Timeout problem with far distance host

[Jürgen Haas]

This has been discussed at many places but this scenario is certainly different. Within my repository I do have one host which is on mainland China, i.e. inside the great firewall. And therefore the connection is significantly slower than anything else. My playbooks do work most of the time, but often I do get this `Timeout (12s) waiting for privilege escalation prompt:` error - only on that host and although it is a real mirror of all the others.

I'm pretty sure this is a problem because of the connection delay and wonder if I can change the timeout just for that host.

[Brian Coca]

You could set ConnectTimeout in ansible_ssh_args for that host


----------
Brian Coca

[Jürgen Haas]

Here is what I've added to the host vars for that particular host:

```

ansible_ssh_args: -o ConnectTimeout=240s -o ControlMaster=auto -o ControlPersist=60s -o ForwardAgent=yes

```

Unfortunately that didn't help. The connection timeout setting is new to this host, the other values are the default in my environment, that's why I've added them here too. Do you think any of those might be the problem? Or do I have any issues in my syntax?

[Brian Coca]

Can you show -vvvv output?


----------
Brian Coca

[Jürgen Haas]

That's strange, it doesn't use those settings although they are in the inventory for that host:

Using module file /usr/lib/python2.7/dist-packages/ansible/modules/core/system/setup.py
<fimfcn1> ESTABLISH SSH CONNECTION FOR USER: jurgenhaas
<fimfcn1> SSH: EXEC ssh -vvv -o ControlMaster=auto -o ControlPersist=60s -o ForwardAgent=yes -o KbdInteractiveAuthentication=no -o PreferredAuthentications=gssapi-with-mic,gssapi-keyex,hostbased,publickey -o PasswordAuthentication=no -o User=jurgenhaas -o ConnectTimeout=10 -o ControlPath=/home/jurgenhaas/.ansible/cp/ansible-ssh-%h-%p-%r fimfcn1 '/bin/sh -c '"'"'sudo -H -S  -p "[sudo via ansible, key=rczwjukcmjelutvbwindjqgbbipfxehz] password: " -u root /bin/sh -c '"'"'"'"'"'"'"'"'echo BECOME-SUCCESS-rczwjukcmjelutvbwindjqgbbipfxehz; /usr/bin/python'"'"'"'"'"'"'"'"' && sleep 0'"'"''

[Brian Coca]

you want:

ansible_ssh_common_args: -o ConnectTimeout=240s -o ControlMaster=auto
-o ControlPersist=60s -o ForwardAgent=yes


ansible_ssh_args does not exist, you DO have ssh_args in ansible.cfg
but that is 'global'.


--
----------
Brian Coca

[Jürgen Haas]

OK, that certainly makes a difference. The output of -vvvv is now as expected and I'll deploy that to may infrastructure and will watch the behavior the next couple of days, if that stabilizes the operations on that host.

Thanks Brian for your help.

[Jürgen Haas]

It's really strange. Still getting that timeout error although it is set differently:

TASK [Upgrade packages] ********************************************************

task path: /opt/ansible/playbooks/sanity.yml:36

Using module file /usr/lib/python2.7/dist-packages/ansible/modules/core/packaging/os/apt.py

<fimfcn1> ESTABLISH SSH CONNECTION FOR USER: jurgenhaas

<fimfcn1> SSH: EXEC ssh -vvv -o ControlMaster=auto -o ControlPersist=60s -o ForwardAgent=yes -o KbdInteractiveAuthentication=no -o PreferredAuthentications=gssapi-with-mic,gssapi-keyex,hostbased,publickey -o PasswordAuthentication=no -o User=jurgenhaas -o ConnectTimeout=10 -o ConnectTimeout=240s -o ControlMaster=auto -o ControlPersist=60s -o ForwardAgent=yes -o ControlPath=/home/jurgenhaas/.ansible/cp/ansible-ssh-%h-%p-%r fimfcn1 '/bin/sh -c '"'"'sudo -H -S  -p "[sudo via ansible, key=ftwxuktzuptwgsqvjrvadpkstknljlfl] password: " -u root /bin/sh -c '"'"'"'"'"'"'"'"'echo BECOME-SUCCESS-ftwxuktzuptwgsqvjrvadpkstknljlfl; /usr/bin/python'"'"'"'"'"'"'"'"' && sleep 0'"'"''

fatal: [fimfcn1]: FAILED! => {

    "failed": true, 

    "msg": "Timeout (12s) waiting for privilege escalation prompt: "

}

Any other idea I could go about? That error comes fairly early I have to say too, it isn't waiting that long.

[Brian Coca]

Also note that the 'displayed' timeout (12s) is the default value, not
inferred from ssh_args.


----------
Brian Coca