emollusion

Hi! I am using Semaphore 2.7.22 and run it as a binary.

Hi! Removed the bug prefix. Great work! A small hickup in the automation so no big deal really, keep it in the pipeline and "fix" it when possible. Thank you!

Hi! Since this is still open and no full example has been provided here, I though i'd provide one example. Been tinkering with this today. However, our case makes use...

Hmm, looked through the code a bit better and see that it is referring to a conf setting, so my initial request above might be in error, So it seems...

Maybe this is a bit further for the work needed, but turning to PAM instead of LDAP all together could be a good improvement as well. PAM is already used...

Yes, the user telegraf has group pdns. This also corresponds to the working example above where the directory is group writeable, not world read nor write.

I have not found any reference in the [PowerDNS settings](https://doc.powerdns.com/authoritative/settings.html) for socket-mode, only [socket-dir](https://doc.powerdns.com/authoritative/settings.html#socket-dir). I tested the corresponding setting for PowerDNS as stated for the Recursor, socket-mode, but PowerDNS did...

> like this: ansible_ssh_common_args: '-o ProxyCommand="ssh -i {{ current_access_key_path }} -W %h:%p -q {{ bastion_user }}@{{ bastion_host }}"' Either like this, or using an ssh-agent with the key loaded into...

Confirming this behaviour in version 2.9.37

+1 on this one, good to have when implementing "machine to machine" where cert auth is not possible.