SOLVED: Amahi7 installation problems

[zion2008]

I followed the network troubleshooter.
From my hda I could not execute the nslookup command. It was not installed.
All the other steps went succesfull however up untill step 8
HDA Dashboard accessible?
I gave the command: curl http://hda
response: curl: (7) Failed connect to hda:80; Connectionn refused.

I entered the command: service httpd status
The response was:
Redirecting to /bin/systemctl status httpd.service
httpd.service - The Apache HTTP Server
Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled)
Active: failed (Result: exit-code) since Fri 2011-11-25 00:22:31 CET; 1 years 8 months ago
Process: 671 ExecStop=/usr/sbin/httpd $OPTIONS -k graceful-stop (code=exited, status=0/SUCCESS)
Process: 472 ExecStart=/usr/sbin/httpd $OPTIONS -DFOREGROUND (code=exited, status=1/FAILURE)

Nov 25 00:22:31 ams.ldqk httpd[472]: AH00557: httpd: apr_sockaddr_info_get() failed for ams.ldqk
Nov 25 00:22:31 ams.ldqk httpd[472]: AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using... message
Nov 25 00:22:31 ams.ldqk systemd[1]: httpd.service: main process exited, code=exited, status=1/FAILURE
Nov 25 00:22:31 ams.ldqk httpd[671]: AH00557: httpd: apr_sockaddr_info_get() failed for ams.ldqk
Nov 25 00:22:31 ams.ldqk httpd[671]: AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using... message
Nov 25 00:22:31 ams.ldqk httpd[671]: httpd (no pid file) not running
Nov 25 00:22:31 ams.ldqk systemd[1]: Failed to start The Apache HTTP Server.
Nov 25 00:22:31 ams.ldqk systemd[1]: Unit httpd.service entered failed state.

So definitely a problem with the http service.

Looking further
[root@ams logs]# ls -als
total 12
4 drwx------. 2 root root 4096 Aug 12 22:02 .
4 drwxr-xr-x. 12 root root 4096 Aug 14 18:29 ..
0 -rw-r--r-- 1 root root 0 Aug 12 22:02 access_log
4 -rw-r--r-- 1 root root 1550 Nov 25 2011 error_log

the error in the log
[Fri Nov 25 00:22:31.497525 2011] [suexec:notice] [pid 472] AH01232: suEXEC mechanism enabled (wrapper: /usr/sbin/suexec)
[Fri Nov 25 00:22:31.504900 2011] [unique_id:alert] [pid 472] (EAI 3)Temporary failure in name resolution: AH01564: unable to find IPv4 address of "ams.ldqk"
AH00016: Configuration Failed

Don't look at the dat/time I reset the hardware clock.
Do you have an answer for the displayed error codes?

zion2008

[bigfoot65]

As root user, you can do:

Code: Select all

systemctl status httpd systemctl restart httpd

This will show if HTTPD is running, then restart it. There is a message after running the status that gives you another command you can run to get more details of why it fails.

Not sure is this is part of your problem, but thought I would offer it.

[zion2008]

entered the command
systemctl status httpd

result is:
[root@ams ~]# systemctl status httpd
httpd.service - The Apache HTTP Server
Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled)
Active: failed (Result: exit-code) since Fri 2011-11-25 00:22:31 CET; 1 years 8 months ago
Process: 671 ExecStop=/usr/sbin/httpd $OPTIONS -k graceful-stop (code=exited, status=0/SUCCESS)
Process: 472 ExecStart=/usr/sbin/httpd $OPTIONS -DFOREGROUND (code=exited, status=1/FAILURE)

Nov 25 00:22:31 ams.ldqk httpd[472]: AH00557: httpd: apr_sockaddr_info_get() failed for ams.ldqk
Nov 25 00:22:31 ams.ldqk httpd[472]: AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using... mes
Nov 25 00:22:31 ams.ldqk systemd[1]: httpd.service: main process exited, code=exited, status=1/FAILURE
Nov 25 00:22:31 ams.ldqk httpd[671]: AH00557: httpd: apr_sockaddr_info_get() failed for ams.ldqk
Nov 25 00:22:31 ams.ldqk httpd[671]: AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using... mes
Nov 25 00:22:31 ams.ldqk httpd[671]: httpd (no pid file) not running
Nov 25 00:22:31 ams.ldqk systemd[1]: Failed to start The Apache HTTP Server.
Nov 25 00:22:31 ams.ldqk systemd[1]: Unit httpd.service entered failed state.

Then entered
systemctl restart httpd

result is:
[root@ams ~]# systemctl restart httpd
[root@ams ~]#

Then entered again systemctl status httpd

result is:
[root@ams ~]# systemctl status httpd
httpd.service - The Apache HTTP Server
Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled)
Active: active (running) since Wed 2013-08-14 23:14:44 CEST; 40s ago
Process: 671 ExecStop=/usr/sbin/httpd $OPTIONS -k graceful-stop (code=exited, status=0/SUCCESS)
Main PID: 2520 (httpd)
Status: "Total requests: 0; Current requests/sec: 0; Current traffic: 0 B/sec"
CGroup: name=systemd:/system/httpd.service
ââ2520 /usr/sbin/httpd -DFOREGROUND
ââ2539 PassengerWatchdog
ââ2542 PassengerHelperAgent
ââ2544 Passenger spawn server
ââ2547 PassengerLoggingAgent
ââ2554 /usr/sbin/httpd -DFOREGROUND
ââ2555 /usr/sbin/httpd -DFOREGROUND
ââ2557 /usr/sbin/httpd -DFOREGROUND
ââ2558 /usr/sbin/httpd -DFOREGROUND
ââ2559 /usr/sbin/httpd -DFOREGROUND

Aug 14 23:14:44 ams.ldqk systemd[1]: Started The Apache HTTP Server.


This seems OK by me.

Opened a browser from a client and entered http://hda and ..... I got my Amahi initialisation page asking me for a userid and passwords.
For now I'm happy because I can get to configure my HDA.

Maybe also entering a line in the troubleshooting page?
Thanks Bigfoot65. It works for now.

zion2008

[bigfoot65]

We think this may be related to what hostname is inserted in /etc/hostname. If you edit that file as root user and ensure the only thing in it is

localhost

, then httpd should start on reboot.

Might want to add that to the wiki as well.

[zion2008]

I'll startup my machine with the HDA on it tonight and then I will see if the problem is there again or not.
If so I'll enter localhost in /etc/hostname

I wonder if this relates to the values you can enter at the Network Interface part/screen when using the Amahi Express install. Default it is localmachine.localhost.

[bigfoot65]

On my machine, the default value was changed when the network iterface received an IP address. In my case, it was my production HDA, so it received a hostname from it.

Not sure if it matters but changing it seems to work. Might be a quirky bug that we will need to track.

[zion2008]

checked the value in /etc/hostname

[root@ams etc]# cat hostname
ams.ldqk

changed this to
[root@ams etc]# vi hostname
[root@ams etc]# cat hostname
localhost
[root@ams etc]#

then rebooted the machine.
[root@localhost etc]# cat hostname
localhost
[root@localhost etc]# ps -ef|grep http
root 473 1 0 23:39 ? 00:00:00 /usr/sbin/httpd -DFOREGROUND
apache 753 473 0 23:39 ? 00:00:00 /usr/sbin/httpd -DFOREGROUND
apache 754 473 0 23:39 ? 00:00:00 /usr/sbin/httpd -DFOREGROUND
apache 756 473 0 23:39 ? 00:00:00 /usr/sbin/httpd -DFOREGROUND
apache 757 473 0 23:39 ? 00:00:00 /usr/sbin/httpd -DFOREGROUND
apache 758 473 0 23:39 ? 00:00:00 /usr/sbin/httpd -DFOREGROUND
root 884 863 0 23:41 pts/0 00:00:00 grep --color=auto http
[root@localhost etc]#
You're right it works now. Only thing is that the localmachine name I assigned during install is now back to default value.

Now I think this is a bug that needs to be resolved in some way. Not sure what the solution should be though.

Some thoughts;
- remove the editable field in the network part of the installation and default localhost value is provided
- improve the installation script
- give a clear instruction in the network part of the installation what changes are permissable and even more important in what format

Shall I make another entry in the wiki page?

Thanks anyway for the help.

Now next step for me is adding a external usb hard drive and use greyhole.

Regards zion2008

[bigfoot65]

Glad to hear it's working. Not sure what the answer is here. Please file a bug if there is not already one there.

As for the wiki, yes please add it. I read your other guidance and was a little confused. Can you revise to make it easier to understand for others?

[zion2008]

I revised my post and added a second one for the http service. I do have the feeling that they are related and the root cause might be in the interpretation of the localhost.localdomain field in combination with the express install script.
zion2008

[bigfoot65]

Thanks for documenting. I think you might be right. This will require more research to sort out. For now, setting it to localhost seems to solve the issue.