apache won't restart after adding VirtualHost conf file, why not?

Question

In a CentOS 7 server, I get the following error when I type sudo apachectl restart after I add an include file at the bottom of httpd.conf:

Job for httpd.service failed. See 'systemctl status httpd.service' and 'journalctl -xn' for details.

When I then type sudo systemctl status httpd.service -l, the result is:

httpd.service - The Apache HTTP Server
   Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled)
   Active: failed (Result: exit-code) since Tue 2014-12-23 20:10:37 EST; 2min 15s ago
  Process: 2101 ExecStop=/bin/kill -WINCH ${MAINPID} (code=exited, status=0/SUCCESS)
  Process: 2099 ExecStart=/usr/sbin/httpd $OPTIONS -DFOREGROUND (code=exited, status=1/FAILURE)
 Main PID: 2099 (code=exited, status=1/FAILURE)
   Status: "Total requests: 0; Current requests/sec: 0; Current traffic:   0 B/sec"

Dec 23 20:10:37 ip-address httpd[2099]: AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using fe80::e23f:49ff:feb7:2a21. Set the 'ServerName' directive globally to suppress this message
Dec 23 20:10:37 ip-address systemd[1]: httpd.service: main process exited, code=exited, status=1/FAILURE
Dec 23 20:10:37 ip-address systemd[1]: Failed to start The Apache HTTP Server.
Dec 23 20:10:37 ip-address systemd[1]: Unit httpd.service entered failed state.

I can get apache to restart if I comment out the include directive, and I can recreate the error again by un-commenting the include directive. How can I get apache to start properly using the contents of the include file?

The line at the bottom of httpd.conf that triggers the error is: IncludeOptional sites-enabled/*.conf. The only .conf file in the sites-enabled folder is mydomain.com.conf, which has the following contents:

<VirtualHost *:80>
    ServerName www.mydomain.com
    ServerAlias mydomain.com
    DocumentRoot /var/www/mydomain.com/public_html
    ErrorLog /var/www/mydomain.com/error.log
    CustomLog /var/www/mydomain.com/requests.log combined
</VirtualHost>

The httpd.conf is the same as what comes pre-installed with httpd, except for the one line include directive above. I know because I did sudo yum remove httpd mod_ssl and sudo yum install httpd mod_ssl right before triggering this error. The entire httpd.conf can be read at a file sharing site by clicking on this link.

I encountered this problem when explicitly following the steps in this tutorial.

When I comment out the include file, http/mydomain.com successfully serves up the static content located in /var/www/html, which is the DocumentRoot defined in httpd.conf. The problem seems to be coming from the VirtualHost directive in the include file shown above. To aid in diagnosis, I have included in EDIT#3 below links to all of the .conf files that are contained in the three include directives in httpd.conf.

EDIT #1

When I try m32's advice to change the /etc/hostname to define mydomain.com, apache still will not restart, and the systemctl status httpd.service results in the following:

[sudo_user_account@server-ip-address ~]$ sudo systemctl status httpd.service -l
httpd.service - The Apache HTTP Server
   Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled)
   Active: failed (Result: exit-code) since Tue 2014-12-23 14:25:35 EST; 20s ago
  Process: 31993 ExecStop=/bin/kill -WINCH ${MAINPID} (code=exited, status=0/SUCCESS)
  Process: 31991 ExecStart=/usr/sbin/httpd $OPTIONS -DFOREGROUND (code=exited, status=1/FAILURE)
 Main PID: 31991 (code=exited, status=1/FAILURE)
   Status: "Total requests: 1; Current requests/sec: 0; Current traffic:   0 B/sec"

Dec 23 14:25:35 hostname systemd[1]: httpd.service: main process exited, code=exited, status=1/FAILURE
Dec 23 14:25:35 hostname systemd[1]: Failed to start The Apache HTTP Server.
Dec 23 14:25:35 hostname systemd[1]: Unit httpd.service entered failed state.

EDIT #2

I also tried eyoung100's advice to change the contents of my /etc/hosts file, as defined in the following image, but I still get the same error defined in EDIT#1 above.

EDIT#3

Per DerekC's request, I ran sudo apachectl configtest and got:

AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using fe80::e23f:49ff:feb7:2a21. Set the 'ServerName' directive globally to suppress this message
Syntax OK

In addition, per GarethTheRed's suggestion, I examined the include directives in httpd.conf. There are three include directives in httpd.conf. I have listed the three below, along with all the files located in each directive's folder. These are all the standard .conf files installed with httpd. I have not modified any of them yet. You can view each of the .conf files at a file sharing site by clicking on the links below:

Include conf.modules.d/*.conf references the following files in the conf.modules.d directory: 00-base.conf
00-dav.conf
00-lua.conf
00-mpm.conf
00-proxy.conf
00-ssl.conf
00-systemd.conf
01-cgi.conf

IncludeOptional conf.d/*.conf references the following files in the conf.d directory: autoindex.conf
ssl.conf
userdir.conf
welcome.conf
There is also a README file that I am omitting here.

In addition, the IncludeOptional sites-enabled/*.conf directive and it's contents were outlined thoroughly in the OP above.

Are any of these include files conflicting with the VirtualHost settings in mydomain.com.conf?

EDIT#4

Per garethTheRed's suggestion, I moved mydomain.com.conf to the conf.d directory and then started commenting out lines in mydomain.com.conf one by one until httpd was able to restart. I then started un-commenting lines to see how many lines could remain and have httpd still restart. I was able to get httpd to restart, but systemctl status httpd.service -l continues to produce the same warning:

AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using fe80::e23f:49ff:feb7:2a21. Set the 'ServerName' directive globally to suppress this message

The VirtualHost syntax that allows httpd to start (though continuing to generate the above warning) is as follows:

<VirtualHost *:80>
    ServerName www.mydomain.com
    ServerAlias mydomain.com
    DocumentRoot /var/www/mydomain.com/public_html
</VirtualHost>

Note that I had to omit the following lines, whose presence escalates the warning into a complete inability to start http:

# ErrorLog /var/www/mydomain.com/error.log
# CustomLog /var/www/mydomain.com/requests.log combined

Also, I ran sudo journalctl -xelu httpd and the terminal replied by repeating the following many times:

-- 
-- Unit httpd.service has finished shutting down.
Dec 24 17:48:43 server-ip-address systemd[1]: Stopped The Apache HTTP Server.
-- Subject: Unit httpd.service has finished shutting down
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit httpd.service has finished shutting down.
Dec 24 17:48:48 server-ip-address systemd[1]: Starting The Apache HTTP Server...
-- Subject: Unit httpd.service has begun with start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit httpd.service has begun starting up.
Dec 24 17:48:48 server-ip-address httpd[10364]: AH00558: httpd: Could not reliably d
Dec 24 17:48:48 server-ip-address systemd[1]: Started The Apache HTTP Server.
-- Subject: Unit httpd.service has finished start-up
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit httpd.service has finished starting up.
-- 
-- The start-up result is done.
lines 887-909/909 (END)

Note: The above results remain the same regardless of whether I use eyoung100's hosts file or m32's host file. For this question to be answered, I think I should be able to create log files and also avoid the servername warning. Otherwise, I fear that subsequent steps of configuring httpd will be prone to lingering errors.

What exact "nebulous" error is produced when you add a `ServerName` directive? Add it to your question. — wurtel, Dec 23 '14 at 08:14
Is there an `IncludeOptional` at the end of original `httpd.conf` file? I believe there is on a default install. The issue may be that you're adding the extra `IncludeOptional` at the end which forces `apache` to read the included config file twice and consequently causes to it fail. — garethTheRed, Dec 24 '14 at 08:09
For a start, you need to move your `mydomain.com.conf` to `/etc/httpd/conf.d` (where all the others are) and delete `/etc/httpd/sites-enabled` (that's a Debian/Ubuntu thing) and the `IncludeOptional` you added to `httpd.conf`. You can ignore the `AH00558` message for now as it doesn't stop `apache`. Next remove a line from your `conf` file until you find the culprit (or alternatively, start with an empty one and restore one line at a time). — garethTheRed, Dec 24 '14 at 21:53
Also, have a look at, and add the output of `journalctl -xelu httpd` to your question. — garethTheRed, Dec 24 '14 at 22:06
My bad. That `journalctl` shouls be preceded by `sudo`.. The directory for your log files should be writeable by the user that `apache` runs as. — garethTheRed, Dec 24 '14 at 22:44
@garethTheRed and anyone from google, you do not need to remove IncludeOptional sites-enabled/*.conf. That was a red herring. The real issue was the log files as pointed out below by garethTheRed. I followed the same tutorial as the OP and had the same issue. Removing the log file lines from the virtual host conf file fixed it. — Louise Eggleton, May 16 '16 at 11:29
@garethTheRed Your contributions to this site are greatly appreciated. Even when we are not posting questions that might put us individually on your radar screen. We still use and appreciate the things that you teach all of us. Thank you. — CodeMed, May 16 '16 at 16:45

garethTheRed · Accepted Answer · 2014-12-26T20:17:46.387

The logs are causing the errors because apache cannot write to the root of your website. Even if you were to fix the file permissions, you'd still be blocked by SELinux; which only allows apache to write logs to /var/log/httpd. The easiest solution would be to change your website to log to this directory - maybe with a filename that contains the website name in order to differentiate it from other logs.

ErrorLog /var/log/httpd/mydomain_com_error.log
CustomLog /var/log/httpd/mydomain_com_requests.log combined

To set the hostname of the server and get rid of the AH00558 warning, simply use:

hostnamectl set-hostname --static <FQDN of your machine>

e.g.

hostnamectl set-hostname --static mydomain.com

I encountered another error as I continue to work on this. Are you willing to help me with it also? Here is the link: http://unix.stackexchange.com/questions/176052/httpd-doesnt-serve-relative-url-from-tomcat-why-not?noredirect=1 — CodeMed, Dec 26 '14 at 21:45

m32 · Answer 2 · 2014-12-24T03:09:47.983

2

I got this error when I forgot to set the hostname for the machine. Have you done so?
Run this command:

echo 'example.com' >> /etc/hostname

Verify with hostname

Edit:
With the updated OP, it looks like your hosts file isn't setup right. Here's what mine looks like;

127.0.0.1 localhost.localdomain localhost
# Auto-generated hostname. Please do not remove this comment.
166.66.666.66 m32.me m32
::1     ip6-localhost ip6-loopback

Try setting it to something simple like that. Of course, replace all the information with your own.
166.66.666.66 with your own IP
m32.me and m32 with your domain, and domain without the TLD

edited Dec 24 '14 at 03:09

answered Dec 23 '14 at 08:35

m32

456
4
4

That did not fix the problem, I still get the `AH00558` error when I use your example hosts file, with my `ip` and `mydomain.com mydomain`. – CodeMed Dec 24 '14 at 23:58

apache won't restart after adding VirtualHost conf file, why not?

2 Answers2

Linked