puppet infra status timeout option is unintuitive

[bastelfreak]

Describe the Bug

puppet infra status has a timeout option. I think that's used to tell the API how long it should try to calculate the response and abort if it doesn't finish in time. However it's a bit unintuitive. I've a primary with a few compilers. One of them isn't reachable via network. puppet infra status waits 120 seconds for a response, despite the default timeout of 5 seconds. I think this hardcoded 120s should also default to 5s and honour --timeout.

Truncated output of puppet infra status --debug

Debug: Caching connection for https://server:8081
Debug: Creating new connection for https://server:8140
Debug: Starting connection for https://server:8140
Debug: Exception raised while trying to fetch status for https://server:8140/status/v1/services
Debug: Puppet::HTTP::ConnectionError: Request to https://server:8140/status/v1/services?timeout=5 timed out connect operation after 120.051 seconds

Expected Behavior

puppet infra status should only wait 5s per node, or 5 s per API call per node, for a response.

Steps to Reproduce

• Install primary
• install a compiler
• shutdown the compiler
• run puppet infra status --debug

Environment

• PE 2021.7.4

Additional Context