Grab kube_node_info as kubernetes_state.node.count

[therc]

What does this PR do?

It picks the kubelet's kube_node_info metric and exports it as kubernetes_state.node.count.

Motivation

Some people might want to graph kubelet counts grouped by kubelet_version or os_image, e.g. to track a Kubernetes upgrade over time.

Review checklist (to be filled by reviewers)

• PR title must be written as a CHANGELOG entry (see why)
• Files changes must correspond to the primary purpose of the PR as described in the title (small unrelated changes should have their own PR)
• PR must have changelog/ and integration/ labels attached
• Feature or bugfix must have tests
• Git history must be clean
• If PR adds a configuration option, it must be added to the configuration file.

[therve]

Thanks for the patch, looks reasonable. The existing tests need to be fixed, and it'd be nice to assert the new metric as well. We need to update metadata.csv, too.

[codecov]

Codecov Report

Merging #4383 into master will increase coverage by 5.96%.
The diff coverage is 100%.

Impacted Files	Coverage Δ
...atadog_checks/kubernetes_state/kubernetes_state.py	76.92% <ø> (ø)
kubernetes_state/tests/test_kubernetes_state.py	95% <100%> (ø)
disk/datadog_checks/disk/__init__.py
...checks_dev/tests/plugin/test_environment_runner.py
datadog_checks_base/tests/test_similar.py
win32_event_log/tests/test_check.py
...service/datadog_checks/windows_service/__init__.py
..._checks_base/datadog_checks/base/checks/network.py
...v/tests/tooling/config_validator/test_validator.py
...adog_checks_base/datadog_checks/base/utils/date.py
... and 163 more

[codecov]

Codecov Report

Merging #4383 into master will increase coverage by 7.55%.
The diff coverage is 100%.

Impacted Files	Coverage Δ
...atadog_checks/kubernetes_state/kubernetes_state.py	79.02% <ø> (ø)
kubernetes_state/tests/test_kubernetes_state.py	95.94% <100%> (ø)
windows_service/tests/common.py
disk/datadog_checks/disk/__init__.py
...checks_dev/tests/plugin/test_environment_runner.py
datadog_checks_base/tests/test_similar.py
win32_event_log/tests/test_check.py
...service/datadog_checks/windows_service/__init__.py
..._checks_base/datadog_checks/base/checks/network.py
...v/tests/tooling/config_validator/test_validator.py
... and 154 more

[therc]

I knew there was more. I fixed metadata (collecting the metric once a minute is plenty) and the tests. I didn't clean up unusual characters in the tags (spaces, colons, slashes), e.g. in os_image and container_runtime_version: should it be done?

[hkaj]

Thanks @therc!

The interval in metadata.csv doesn't impact the collection freq, it's used for in-app interpolation between the counter and rate types. 60s is fine there, but the default (empty value) is fine, too.

As for unusual characters, this is handled automatically. Characters that are not valid in tag values are replaced with _ (eg: os_image:container-optimized_os_from_google).

cc @JulienBalestra until the feature we talked about lands I think you will find this useful.

[therc]

Thanks, @hkaj. I saw that, as far as the tests are concerned, the invalid characters are preserved, but it makes sense for them to be sanitised downstream (in the statsd client, dd-agent or even on the server side, perhaps all).

[therc]

@therve is this good to go?

changes were made