Tcp_conn_validator doesn't respect timeout if the connection is dropped

[gmenuel]

Affected Puppet, Ruby, OS and module versions/distributions

• Puppet: 7.26.0
• Ruby: 2.7.8p225
• Distribution: Debian 10
• Module version: 2.0.0

How to reproduce (e.g Puppet code you use)

time puppet apply --debug -e "tcp_conn_validator { '1.2.3.4:443': timeout => 10}"

What are you seeing

The Tcp_conn_validator resource takes 2 minutes to execute.

What behaviour did you expect instead

The Tcp_conn_validator should fail after 10 seconds.

Output log

[...]
Info: Applying configuration version '1709733218'
Notice: Failed to connect to the host within timeout window of 10 seconds; giving up.
Error: Unable to connect to the  host. (1.2.3.4:443)
Error: /Stage[main]/Main/Tcp_conn_validator[1.2.3.4:443]/ensure: change from 'absent' to 'present' failed: Unable to connect to the  host. (1.2.3.4:443)
Debug: Class[Main]: Resource is being skipped, unscheduling all events
Debug: Stage[main]: Resource is being skipped, unscheduling all events
Debug: Finishing transaction 12780
Debug: Storing state
Debug: Pruned old state cache entries in 0.00 seconds
Debug: Stored state in 0.01 seconds
Notice: Applied catalog in 120.12 seconds
[...]

Any additional information you'd like to impart

The 2 minutes timeout seems to be the http_connect_timeout of Puppet, which is used in https://github.com/voxpupuli/puppet-healthcheck/blob/master/lib/puppet_x/puppet-community/tcp_validator.rb#L31 . I think this timeout should at least be configurable. And it should be at most the timeout of the Tcp_conn_resource.