Previously[1] we attempted to do the same, but there was a
misunderstanding about the GC behavior and it caused the alert to be
fired even before GC comes into play.

According to[2][3] kubelet GC kicks in only when `imageGCHighThresholdPercent` is hit which is set to 85% by default. However `NodeFilesystemSpaceFillingUp` is set to fire as soon as 80% usage is hit.

This commit changes the `fsSpaceFillingUpWarningThreshold` to 15% so
that we give ample time to GC to reclaim unwanted images. This commit
also changes `fsSpaceFillingUpCriticalThreshold` to 10% which gives more time to admins to react to warning before sending critical alert.

[1] https://github.com/prometheus-operator/kube-prometheus/pull/1357
[2] https://docs.openshift.com/container-platform/4.10/nodes/nodes/nodes-nodes-garbage-collection.html#nodes-nodes-garbage-collection-images_nodes-nodes-configuring
[3] https://kubernetes.io/docs/reference/config-api/kubelet-config.v1beta1/



Signed-off-by: Arunprasad Rajkumar <arajkuma@redhat.com>

Signed-off-by: ArthurSens <arthursens2005@gmail.com>

Signed-off-by: ArthurSens <arthursens2005@gmail.com>

Signed-off-by: ArthurSens <arthursens2005@gmail.com>

Signed-off-by: ArthurSens <arthursens2005@gmail.com>

Problem: Currently the prometheus-adapter pods are restarted at the same
time even though the deployment is configured with strategy RollingUpdate.
This happens because the kubelet does not know when the prometheus-adapter
pods are ready to start receiving requests.

Solution: Add both readinessProbe and livenessProbe to the
prometheus-adapter, this way the kubelet will know when either the pod
stoped working and should be restarted or simply when it ready to start
receiving requests.

Issue: https://bugzilla.redhat.com/show_bug.cgi?id=2048333

Signed-off-by: paulfantom <pawel@krupa.net.pl>
Signed-off-by: Paweł Krupa (paulfantom) <pawel@krupa.net.pl>
(cherry picked from commit d3ea3147a8fa643413fbcba368c0de88aec8f7eb)
(cherry picked from commit d24c347b2742d9474c8f441f2831262c63b8c79b)

Signed-off-by: GitHub <noreply@github.com>
(cherry picked from commit df68b8d1da5d2d91b9502d4be67063c2c497e0cb)

Signed-off-by: GitHub <noreply@github.com>
(cherry picked from commit 86e16b539cc57710b50f4692848cab5645e3d2bc)

(cherry picked from commit b4bf38ba6c0f4ad34bc080b0c655454c3ab1fbdb)
(cherry picked from commit c21bf4fbfa478fd163c091054d9fcc98836d7045)

(cherry picked from commit f8c00b9963cc63a3cf98dd1c825943d4df92d9c4)
(cherry picked from commit f09b8e5de2e46db85f090549d37eeb878a81842f)

Signed-off-by: ArthurSens <arthursens2005@gmail.com>

Signed-off-by: GitHub <noreply@github.com>

Signed-off-by: ArthurSens <arthursens2005@gmail.com>

This is a follow up fix of https://github.com/prometheus-operator/kube-prometheus/pull/1613

. @simonpasquier recommended to sanitize all denylist metrics.

Signed-off-by: Arunprasad Rajkumar <arajkuma@redhat.com>

The following metrics are missing from kube-state-metrics:
- kube_pod_container_status_terminated_reason
- kube_pod_init_container_status_terminated_reason
- kube_pod_status_scheduled_time

Previously, some metrics were removed from kube-state-metrics by adding the following --metric-denylist argument to the kube-state-metrics container

```
--metric-denylist=
kube_.+_created,
kube_.+_metadata_resource_version,
kube_replicaset_metadata_generation,
kube_replicaset_status_observed_generation,
kube_pod_restart_policy,
kube_pod_init_container_status_terminated,
kube_pod_init_container_status_running,
kube_pod_container_status_terminated,
kube_pod_container_status_running,
kube_pod_completion_time,
kube_pod_status_scheduled
```

--metric-denylist: Comma-separated list of metrics not to be enabled. This list comprises of exact metric names and/or regex patterns. The allowlist and denylist are mutually exclusive.

However, all the list of metrics is managed as RegEx, thus "kube_pod_container_status_terminated" denies .*kube_pod_container_status_terminated.*, that's why kube_pod_init_container_status_terminated_reason is missing

Co-authored-by: Florian Gleizes <fgleizes@redhat.com>
Signed-off-by: Arunprasad Rajkumar <arajkuma@redhat.com>

Signed-off-by: Paweł Krupa (paulfantom) <pawel@krupa.net.pl>

Signed-off-by: GitHub <noreply@github.com>