[FEATURE REQUEST] Dedicated health check page

[luckydonald]

Description

Feature
Docker exposes the state of a few last healthchecks.
It would be great to have a page which shows them.

I think about a similar approach as travis-ci does:

So there would be an entry for every health check enabled container

Failing being red marked with a cross ❌
Successfull being green with a check mark ✅
Running being yellow with 2 ellipsis dots ••

Technical details:

• Portainer version: 1.15.2

[WTFKr0]

I like the idea, but I think it has to be more large than that
I think portainer is lacking an Ops view, which show :

• Status of services / containers
• HealthChecks if available
• Image and versions and last update date

In fact, a monitoring view of what i got on my docker / swarm and what is the status
A view that we can fullscreen will be awesome for Ops I think

[deviantony]

Agree with @WTFKr0 as well but this should be treated inside another issue.

I'll tag this one as an evolution to the actual healthcheck panels available in the container-details view.

[iotgopher]

Having what WTFKr0 describes would be very useful for devops
A view like that could be for a single endpoint or for all of them

[eine]

My two cent:

• I conceive the dashboard as the Ops view @WTFKr0 comments. Indeed, as a limited-feature grafana alike with (optional) tight integration with a sibling stack running prometheus and grafana. See Monitoring and alerting #943.
• When I read the first message by @luckydonald , what came to my mind was to replace the State label in the containers view with a btn-group composed of 3-4 buttons. This is more similar to travis, where checks for all the branches are shown in a single view. However, I don't know if this can be applied to the stacks view.

[luckydonald]

Currently the only way to see an overview of health checks are the container list.

However, it has no real way to filter them by health status, and all in all it's not easy to find your health check enabled containers between all the rest.

In non-swarm environments they don't automatically restart, so having a quick overview would be great.

Having a separate, dedicated healthcheck page would be great.
Each row would be a container.
Inside that container, there are the status of the last healthchecks (like travis: last builds)
This would be provided via inspect -> State.Health.Log

(Note the exit status changed from 0 to 1 in the last 2 checks)

---

The Dev Ops-Dashboard, especially the Image and versions and last update date would be a separate feature, however.
However, maybe a first step into that direction could be adding a unhealty/healty count to the main dashboard:

[luckydonald]

For reference, picture of branches in travis.

[eine]

@luckydonald, if the labels in the containers list view were replaced with four small boxes representing the four last healthchecks, and filtering out containers without healthcheck enabled was added, would you need a separate page/view? What additional info would that page include?

[luckydonald]

@1138-4eb Maybe it is about including less information.
A page where I can see if they are healthy or not.
Without all that unneeded information,
like image name, ports, etc.

Just the container name, health overview, and maybe even some quick action buttons, like restarting that container.

Unhealthy containers first, next the booting ones, and lastly the good ones.