junos_command

Archives Ansible Project
1

I am failing to make junos_command work

2017-04-25 15:28:49,152 p=4058 u=user | PLAY [sdwan-routers] ****************************************************************************************************
2017-04-25 15:28:49,171 p=4058 u=user | TASK [junos_command] ****************************************************************************************************
2017-04-25 15:28:49,572 p=4084 u=user | creating new control socket for host router.example.net:830 as user None
2017-04-25 15:28:49,572 p=4084 u=user | control socket path is /home/user/.ansible/pc/c08d46d021
2017-04-25 15:28:49,572 p=4084 u=user | current working directory is /home/user/ansible/sdwan
2017-04-25 15:28:49,572 p=4084 u=user | using connection plugin netconf
2017-04-25 15:28:49,629 p=4084 u=user | network_os is set to junos
2017-04-25 15:28:49,629 p=4084 u=user | ssh connection done, stating ncclient
2017-04-25 15:28:59,694 p=4084 u=user | failed to create control socket for host router.example.net
2017-04-25 15:28:59,695 p=4084 u=user | Traceback (most recent call last):
File “/usr/bin/ansible-connection”, line 298, in main
server = Server(sf_path, pc)
File “/usr/bin/ansible-connection”, line 135, in init
self.conn._connect()
File “/usr/lib/python2.7/dist-packages/ansible/plugins/connection/netconf.py”, line 95, in _connect
device_params={‘name’: self._network_os}
File “/usr/lib/python2.7/dist-packages/ncclient/manager.py”, line 149, in connect
return connect_ssh(*args, **kwds)
File “/usr/lib/python2.7/dist-packages/ncclient/manager.py”, line 118, in connect_ssh
session.connect(*args, **kwds)
File “/usr/lib/python2.7/dist-packages/ncclient/transport/ssh.py”, line 372, in connect
raise SSHError(“Could not open socket to %s:%s” % (host, port))
SSHError: Could not open socket to router.example.net:830

2017-04-25 15:29:19,598 p=4084 u=user | number of connection attempts exceeded, unable to connect to control socket
2017-04-25 15:29:19,599 p=4084 u=user | persistent_connect_interval=1, persistent_connect_retries=30
2017-04-25 15:29:19,625 p=4058 u=user | fatal: [router.example.net]: FAILED! => {“changed”: false, “failed”: true, “msg”: “unable to open shell. Please see: https://docs.ansible.com/ansible/network_debug_troubleshooting.html#unable-to-open-shell”, “rc”: 255}
2017-04-25 15:29:19,626 p=4058 u=user | PLAY RECAP **************************************************************************************************************
2017-04-25 15:29:19,626 p=4058 u=user | router.example.net : ok=0 changed=0 unreachable=0 failed=1

I can only ssh to the router through a proxy host and the router does not have any port 830 open and I verified.

Please advise a workaround…

Thanks

2

I might be mistaken, but when dealing with network devices there should be something like:

connection: local

somewhere at the beginning of the playbook.

And the content of the playbook itself would be helpful here I think.

Please also check:
https://www.ansible.com/webinars-training/ansible-playbooks-using-network-modules

Od: ansible-project@googlegroups.com [mailto:ansible-project@googlegroups.com] W imieniu Asif Iqbal
Wysłano: 25 kwietnia 2017 21:31
Do: Ansible Project ansible-project@googlegroups.com
Temat: [ansible-project] junos_command

3

HI Asif, junos_command requires netconf to be enabled on the remote device.

4

Based on the error message, authentication parameters are not passed properly to the module.

I see it’s using netconf; so i think you have used connection method right…

Cheers

5

Boker Tov Asaf, did you managed to solve it?

I have exactly the same problem with ex4200 and Ansible 2.3.3. netconf is enabled on port 830 and accessible:

[ playbooks]# ssh -s -p 830 ansible@juniper_12 netconf

Welcome to Juniper ex4300-48t

Password:

urn:ietf:params:netconf:base:1.0

urn:ietf:params:netconf:capability:candidate:1.0

urn:ietf:params:netconf:capability:confirmed-commit:1.0

urn:ietf:params:netconf:capability:validate:1.0

urn:ietf:params:netconf:capability:url:1.0?scheme=http,ftp,file

urn:ietf:params:xml:ns:netconf:base:1.0

urn:ietf:params:xml:ns:netconf:capability:candidate:1.0

urn:ietf:params:xml:ns:netconf:capability:confirmed-commit:1.0

urn:ietf:params:xml:ns:netconf:capability:validate:1.0

urn:ietf:params:xml:ns:netconf:capability:url:1.0?protocol=http,ftp,file

http://xml.juniper.net/netconf/junos/1.0

http://xml.juniper.net/dmi/system/1.0

11528

]]>]]>

This is debug log:

2018-02-09 01:13:25,964 p=17223 u=root | creating new control socket for host 192.168.89.240:830 as user ansible
2018-02-09 01:13:25,965 p=17223 u=root | control socket path is /root/.ansible/pc/1007ccc67b
2018-02-09 01:13:25,965 p=17223 u=root | current working directory is /etc/ansible/playbooks
2018-02-09 01:13:25,966 p=17223 u=root | using connection plugin netconf
2018-02-09 01:13:25,968 p=17223 u=root | failed to create control socket for host 192.168.89.240
2018-02-09 01:13:25,969 p=17223 u=root | Traceback (most recent call last):
File “/opt/ansible/bin/ansible-connection”, line 298, in main
server = Server(sf_path, pc)
File “/opt/ansible/bin/ansible-connection”, line 134, in init
self.conn = connection_loader.get(play_context.connection, play_context, sys.stdin)
File “/opt/ansible/lib/ansible/plugins/init.py”, line 358, in get
self._module_cache[path] = self._load_module_source(name, path)
File “/opt/ansible/lib/ansible/plugins/init.py”, line 343, in _load_module_source
module = imp.load_source(full_name, path, module_file)
File “/opt/ansible/lib/ansible/plugins/connection/netconf.py”, line 39, in
raise AnsibleError(“ncclient is not installed”)
AnsibleError: ncclient is not installed

If you realised how to fix it, please share.
Thanks!

6

AnsibleError: ncclient is not installed

Please install ncclient and retry.

Command to install ncclient using pip:

pip install ncclient

Regards,
Ganesh

7

Hi Ganesh,
Thanks!
I moved FWD and installed it from git, now it is listed in packages:

[root@server]# pydoc modules | grep -i ncc
UserDict datetime ncclient (package) sre_parse
[root@server]#

but I still have same error, absolutely the same.

It looks to me as Ansible configuration issue.
I don’t exactly know where to put ncclient module(path where Ansible looks for it) which I installed with setup.py.
At the moment I just added its patch to env (so its listed by pydoc):

ANSIBLE_LIBRARY=/opt/ansible/lib/ansible/modules:/root/Ansible_install/ncclient-0.4.2

ANSIBLE_HOME=/opt/ansible
PYTHONPATH=/opt/ansible/lib:/root/Ansible_install/ncclient-0.4.2

[root@playbooks]# ansible --version
7198 1518284728.95276: starting run
ansible 2.3.3.0
config file = /etc/ansible/ansible.cfg
configured module search path = [u’/opt/ansible/lib/ansible/modules’, u’/root/Ansible_install/ncclient-0.4.2’]
python version = 2.6.8 (unknown, Nov 7 2012, 14:47:34) [GCC 4.1.2 20080704 (Red Hat 4.1.2-52)]
[root@playbooks]#

I know, Im not cool by running this directly as root, but this is just test env.

Thanks.

8

OK, looks like I found the root cause:

VIP_CR:[root@upm1 playbooks]# python
Python 2.6.8 (unknown, Nov 7 2012, 14:47:34)
[GCC 4.1.2 20080704 (Red Hat 4.1.2-52)] on linux2
Type “help”, “copyright”, “credits” or “license” for more information.
>>> from ncclient import manager
Traceback (most recent call last):
File “”, line 1, in
File “build/bdist.linux-i686/egg/ncclient/manager.py”, line 22, in
File “build/bdist.linux-i686/egg/ncclient/operations/init.py”, line 16, in
File “build/bdist.linux-i686/egg/ncclient/operations/rpc.py”, line 18, in
File “build/bdist.linux-i686/egg/ncclient/xml_.py”, line 22, in
ImportError: /usr/lib/libexslt.so.0: symbol gcry_cipher_setkey, version GCRYPT_1.2 not defined in file libgcrypt.so.11 with link time reference
>>>

ncclient needs to resolve dependencies.

9

OK, Finally I got ncclient installed and it works fine (all dependencies solved),
But this is what confusing me now:

2018-02-11 02:01:07,225 p=17431 u=root | ssh connection done, stating ncclient
2018-02-11 02:01:09,832 p=17431 u=root | ncclient manager object created successfully
2018-02-11 02:01:09,833 p=17431 u=root | connection established to 192.168.89.240 in 0:00:02.721349
2018-02-11 02:01:10,108 p=17431 u=root | incoming request accepted on persistent socket
2018-02-11 02:01:10,109 p=17431 u=root | socket operation is CONTEXT
2018-02-11 02:01:10,111 p=17431 u=root | socket operation is EXEC
2018-02-11 02:01:10,112 p=17431 u=root | socket operation completed with rc 0
2018-02-11 02:01:10,143 p=16797 u=root | open_session() returned 0 ok
2018-02-11 02:01:10,577 p=16797 u=root | Using module file /opt/ansible-2.3.3.0/lib/ansible/modules/network/junos/junos_command.py
2018-02-11 02:01:10,713 p=16797 u=root | The full traceback is:
Traceback (most recent call last):
File “/tmp/ansible_iVnNY8/ansible_module_junos_command.py”, line 164, in ?
from functools import partial
ImportError: No module named functools

2018-02-11 02:01:10,714 p=16797 u=root | fatal: [hsbu12]: FAILED! => {
“changed”: false,
“failed”: true,

This module is installed and listed but, junos_command.py can’t find it.
Does anybody have any ideas?

Thanks in advance!