Apache not starting on CentOS 7

JaredBusch

The server is used for ownCloud. It was working. Nothing was changed related to httpd setup. yum update was ran and the server was rebooted.

[root@oc httpd]# ls -l
total 12
drwxr-xr-x. 2 root root   35 Dec 21 14:26 conf
drwxr-xr-x. 2 root root 4096 Jan  3 01:14 conf.d
drwxr-xr-x. 2 root root 4096 Jan  3 01:14 conf.modules.d
lrwxrwxrwx. 1 root root   19 Dec 21 14:26 logs -> ../../var/log/httpd
lrwxrwxrwx. 1 root root   29 Dec 21 14:26 modules -> ../../usr/lib64/httpd/modules
lrwxrwxrwx. 1 root root   10 Dec 21 14:26 run -> /run/httpd
drwxr-xr-x. 2 root root 4096 Jan  3 01:20 ssl
[root@oc httpd]# cd  ssl
[root@oc ssl]# ls -l
total 16
-rw-r--r--. 1 root root 2102 Jan  3 01:06 oc.acidist.com.crt
-rw-r--r--. 1 root root 3458 Jan  2 12:22 oc.acidist.com.encrypted.key
-rw-r--r--. 1 root root 3294 Jan  2 12:27 oc.acidist.com.key
-rw-r--r--. 1 root root 2106 Jan  3 01:07 root_bundle.crt
[root@oc ssl]# systemctl start httpd
Job for httpd.service failed because the control process exited with error code. See "systemctl status httpd.service" and "journalctl -xe" for details.
[root@oc ssl]# systemctl status httpd -l
● httpd.service - The Apache HTTP Server
   Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled; vendor preset: disabled)
   Active: failed (Result: exit-code) since Sun 2016-01-17 17:52:51 CST; 5s ago
     Docs: man:httpd(8)
           man:apachectl(8)
  Process: 3232 ExecStop=/bin/kill -WINCH ${MAINPID} (code=exited, status=1/FAILURE)
  Process: 3231 ExecStart=/usr/sbin/httpd $OPTIONS -DFOREGROUND (code=exited, status=1/FAILURE)
 Main PID: 3231 (code=exited, status=1/FAILURE)

Jan 17 17:52:51 oc.acidist.com systemd[1]: Starting The Apache HTTP Server...
Jan 17 17:52:51 oc.acidist.com httpd[3231]: AH00526: Syntax error on line 101 of /etc/httpd/conf.d/ssl.conf:
Jan 17 17:52:51 oc.acidist.com httpd[3231]: SSLCertificateFile: file '/etc/httpd/ssl/oc.acidist.com.crt' does not exist or is empty
Jan 17 17:52:51 oc.acidist.com systemd[1]: httpd.service: main process exited, code=exited, status=1/FAILURE
Jan 17 17:52:51 oc.acidist.com kill[3232]: kill: cannot find process ""
Jan 17 17:52:51 oc.acidist.com systemd[1]: httpd.service: control process exited, code=exited status=1
Jan 17 17:52:51 oc.acidist.com systemd[1]: Failed to start The Apache HTTP Server.
Jan 17 17:52:51 oc.acidist.com systemd[1]: Unit httpd.service entered failed state.
Jan 17 17:52:51 oc.acidist.com systemd[1]: httpd.service failed.

scottalanmiller

Looks like it thinks it has a syntax error in that config file.

scottalanmiller

What is the status of this file: /etc/httpd/ssl/oc.acidist.com.crt

JaredBusch

@scottalanmiller said:

What is the status of this file: /etc/httpd/ssl/oc.acidist.com.crt

Not sure what you mean.. ls -l results in this

-rw-r--r--. 1 root root 2102 Jan  3 01:06 oc.acidist.com.crt

as note in the text above.

scottalanmiller

If you look in the file, it looks normal?

JaredBusch

@scottalanmiller said:

If you look in the file, it looks normal?

Yes. looks like a normal cert..

[root@oc ssl]# cat oc.acidist.com.crt
-----BEGIN CERTIFICATE-----
MIIF4jCCBMqgAwIBAgIQIyVm8Bdz9GRKswiZTiYtUzANBgkqhkiG9w0BAQsFADB4

#snipped a bunch out of the middle..

Y3PBtYhJyVJ3Gh8sClT+XGS7WK8mSg==
-----END CERTIFICATE-----
[root@oc ssl]#

scottalanmiller

Yeah, doesn't look corrupt, hmmm...

JaredBusch

@scottalanmiller said:

Yeah, doesn't look corrupt, hmmm...

I grabbed the entire thing and dumped it into notepad++ and saved it as test.crt on my desktop.. it is valid.

JaredBusch

I realized, OpenSSL updated in that last yum update. wonder if it broke something.

This is a pretty basic install.

scottalanmiller

Seems likely under the circumstances.

JaredBusch

@scottalanmiller said:

Seems likely under the circumstances.

I just turned off SSL and it httpd started.

but all clients for ownCloud use https.. /sigh...

dafyre

@JaredBusch said:

@scottalanmiller said:

Seems likely under the circumstances.

I just turned off SSL and it httpd started.

but all clients for ownCloud use https.. /sigh...

I'm using Letsencrypt for my ownCloud [personal] set up. It works great!

JaredBusch

@dafyre said:

@JaredBusch said:

@scottalanmiller said:

Seems likely under the circumstances.

I just turned off SSL and it httpd started.

but all clients for ownCloud use https.. /sigh...

I'm using Letsencrypt for my ownCloud [personal] set up. It works great!

I am doing so also, but this specific system is actually not accessible off network (at this time). So I chose to use a StartSSL certificate.

JaredBusch

I have not gotten back to this yet. I just updated a couple users to use http for the short term. this is on my today or tomorrow list depending on how today goes.

JaredBusch

This has magically resolved itself.

I was finally able to come back to this tonight. Prior to doing anything, I

rebooted the server again.
ensured ownCloud was working via http.
uncommented the certificates in ssl.conf
restarted httpd
it successfully restart.

I changed nothing in the last week. I never even logged into the damned server.

I don't need this shit. I need to know what is happening so that when I reboot again next month, I can be confident the damned thing comes up.

scottalanmiller

Nothing scarier than things that fix themselves

JaredBusch

@scottalanmiller said:

Nothing scarier than things that fix themselves

Seriously upset about this. I will be rebooting the machine a few times today trying to make it do it again.