Usage report breakdown by user does not properly group users

[choldgraf]

Bug description

The Grafana report that breaks down RAM by user does not properly group all of the users.

We think that this is because of the following dashboard code:

grafana-dashboards/dashboards/usage-report.jsonnet

Lines 35 to 53 in 45fd723

	// Computes sum of pod memory requests, grouped by username, for notebook pods
	prometheus.target(
	|||
	kube_pod_labels{
	label_app="jupyterhub",
	label_component="singleuser-server",
	namespace=~"$hub"
	}
	* on (namespace, pod) group_left()
	sum(
	container_memory_working_set_bytes{
	namespace=~"$hub",
	container="notebook",
	hub_jupyter_org_node_purpose="user",
	name!="",
	}
	) by (namespace, pod)
	|||,
	legendFormat='{{label_hub_jupyter_org_username}} ({{namespace}})',

From a conversation @GeorgianaElena noted:

Notice the group_left in the query which generated some duplicated entries in the graph, even though we were up to get the sum over each pod.
I've changed the group_left to a group_right and that generated the graph I shared in the previous message which looked like it produced a single value for the mem used by each notebook pod.

Expected behaviour

There should be a single plot per user, not multiple plots per user.

Actual behaviour

How to reproduce

• Go to the Usage report dashboard in the Grafana
• This should show multiple plots per user

[welcome]

Thank you for opening your first issue in this project! Engagement like this is essential for open source projects! 🤗

If you haven't done so already, check out Jupyter's Code of Conduct. Also, please try to follow the issue template as it helps other other community members to contribute more effectively.

You can meet the other Jovyans by joining our Discourse forum. There is also an intro thread there where you can stop by and say Hi! 👋

Welcome to the Jupyter community! 🎉