that works and then figure out how to pull that out via rc.conf

It used to be called apache22_http_accept_enable="NO" in /etc/rc.conf but that maybe apache24 or just apache these days.
 
It's not an error and you don't actually need it for Apache to function properly.

None of the messages balanga posted are errors that need fixing. The first one is a warning, the second and third are informational. And none of them are the reason for Apache not listening.
 
Use sockstat in the environment where you think Apache is running - see if anything is listening (and on what IPs) on port 80 (and/or 443 if you are using ssl).

If in a jail, is all the networking set up correctly, so that you can connect to the jail from other machines on the network (e.g. can you ssh in from another machine?)

Do you have a firewall running in the jail?
 
For some reason I get this error on a particular laptop.

I do not have this problem accessing the same web server in a jail from three other laptops so have no idea what is going on.

It obviously is not a problem in the apache configuration, but it would be nice to know why this error shows up. Maybe my browser is pulling something out from cache....
 
Better test, nc -zv 192.168.1.20 80. Does that get you a connection refused, timed out or successful? Refused would mean there's nothing listening on that port, timed out might indicate some networking or firewall issues, and connection successful, well, that's what you want to see.
 
I can also connect with lynx so there must be something be wrong with chrome.
Try a different browser? Or some browser plugin in chrome that's interfering?

In any case, if you get a successful connection but the browser reports it took too long to respond, you might want to have a closer look at the PHP page or whatever scripting engine you enabled. It sounds like Apache correctly accepts the connection (the netcat test confirmed you can at least make a TCP connection) and hands it off to the scripting engine, but it fails to respond (therefor the page itself takes too long load) or the page itself is stuck. Maybe a javascript in an infinite loop? Lynx doesn't support javascript, your chrome browser does.
 
Back
Top