kyl191 / ansible-role-openvpn Goto Github PK

https://github.com/geerlingguy/ansible-role-php/blob/master/.github/stale.yml

or github actions natively supports stale & greeting, but not as configurable

We are not able to select the directory where to fetch users's configuration,

We juste need to set a variable, wich is defined on /tmp/ansible by default.

I send a PR in that sense,

Bye

Interestingly, the ldap plugin is natively available for Fedora: https://src.fedoraproject.org/rpms/openvpn-auth-ldap

But the CentOS versions aren't updated, looks like dependencies are the issue.

Trying a copr version of it isn't too helpful: https://copr.fedorainfracloud.org/coprs/kyl191/openvpn-auth-ldap/
gcc-objc was removed in RHEL8, there's no EPEL replacement, so it has to be pulled in from an external repo. (copr supports this, thankfully)

re2c just isn't updated: https://src.fedoraproject.org/rpms/re2c
I can build it locally on CentOS 8, so that should be easier to solve.

If you try make management interface listen via TCP (setting openvpn_management_bind to something like "localhost 7505"), the resulting config still contains management-client-user option, which results in OpenVPN server failing to start with this error message:
Options error: --management-client-(user|group) can only be used on unix domain sockets

:-/

templates/ca.conf.j2:

default_crl_days = 30

Which means that upon "gencrl" "lastupdate" is set to the run timestamp and "nextupdate" is set to 30 days in the future.
After that, nothing in ansible-role-openvpn seems to update "nextupdate" other than doing an actual manual revoke / manual running revoke.sh .
When more than 30 days have passed since either the initial creation or the last revoke,
openvpn 2.4 or later then freaks out because the CRL has expired and doesn't let anyone connect with crl-verify enabled. (https://community.openvpn.net/openvpn/wiki/CertificateRevocationListExpired)

templates/crl-cron.sh.j2:
only checks there's a less than 10 days difference between "lastupdate" and "nextupdate", but my impression is that difference will always be "default_crl_days", so it never triggers the regenerate/renew.
Was the intention to do a check of current timestamp (date +%s) vs "nextupdate" instead?
Straight up having crl-cron.sh run revoke.sh without the "less than 10 days" check should probably work OK instead, since that would do a weekly regenerate/renew instead of 20-27 days after the last revoke/renew.
(keeping the short renew period should from what I could google be a bit safer (shorter caching, shorter keeping antique revoked cerificates listed in it) than the often used workaround of throwing default_crl_days to 3650 (10 years)...)

templates/server.conf.j2:

{% if openvpn_crl_path is defined %}
crl-verify {{openvpn_crl_path}}
{% endif %}
{% if openvpn_use_crl|bool %}
crl-verify {{openvpn_key_dir}}/ca-crl.pem
{% endif %}

I'm guessing that was meant to be "openvpn_use_crl" as boolean on/off toggle with
"{{openvpn_key_dir}}/ca-crl.pem" as default if "openvpn_crl_path" isn't defined?
In its current form it writes 2 "crl-verify" lines if both options are defined...

(I'm decent at reading/decoding programming languages, but a newbie at writing ansible/jinja2, so unfortunately couldn't put the above in the form of a pull request...)

subj

Hi, I want to send you PR, but
ERROR: Permission to kyl191/ansible-role-openvpn.git denied to de1ayer. fatal: Could not read from remote repository.
May I ask to grant some access?

It would be good to be able to pin the openvpn package and the other packages to a specific version to make sure that installations are identical even if new versions of the packages are released. For example:

- name: Install openvpn
  package:
   name: "{{ item }}"
   state: present
  with_items:
    - openvpn=2.4.7-1ubuntu2
    - openssl=1.1.1f-1ubuntu2.2

I want to use OpenVPN to talk from the server to clients (clients are behind NAT behind a dynamic IP address). It turned out that I needed MASQUERADE instead of SNAT to achieve that, how would you feel about adding an option for that? I can do the actual work if it's something you'd accept.

I need some examples on how to use your repository for:

1. install the server,
2. create 2 users (user and password)
3. install a client to use user1 and password1

Hi all,

I'm having a few issues, when running this playbook I'm getting the following error when running the handlers at the end (specifically restart openvpn & save iptables).
ERROR! Timeout (12s) waiting for privilege escalation prompt
and when I tried running using "-c paramiko" i got this:
UNREACHABLE! => {"changed": false, "msg": "Authentication timeout.", "unreachable": true}

Any ideas what's going on here?
I can manually SSH to the server and perform both tasks (restart openvpn & save iptables) with no issue.

Centos 7 minimal 2003 don't have installed crontab

Role application stops on error

TASK [kyl191.openvpn : Add cron to check every Saturday if the CRL needs to be renewed] ***************************************************************************************************************************** fatal: [vpn-server-01]: FAILED! => {"changed": false, "msg": "Failed to find required executable crontab in paths: /usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin"}

Decision
yum install cronie

deployed openvpn using this role, and tred to use authentication plugin from duo.
no call to duo server when server.conf has "proto udp6". works fine when "proto udp"

Great work !
Just having a small issue at the end. Any clue ?
On Ubuntu 16.04 LTS

TASK [kyl191.openvpn : Setup openvpn auto-start & start] ************************************************************************************************************************************************************************************************************
fatal: [47.254.156.204]: FAILED! => {"changed": false, "msg": "Unable to start service openvpn@openvpn_udp_1194.service: Job for openvpn@openvpn_udp_1194.service failed because the control process exited with error code. See "systemctl status openvpn@openvpn_udp_1194.service" and "journalctl -xe" for details.\n"}

Hello there,
It's been a while 🙂.

I'm deploying on Rocky Linux 8. I got an issue with the ldap auth module.
Here the log

PLUGIN_INIT: could not load plugin shared object /usr/lib64/openvpn/plugin/lib/openvpn-auth-ldap.so: /usr/lib64/openvpn/plugin/lib/openvpn-auth-ldap.so: cannot open shared object file: No such file or directory: No such file or directory (errno=2)

On Rocky 8 (and on RHEL 8) the correct path is /usr/lib64/openvpn/plugins/openvpn-auth-ldap.so (just a 's' added to 'plugin' actually).

We can push DNS with dhcp-option DNS, can we also get an option to push dhcp-option DOMAIN ?

There is a bad var usage in the task server_keys.yml.

fatal: [vpn-test.mylab.com]: FAILED! => {"msg": "The conditional check 'file_result.state == 'absent'' failed. The error was: error while evaluating conditional (file_result.state == 'absent'): 'dict object' has no attribute 'state'\n\nThe error appears to be in '/home/lmaurice/ws/infra/ansible-infrastructure/galaxy_roles/openvpn/tasks/server_keys.yml': line 137, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n- name: create certificate revocation list database if required\n  ^ here\n"}

I solved the problem replacing in tasks/server_keys.yml line 137 :
This

- name: create certificate revocation list database if required
  file:
    path: "{{openvpn_key_dir}}/index.txt"
    state: touch
  when: file_result.state == 'absent'

By this

- name: create certificate revocation list database if required
  file:
    path: "{{openvpn_key_dir}}/index.txt"
    state: touch
  when: '"absent" in file_result'

#117 and #120 are around the specific listing of Fedora & CentOS versions that started using systemd.

More distros can use the role than are otherwise in the list (https://github.com/kyl191/ansible-role-openvpn/blob/master/tasks/set_facts.yml#L16), adding support for other distros needs to be done manually.

ansible-role-openvpn should run on systems with SELinux enabled

Hello,

I am curious about how crl-cron.sh is supposed to work.

Here is the content of the file

#!/bin/sh 
lastUpdate=$(date --date "$(openssl crl -in /etc/openvpn/keys/ca-crl.pem -noout -lastupdate | cut -d'=' -f2)" +%s) 
nextUpdate=$(date --date "$(openssl crl -in /etc/openvpn/keys/ca-crl.pem -noout -nextupdate | cut -d'=' -f2)" +%s) 
if [ $(( (nextUpdate - lastUpdate) / 86400 )) -le 10 ]; then 
    sh /etc/openvpn/keys/revoke.sh   
fi

lastupdate and nextupdate are static values ; they are both not changing unless explicitely modified (see
https://community.openvpn.net/openvpn/wiki/CertificateRevocationListExpired ).

that's why I am not understanding the if statement. It will systematically be 30 (days) ?

Can someone explain?

Today I had to manually launch the script /etc/openvpn/keys/revoke.sh because there nextUpdate < now.

Thanks

I just tested the role on a Centos8 VM and it seems to work perfectly.
Maybe you can add this OS to the list of supported one. And add it to Travis Ci as well.

several tasks use the internal bash utility 'command' which cannot be executed by sudo, e.g. the following task on Ubuntu 18.04:

- name: Check for iptables
  command: command -v iptables
...

can trigger the following error:

TASK [openvpn : Check for iptables] ***************************************************************
task path: /home/ubuntu/git/clusterops/ansible/roles/openvpn/tasks/firewall.yml:15
<localhost> ESTABLISH LOCAL CONNECTION FOR USER: ubuntu
<localhost> EXEC /bin/sh -c 'echo ~ubuntu && sleep 0'
<localhost> EXEC /bin/sh -c '( umask 77 && mkdir -p "` echo /home/ubuntu/.ansible/tmp/ansible-tmp-1564654912.301547-91073605449550 `" && echo ansible-tmp-1564654912.301547-91073605449550="` echo /home/ubuntu/.ansible/tmp/ansible-tmp-1564654912.301547-91073605449550 `" ) && sleep 0'
Using module file /home/ubuntu/miniconda3/lib/python3.7/site-packages/ansible/modules/commands/command.py
<localhost> PUT /home/ubuntu/.ansible/tmp/ansible-local-12565unktj_z2/tmpl43quayn TO /home/ubuntu/.ansible/tmp/ansible-tmp-1564654912.301547-91073605449550/AnsiballZ_command.py
<localhost> EXEC /bin/sh -c 'chmod u+x /home/ubuntu/.ansible/tmp/ansible-tmp-1564654912.301547-91073605449550/ /home/ubuntu/.ansible/tmp/ansible-tmp-1564654912.301547-91073605449550/AnsiballZ_command.py && sleep 0'
<localhost> EXEC /bin/sh -c 'sudo -H -S  -p "[sudo via ansible, key=hlycgfpiptyjmgxvykfljpqyuoalticm] password:" -u root /bin/sh -c '"'"'echo BECOME-SUCCESS-hlycgfpiptyjmgxvykfljpqyuoalticm ; /usr/bin/python /home/ubuntu/.ansible/tmp/ansible-tmp-1564654912.301547-91073605449550/AnsiballZ_command.py'"'"' && sleep 0'
<localhost> EXEC /bin/sh -c 'rm -f -r /home/ubuntu/.ansible/tmp/ansible-tmp-1564654912.301547-91073605449550/ > /dev/null 2>&1 && sleep 0'
ok: [localhost] => {
    "changed": false,
    "cmd": "command -v iptables",
    "failed_when_result": false,
    "invocation": {
        "module_args": {
            "_raw_params": "command -v iptables",
            "_uses_shell": false,
            "argv": null,
            "chdir": null,
            "creates": null,
            "executable": null,
            "removes": null,
            "stdin": null,
            "stdin_add_newline": true,
            "strip_empty_ends": true,
            "warn": true
        }
    },
    "msg": "[Errno 2] No such file or directory",
    "rc": 2
}

TASK [openvpn : Fail on both firewalld & ufw] *****************************************************
task path: /home/ubuntu/git/clusterops/ansible/roles/openvpn/tasks/firewall.yml:22
skipping: [localhost] => {
    "changed": false,
    "skip_reason": "Conditional result was False"
}

TASK [openvpn : Fail on no firewall detected] *****************************************************
task path: /home/ubuntu/git/clusterops/ansible/roles/openvpn/tasks/firewall.yml:27
fatal: [localhost]: FAILED! => {
    "changed": false,
    "msg": "No firewall detected, install one before proceeding (firewalld||ufw||iptables)"
}

PLAY RECAP ****************************************************************************************
localhost                  : ok=34   changed=0    unreachable=0    failed=1    skipped=16   rescued=0    ignored=0   

Can we replace them with programs that are less dependent on bash environment, e.g. 'which iptables' instead of 'command -v iptables'?

When revoking the clients the CA cert is not removed from the server.

Current Code

- name: Revoke certificates
  command: sh revoke.sh {{ item }}.crt
  changed_when: true
  args:
    chdir: "{{ openvpn_key_dir }}"
  with_items:
    - '{{ openvpn_revoke_these_certs }}'
    - '{{ openvpn_cert_sync_revoke | default([]) }}'

New Code

- name: Revoke client crt
  file:
    path: "{{ openvpn_key_dir }}/{{ item }}.crt"
    state: absent
    force: true
  with_items:
    - '{{ openvpn_revoke_these_certs }}'
    - '{{ openvpn_cert_sync_revoke | default([]) }}'

Inansible-role-openvpn/tasks/ufw.yml the default input policy is hardcoded to 'allow'.

When providing items in the clients variable, the role creates the client's private key, CSR and certificate on the target. However, the target only needs the certificate in order to add the client to the revocation list later on.

The best scenario would be to generate the clients' key on their own side, but it is out of scope of this role. However, the role should at least remove the private key, CSR and ovpn files at the end of the client_keys task.

It looks like it's design for running a role in a Docker image, which is the tests do right now, except Github Actions don't allow us to actually start containers with the proper entrypoint (systemd).
And there's something about the docker create && docker start and docker run cycle that doesn't allow containers to last on GH infra, despite docker run working fine on other platforms.
Maybe molecule will overcome that?

Crib liberally from https://github.com/diodonfrost/ansible-role-podman/blob/master/.github/workflows/molecule.yml

crl-cron.sh need to include full path for the ca-crl.pem or it will throw errors in cron logs

What happened:

Currently, I installed the role with an LDAP configuration. However, during the compiling in compile_ldap_plugin.yml it results in:

{
    "changed": true,
    "cmd": "autoconf\nautoheader\n./configure --prefix=/usr/lib64/openvpn/plugin --with-openvpn=/sbin/openvpn CFLAGS=\"-fPIC\" OBJCFLAGS=\"-std=gnu11\"\nmake\nmake install\n",
    "delta": "0:00:00.050354",
    "end": "2022-01-11 09:39:29.705113",
    "msg": "non-zero return code",
    "rc": 127,
    "start": "2022-01-11 09:39:29.654759",
    "stderr_lines": [
        "/bin/sh: autoconf: command not found",
        "/bin/sh: line 1: autoheader: command not found",
        "./configure: line 473: sed: command not found",
        "./configure: line 472: expr: command not found",
        "./configure: line 473: sed: command not found",
        "./configure: line 487: sed: command not found",
        ": error: cannot create .lineno; rerun with a POSIX shell",
        "/bin/sh: line 3: make: command not found",
        "/bin/sh: line 4: make: command not found"
    ],
    "stdout": "",
    "stdout_lines": []
}

Are there any ideas why I ran into that problem?

What you expected to happen:

Compilation should run through

Anything else we need to know?:

I checked the $PATH variable and I don't see any misconfiguration: /home/centos/.local/bin:/home/centos/bin:/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin

• sed: /usr/bin/sed (which sed)
• autoconf /usr/bin/autoconf
• autoheader /usr/bin/autoheader
• make /usr/bin/make

System

CentOS Linux release 8.3.2011

For the openssl operations should this role make use of the collection 'community.crypto.x509_certificate_info'?

What do people think, i am happy to do the work if others agree.

Openvpn server is throwing this warning on the logs:

DEPRECATED OPTION: --cipher set to 'AES-256-CBC' but missing in --data-ciphers (AES-256-GCM:AES-128-GCM). Future OpenVPN version will ignore --cipher for cipher negotiations. Add 'AES-256-CBC' to --data-ciphers or change --cipher 'AES-256-CBC' to --data-ciphers-fallback 'AES-256-CBC' to silence this warning.

There do not seem to be options to set up this on the module.

The openvpn_custom_dns option which is described in the README file does not actually work. Even if I set openvpn_custom_dns to a list of servers and also set openvpn_set_dns to true, my custom DNS server is not pushed to the clients.

I think the problem is that templates/server.conf.j2 uses the openvpn_dns_servers variable (in https://github.com/kyl191/ansible-role-openvpn/blob/master/templates/server.conf.j2#L64) rather than using the values from openvpn_custom_dns.

I see two solutions:

• document the current behavior, which in my understanding is this:

  • openvpn_dns_servers is the list of DNS servers to use
  • openvpn_custom_dns is a bool value that must be set to true to actually use the entries from openvpn_dns_servers (rather than the Google/Cloudflare servers)
  • openvpn_set_dns must be set to true as well to use the values from openvpn_dns_servers

• simplify the current behavior: remove openvpn_custom_dns from code and documentation, and use only openvpn_dns_servers

Hi, I just got this error on Raspbian Stretch (raspberry pi) when connecting to my server.

Was connecting fine until today (been using it for 2-3 months now)

How can I fix this problem, tried running the role again to no avail

Thanks !

Please add version tags, so one can lock role version when using it from Galaxy. Thanks

The template sets compress lzo, this configuration option is deprecated in 2.4 and was removed in 2.5.

I suggest to make including this setting configurable and maybe not activate it by default. Maybe setting lz4 compression should be made possible.

The install fails with the following error
{"changed": false, "failed": true, "msg": "Could not find the requested service iptables: cannot enable"}

Ansible version: 2.2.2.0 tested on 2.2.1.0 also
OS: Mac OSX
External Host: Centos 7 AWS AMI ami-7abd0209

Hi Team,

I tried to configure the ldap authentication but auth/ldap.conf file is not getting updated.

Below is my setup-openvpn.yml file. Could you please suggest what is wrong in it.

• hosts: vpn01
  become: true
  gather_facts: true
  roles:
  • role: kyl191.openvpn
    openvpn_port: 1194
    clients:
    • client1
      openvpn_push:
    • route 10.20.30.0 255.255.255.225
    • route 10.x.x.0 255.255.255.0
    • dhcp-option DNS 10.x.x.46
      iptables_service: firewalld
      openvpn_use_ldap: true
      ldap:
      url: 'ldaps://ldap01'
      bind_dn: cn=Manager,dc=example,dc=uk
      bind_password: test
      tls_ca_cert_file: /etc/openldap/certs/ca_certs.pem
      base_dn: dc=example,dc=uk
      search_filter: (&(uid=%u))
      require_group: true
      group_base_dn: ou=vpn,dc=example,dc=uk
      group_search_filter: (|(cn=DevOps)(cn=users))

Hello,

I tried to provision openvpn on Amazon Linux 2 AMI (ami-02f9ea74050d6f812) and Centos 7.

- hosts: all
  gather_facts: true
  roles:
    - role: kyl191.openvpn
      openvpn_sync_certs: true # BUG
      openvpn_duplicate_cn: true
      openvpn_server_hostname: xxx.xxx.com
      become: true
      clients:
        - victor
        - celine
      openvpn_push:
        - dhcp-option DNS X.X.X.X

When enabling openvpn_sync_certs I have an error when there is no client and another if there are already clients.
Error when there are no clients yet:

TASK [kyl191.openvpn : Get existing certs] *******************************************************************************************************
fatal: [172.31.11.150]: FAILED! => {"changed": true, "cmd": "set -o pipefail\nls /etc/openvpn/keys | grep '\\.csr' | grep -v 'server.csr' | sort | cut -f1 -d'.' > /tmp/existing-certs\n", "delta": "0:00:00.006633", "end": "2021-05-10 09:10:32.823718", "msg": "non-zero return code", "rc": 1, "start": "2021-05-10 09:10:32.817085", "stderr": "", "stderr_lines": [], "stdout": "", "stdout_lines": []}

Error when there are already clients:

TASK [kyl191.openvpn : Get difference between 'clients' variable and existing certs] ***********************************************************************************************************************
fatal: [172.31.15.142]: FAILED! => {"changed": true, "cmd": "set -o pipefail\nsort -o /tmp/client-list /tmp/client-list\ndiff /tmp/client-list /tmp/existing-certs | grep '>' | awk '{print $2}'\n", "delta": "0:00:00.021649", "end": "2021-05-10 07:55:40.363357", "msg": "non-zero return code", "rc": 1, "start": "2021-05-10 07:55:40.341708", "stderr": "", "stderr_lines": [], "stdout": "", "stdout_lines": []}

Cheers and thanks

When I run the playbook I get:

TASK [etc : iptables - save everything (Debian/Ubuntu)] ************************
fatal: [vpn_germany1]: FAILED! => {"changed": false, "cmd": "/etc/init.d/iptables-persistent save", "failed": true, "msg": "[Errno 2] No such file or directory", "rc": 2}

I verified the installation of iptables-persistent:

root@DEBIAN-TEST:~# apt-get install iptables-persistent
Reading package lists... Done
Building dependency tree
Reading state information... Done
iptables-persistent is already the newest version (1.0.4).
The following packages were automatically installed and are no longer required:
linux-headers-4.4.0-21 linux-headers-4.4.0-21-generic linux-headers-4.4.0-31
linux-headers-4.4.0-31-generic linux-image-4.4.0-21-generic
linux-image-4.4.0-31-generic linux-image-extra-4.4.0-21-generic
linux-image-extra-4.4.0-31-generic
Use 'apt autoremove' to remove them.
0 upgraded, 0 newly installed, 0 to remove and 79 not upgraded.

Currently the role can handle only one configuration per server.

It would be usefull to allow the configuration of multiple config.

Ubuntu (https://cloud-images.ubuntu.com/locator/ec2/):
16.10: ubuntu/images/hvm-ssd/ubuntu-yakkety-16.10-amd64-server-20170103 - ami-4ae1fb5d
16.04: ubuntu/images/hvm-ssd/ubuntu-xenial-16.04-amd64-server-20170202 - ami-f0768de6
14.04: ubuntu/images/hvm-ssd/ubuntu-trusty-14.04-amd64-server-20170202.1 - ami-af22d9b9
12.04: ubuntu/images/hvm-ssd/ubuntu-precise-12.04-amd64-server-20170202 - ami-a49e65b2

Debian (user: admin, https://wiki.debian.org/Cloud/AmazonEC2Image/Jessie):
8.7: debian-jessie-amd64-hvm-2017-01-15-1221-ebs - ami-b14ba7a7
7.8: debian-wheezy-amd64-hvm-2015-01-28-ebs - ami-e0efab88

Fedora:
25: Fedora-Cloud-Base-25-20170214.0.x86_64-us-east-1-HVM-gp2-0 - ami-ef70bff9
24: Fedora-Cloud-Base-24-20161207.0.x86_64-us-east-1-HVM-gp2-0 - ami-a11c1ab6
23: Fedora-Cloud-Base-23-20160626.x86_64-us-east-1-HVM-gp2-0 - ami-ec458981

CentOS:
7 (1602): https://aws.amazon.com/marketplace/pp/B00O7WM7QW
6 (2014-09-29): https://aws.amazon.com/marketplace/pp/B00A6KUVBW
Other point releases not available?

@kyl191 what if I want to remove the user, I didn't find anything related to it over the net. Do you have an idea about it. Thanks

On the step "generate client config", the client cert and private key of the client is outputted to the console, despite being a sensitive info. Example:

TASK [kyl191.openvpn : generate client config] ***********************************************************************************************************************************************************
changed: [server] => (item=[{'content': 'LS0tLS1CR.......redacted.........EUtLS0tLQo=', 'source': '/etc/openvpn/keys/client.crt', 'encoding': 'base64', 'invocation': {'module_args': {'src': '/etc/openvpn/keys/client.crt'}}, 'failed': False, 'changed': False, 'item': 'client', 'ansible_loop_var': 'item'}, {'content': 'LS0tLS1C.....redacted......0tLS0K', 'source': '/etc/openvpn/keys/client.key', 'encoding': 'base64', 'invocation': {'module_args': {'src': '/etc/openvpn/keys/client.key'}}, 'failed': False, 'changed': False, 'item': 'client', 'ansible_loop_var': 'item'}])

This should not be printed to stdout. I suggest adding no_log: true to that task, maybe overridable using a variable in defaults.

Made this change (and fixed a syntax error) and all the systemd-based docker images fell over.

Commented out the port randomization and systemd docker images started working again.

Can't replicate it on a local docker image, but Travis fails with

TASK [ansible-role-openvpn : setup openvpn auto-start & start] *****************
task path: /etc/ansible/roles/ansible-role-openvpn/tasks/config.yml:37
fatal: [127.0.0.1]: FAILED! => {"changed": false, "failed": true, "msg": "Unable to start service openvpn@openvpn_udp_49295.service: Job for openvpn@openvpn_udp_49295.service failed because the control process exited with error code.\nSee \"systemctl status openvpn@openvpn_udp_49295.service\" and \"journalctl -xe\" for details.\n"}

and no logs are written to /var/log/openvpn.log.

❄️

When the option

tls_auth_required

is disabled, the tls_auth key still gets to the client config. Because of this, the Windows client cannot connect to the server.

OS: AlmaLinux (8)
Resolution: install checkpolicy on the server

    - name: Install checkpolicy
      dnf:
        name: checkpolicy

Deployment fails with fatal: [xxxxxxxxx.com]: FAILED! => {"changed": false, "msg": "No firewall detected, install one before proceeding (firewalld||ufw||iptables)"}

ufw was installed from https://github.com/weareinteractive/ansible-ufw

Galaxy already checks for it
https://github.com/geerlingguy/ansible-role-php/blob/master/.ansible-lint

When I do "ansible-playbook ... ":
FAILED! => {"changed": false, "failed": true, "msg": "ERROR: Exception caught: set_masquerade_enabled() takes exactly 1 argument (6 given)"}