amygrant
Guru
- Joined
- Apr 4, 2010
- Messages
- 132
- Reaction score
- 3
I build a new Debian 8.6 container under proxmox. Perform an apt-get, apt-update to bring the packages current. I then download and run the wazo_install bash script but it errors out.
Here are the relevant lines of the output from execution of the bash script:
The symbolic link /etc/xivo-agentd/conf.d/050-xivo-consul-config.yml is reported as inaccessible. The link exists with 777 permissions.
How about the file the link points too? That exists as well
The log file /var/log/xivo-agentd.log tells us the problem:
Oddly, I ran across this same error a few short months ago and posted about it. However at the time that was due to user error on my part which I discovered.
I have reproduced this error three times and have tried manually installing (not running the script) xivo-agent via apt-get install but I run into the same UUID error.
I have tried containers of Debian 8.0 and updated the container to 8.6 and have tried a container based on 8.6
I have tried with and without the cloud-init package.
My /etc/hosts:
My /etc/hostname:
This seems to be a bug with the xivo-agent package perhaps limited in scope to installs under a proxmox container environment.
Here are the relevant lines of the output from execution of the bash script:
Code:
Adding system user xivo-agentd
Adding user xivo-agentd to group www-data
[....] Starting: xivo-agentdCould not read config file /etc/xivo-agentd/conf.d/050-xivo-consul-config.yml: [Errno 2] No such file or directory: '/etc/xivo-agentd/conf.d/050-xivo-consul-config.yml'
Could not read config file /etc/xivo-agentd/conf.d/050-xivo-consul-token.yml: [Errno 2] No such file or directory: '/etc/xivo-agentd/conf.d/050-xivo-consul-token.yml'
failed!
invoke-rc.d: initscript xivo-agentd, action "start" failed.
dpkg: error processing package xivo-agentd (--configure):
subprocess installed post-installation script returned error exit status 1
Setting up xivo-amid (16.16~20161205.122841.d0b9ba5) ...
Adding system user xivo-amid
Adding user xivo-amid to group www-data
The symbolic link /etc/xivo-agentd/conf.d/050-xivo-consul-config.yml is reported as inaccessible. The link exists with 777 permissions.
Code:
ls -alt /etc/xivo-agentd/conf.d/
total 8
drwxr-xr-x 3 root root 4096 Dec 21 07:55 ..
drwxr-xr-x 2 root root 4096 Dec 21 07:55 .
lrwxrwxrwx 1 root root 46 Dec 5 12:23 050-xivo-consul-config.yml -> /var/lib/consul/default_xivo_consul_config.yml
lrwxrwxrwx 1 root root 45 Dec 5 12:23 050-xivo-consul-token.yml -> /var/lib/consul/default_xivo_consul_token.yml
How about the file the link points too? That exists as well
Code:
root@voip2:/var/lib/consul# ls -al /var/lib/consul
total 28
drwxr-xr-x 4 consul consul 4096 Dec 21 07:55 .
drwxr-xr-x 54 root root 4096 Dec 21 07:56 ..
-rw-r----- 1 consul consul 106 Dec 21 07:55 default_xivo_consul_config.yml
-rw-r----- 1 consul consul 84 Dec 21 07:55 default_xivo_consul_token.yml
-rw------- 1 consul consul 65 Dec 21 07:55 master_token
drwxr-xr-x 3 consul consul 4096 Dec 21 07:55 raft
drwxr-xr-x 2 consul consul 4096 Dec 21 07:55 serf
The log file /var/log/xivo-agentd.log tells us the problem:
Code:
2016-12-21 07:55:41,840 [31330] (ERROR) (xivo_agent.bin.agentd): undefined environment variable XIVO_UUID
2016-12-21 07:55:41,840 [31330] (CRITICAL) (root): XIVO_UUID environment variable is not set
Traceback (most recent call last):
File "/usr/bin/xivo-agentd", line 5, in <module>
main()
File "/usr/lib/python2.7/dist-packages/xivo_agent/bin/agentd.py", line 137, in main
set_xivo_uuid(config, logger)
File "/usr/lib/python2.7/dist-packages/xivo/config_helper.py", line 144, in set_xivo_uuid
config['uuid'] = get_xivo_uuid(logger)
File "/usr/lib/python2.7/dist-packages/xivo/config_helper.py", line 139, in get_xivo_uuid
raise UUIDNotFound()
Oddly, I ran across this same error a few short months ago and posted about it. However at the time that was due to user error on my part which I discovered.
I have reproduced this error three times and have tried manually installing (not running the script) xivo-agent via apt-get install but I run into the same UUID error.
I have tried containers of Debian 8.0 and updated the container to 8.6 and have tried a container based on 8.6
I have tried with and without the cloud-init package.
My /etc/hosts:
Code:
root@xivo:~# cat /etc/hosts
127.0.0.1 localhost
::1 localhost ip6-localhost ip6-loopback
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
# --- BEGIN PVE ---
127.0.1.1 xivo
# --- END PVE ---
127.0.1.1 xivo.incrediblepbx.com
My /etc/hostname:
Code:
root@xivo:~# cat /etc/hostname
xivo
This seems to be a bug with the xivo-agent package perhaps limited in scope to installs under a proxmox container environment.