-
Notifications
You must be signed in to change notification settings - Fork 717
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Rebase Monitoring to 14.5.0 and update all dependencies #1072
Rebase Monitoring to 14.5.0 and update all dependencies #1072
Conversation
2a28654
to
a4b3b80
Compare
a4b3b80
to
060d3c7
Compare
756f66a
to
1f30dbf
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Since this PR is huge and not reviewable in its original form, I have added 3 commits to make this process way easier to review:
Third To Last Commit
I ran:
a) PACKAGE=rancher-monitoring make charts
without any changes
b) mv charts/rancher-monitoring/rancher-monitoring/<VERSION> charts/rancher-monitoring/rancher-monitoring/charts
c) mv charts/rancher-monitoring/rancher-monitoring-crd/<VERSION> charts/rancher-monitoring/rancher-monitoring-crd/charts
This represents the state of the chart today.
Second To Last Commit
Same steps as above once my changes were made.
This represents the chart that will be shipped in this commit.
Reviewers should review this commit for changes since this contains the final diff that will be shipped from this PR.
Last Commit
I removed the charts/
directory since we do not want to commit that.
Modifications
If any modifications are requested, I will add further commits to this PR instead of modifying these charts commits.
At the end of the PR, all of this will be squash and merged.
@@ -2486,18 +2771,6 @@ prometheus: | |||
runAsUser: 101 | |||
runAsGroup: 101 | |||
|
|||
# Additional volumes on the output StatefulSet definition. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I moved this back to the original location, shown above
@@ -7,4 +7,4 @@ apiVersion: v1 | |||
description: Installs the CRDs for rancher-monitoring. | |||
name: rancher-monitoring-crd | |||
type: application | |||
version: 9.4.204-rc09 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Old chart versions are dropped in #1075
@@ -14,8 +14,63 @@ annotations: | |||
catalog.cattle.io/requests-cpu: 4500m | |||
catalog.cattle.io/requests-memory: 4000Mi | |||
catalog.cattle.io/ui-component: monitoring | |||
apiVersion: v1 | |||
appVersion: 0.38.1 | |||
apiVersion: v2 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Since we've moved to apiVersion v2, the dependencies have all been moved to the Chart.yaml
catalog.cattle.io/hidden: "true" | ||
catalog.cattle.io/os: linux | ||
catalog.rancher.io/certified: rancher | ||
catalog.rancher.io/namespace: cattle-monitoring-system | ||
catalog.rancher.io/release-name: rancher-grafana |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These annotations were added since all dependent charts that track an upstream were split off into top-level charts. This chart needs to be hidden and is only supported by Linux.
replicas: 1 | ||
image: | ||
# image-renderer Image repository | ||
repository: rancher/grafana-grafana-image-renderer |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
New image repository required and will be added in rancher/image-mirror#83
annotations: | ||
catalog.cattle.io/hidden: "true" | ||
catalog.cattle.io/os: linux | ||
catalog.rancher.io/certified: rancher | ||
catalog.rancher.io/namespace: cattle-monitoring-system | ||
catalog.rancher.io/release-name: rancher-kube-state-metrics |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These annotations were added since all dependent charts that track an upstream were split off into top-level charts. This chart needs to be hidden and is only supported by Linux.
annotations: | ||
catalog.cattle.io/hidden: "true" | ||
catalog.cattle.io/os: linux | ||
catalog.rancher.io/certified: rancher | ||
catalog.rancher.io/namespace: cattle-monitoring-system | ||
catalog.rancher.io/release-name: rancher-prometheus-adapter |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These annotations were added since all dependent charts that track an upstream were split off into top-level charts. This chart needs to be hidden and is only supported by Linux.
annotations: | ||
catalog.cattle.io/hidden: "true" | ||
catalog.cattle.io/os: linux | ||
catalog.rancher.io/certified: rancher | ||
catalog.rancher.io/namespace: cattle-monitoring-system | ||
catalog.rancher.io/release-name: rancher-node-exporter |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These annotations were added since all dependent charts that track an upstream were split off into top-level charts. This chart needs to be hidden and is only supported by Linux.
@@ -1,145 +0,0 @@ | |||
dependencies: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This was completely removed since we moved to apiVersion v2 charts
bc6a1db
to
3cbb181
Compare
Tested deploying in a hardened cluster, a windows cluster, and a normal cluster and successfully deployed all components. All images and repositories have been created as well. |
Signed-off-by: Arvind Iyengar <arvind.iyengar@rancher.com>
Script has been moved to #1079. |
3cbb181
to
ce35c7b
Compare
@aiyengar2 In upgrade case, doe the deleted CRDs and the CRs get cleaned up? I see a couple of them deleted by the new version? |
I believe those should be cleaned up by the CRD chart upgrade |
Also, I don't think the CRDs are being deleted, but rather renamed into different files that also have modifications. That's why they appear to show up as deletions |
packages/rancher-grafana/generated-changes/exclude/ci/with-dashboard-json-values.yaml
Show resolved
Hide resolved
Had to push another commit to patch the CRD version to 14.5.1 (for 14.5.100 to be exposed) as well |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
this was a huge rebase, well done!!! as far as I could tell everything looks good. You answered all of my questions and it looks like images, versions, and other port or chart.yaml changes were consistent and expected!
* Switch rancher-monitoring to Git-based package Signed-off-by: Arvind Iyengar <arvind.iyengar@rancher.com> * Move Monitoring dependencies to separate packages * Switch rancher-grafana to Git-based and add annotations * Switch rancher-kube-state-metrics to Git-based and add annotations * Switch rancher-prometheus-adapter to Git-based and add annotations * Switch rancher-node-exporter to Git-based and add annotations * Rebase to kube-prometheus-stack-10.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-10.0.0 * Rebase to kube-prometheus-stack-11.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-11.0.0 * Rebase to kube-prometheus-stack-12.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-12.0.0 * Rebase to kube-prometheus-stack-13.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-13.0.0 * Rebase to kube-prometheus-stack-14.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-14.0.0 * Rebase to kube-prometheus-stack-14.5.0 * Update rancher-monitoring to new base kube-prometheus-stack-14.5.0 * fixup rancher-monitoring * Switch nginx port for Prometheus to 8081 This is necessary since prometheusConfigReloaderImage now takes port 8080 in the Prometheus pod due to a change introduced in Prometheus Operator v0.43.0 in prometheus-operator/prometheus-operator@dcf34c5. This value is hard-coded in the operator in https://github.com/prometheus-operator/prometheus-operator/blob/bcf61aacda782d6c8ffc79f55ea1def9cfd97933/pkg/operator/statefulset.go#L25. * Rebase to grafana-5.7.7 * Update rancher-grafana to new base grafana-5.7.7 * Rebase to grafana-6.0.0 * Update rancher-grafana to new base grafana-6.0.0 * Rebase to grafana-6.6.4 * Update rancher-grafana to new base grafana-6.6.4 * fixup rancher-grafana * Add linux annotations to image-renderer * Use default grafana serviceAccount if unavailable for image-renderer * Rebase to kube-state-metrics-helm-chart-2.13.1 * Update rancher-kube-state-metrics to new base kube-state-metrics-helm-chart-2.13.1 * fixup kube-state-metrics * Rebase to prometheus-node-exporter-1.16.2 * Update rancher-node-exporter to new base prometheus-node-exporter-1.16.2 * fixup node-exporter * Rebase to prometheus-adapter-2.12.1 * Update rancher-prometheus-adapter to new base prometheus-adapter-2.12.1 * fixup prometheus-adapter * Update CRD chart version * Add original chart before all commits were made * Add chart after all commits were made * Delete charts * Patch the chart version to 1 and reset packageVersion to 00 * Patch CRD chart version
* Switch rancher-monitoring to Git-based package Signed-off-by: Arvind Iyengar <arvind.iyengar@rancher.com> * Move Monitoring dependencies to separate packages * Switch rancher-grafana to Git-based and add annotations * Switch rancher-kube-state-metrics to Git-based and add annotations * Switch rancher-prometheus-adapter to Git-based and add annotations * Switch rancher-node-exporter to Git-based and add annotations * Rebase to kube-prometheus-stack-10.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-10.0.0 * Rebase to kube-prometheus-stack-11.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-11.0.0 * Rebase to kube-prometheus-stack-12.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-12.0.0 * Rebase to kube-prometheus-stack-13.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-13.0.0 * Rebase to kube-prometheus-stack-14.0.0 * Update rancher-monitoring to new base kube-prometheus-stack-14.0.0 * Rebase to kube-prometheus-stack-14.5.0 * Update rancher-monitoring to new base kube-prometheus-stack-14.5.0 * fixup rancher-monitoring * Switch nginx port for Prometheus to 8081 This is necessary since prometheusConfigReloaderImage now takes port 8080 in the Prometheus pod due to a change introduced in Prometheus Operator v0.43.0 in prometheus-operator/prometheus-operator@dcf34c5. This value is hard-coded in the operator in https://github.com/prometheus-operator/prometheus-operator/blob/bcf61aacda782d6c8ffc79f55ea1def9cfd97933/pkg/operator/statefulset.go#L25. * Rebase to grafana-5.7.7 * Update rancher-grafana to new base grafana-5.7.7 * Rebase to grafana-6.0.0 * Update rancher-grafana to new base grafana-6.0.0 * Rebase to grafana-6.6.4 * Update rancher-grafana to new base grafana-6.6.4 * fixup rancher-grafana * Add linux annotations to image-renderer * Use default grafana serviceAccount if unavailable for image-renderer * Rebase to kube-state-metrics-helm-chart-2.13.1 * Update rancher-kube-state-metrics to new base kube-state-metrics-helm-chart-2.13.1 * fixup kube-state-metrics * Rebase to prometheus-node-exporter-1.16.2 * Update rancher-node-exporter to new base prometheus-node-exporter-1.16.2 * fixup node-exporter * Rebase to prometheus-adapter-2.12.1 * Update rancher-prometheus-adapter to new base prometheus-adapter-2.12.1 * fixup prometheus-adapter * Update CRD chart version * Add original chart before all commits were made * Add chart after all commits were made * Delete charts * Patch the chart version to 1 and reset packageVersion to 00 * Patch CRD chart version
First commit switches rancher-monitoring to a Git-based package for the next step.
Second commit creates a script that allows commit-by-commit interactive rebases on upstreams for Git-based packages.
Remaining commits were done via scripts in pairs:
Some fixups were added after the fact to each chart when I tested deploying them and found bugs from the rebase.
Related Issue: rancher/rancher#29871