Hello hello,

First of all: thank you for this helm chart !

Is your feature request related to a problem? Please describe.
Our proxy was not working because we used the last version of the react-client-sdk which requires version 7.0 of the relay proxy as mentionned here.

Describe the solution you'd like
Either update the version of the relay proxy to match up-to-date documentation.

Describe alternatives you've considered
Update README.md to emphasis which version of SDK (Client side and Server side) can be used when using this helm chart.

Additional context
Add any other context about the feature request here.

Is your feature request related to a problem? Please describe.
Currently, the helm chart supports envfrom the configMapRef but this does not support secure string secrets (like sdk keys) as it would require them to be in plain text in the values files.

Describe the solution you'd like
Allowing dynamic envFrom pre-existing k8s secrets. This would allow users to create k8s secrets that hold a bunch of key-value pairs (like the external secrets operator does via aws ssm) and the helm chart would create the environment variables from this secret. Removing the need for having secrets in the plain text of the values files. This could be a boolean toggle in the values files.

For now, we can leave the onus of creating/maintaining the k8s secret to the individual user. This way they can use any tool they would like to create the secret, as long as the secret name is <helm release name>-secret-environment-variables it will be picked up and used here.

Is your feature request related to a problem? Please describe.
We're using EKS and we're now running k8s in version 1.22, which doesn't support autoscaling/v2 API for HPA

Describe the solution you'd like
Have a check to use a compatible version of HPA based on k8s version.

Describe alternatives you've considered
We tried overriding the chart locally, but it created us issues with our argocd integration

Additional context
Add any other context about the feature request here.

Is your feature request related to a problem? Please describe.

priorityClassName is not assignable to the LD relay pod.

Describe the solution you'd like

Let priorityClassName be passed into the deployment int he same way as other pod attributes, like annotations or labels.

Describe alternatives you've considered

No real alternatives considered - this should be a pretty minor addition.

Additional context

Pull Request here.

Is your feature request related to a problem? Please describe.
It's not possible to add lifecycle hooks for relay container.

With some ingress controllers e.g. Azure application gateway ingress controller, the backend pool refreshes too slowly and causes downtime without a preStop hook. Documentation

Describe the solution you'd like
An ability to add lifecycle hooks via helm values

Describe alternatives you've considered
No reasonable alternatives afaik as this would be a pretty minor feature.

Additional context
Add any other context about the feature request here.

Describe the bug
Readiness and liveness fails when enabling TLS termination on ld-proxy.
Pods restarts because of Liveness failure

To reproduce

• Create a tls secret named "tls" with a certificate and key.
• Create values.yaml with below to enable TLS.

  relay:
    environment:
      TLS_ENABLED: "true"
      TLS_CERT: "/mnt/secrets/tls.crt"
      TLS_KEY: "/mnt/secrets/tls.key"
    secrets:
    - volumePath: tls.key
      volumeName: tlskey
      secretName: tls
      secretKey: tls.key
    - volumePath: tls.crt
      volumeName: tlscrt
      secretName: tls
      secretKey: tls.crt

• Use below helm command to install the chart

helm install relay --set relay.environment.LD_ENV_YourEnvironment=your-sdk-key -f values.yaml launchdarkly-ld-relay/ld-relay

Expected behavior
The Liveness and Readiness check should be using scheme HTTPS instead of default HTTP when TLS termination is enabled on the ld-proxy side.

        livenessProbe:
          failureThreshold: 3
          httpGet:
            path: /status
            port: api
            scheme: HTTP <<---- in chart, there is no scheme set, so it sets to default HTTP instead of HTTPS
          periodSeconds: 10
          successThreshold: 1
          timeoutSeconds: 1
        readinessProbe:
          failureThreshold: 3
          httpGet:
            path: /status
            port: api
            scheme: HTTP <<---- in chart, there is no scheme set, so it sets to default HTTP instead of HTTPS
          periodSeconds: 10
          successThreshold: 1
          timeoutSeconds: 1

Logs

ld-proxy kubectl pod describe output

  Warning  Unhealthy  9s (x10 over 47s)  kubelet            Readiness probe failed: HTTP probe failed with statuscode: 400
  Warning  Unhealthy  9s (x4 over 39s)   kubelet            Liveness probe failed: HTTP probe failed with statuscode: 400

Helm chart version

NAME                          	CHART VERSION	APP VERSION	DESCRIPTION
launchdarkly-ld-relay/ld-relay	2.2.1        	7.2.1      	LaunchDarkly Relay Proxy

Relay proxy configuration, developer tools
Helm : v3.11.2
LD Version : launchdarkly/ld-relay:7.2.1

Additional context
This issue surfaced after fixing the issue #36

Is your feature request related to a problem? Please describe.
Currently there is no way to configure an AWS ELB when deploying this helm chart. However this is possible via annotations. This would be very helpful to allow deployment of load balancers with their full configuration via this helm chart.

Describe the solution you'd like
Allow annotations to be made to the service like what is possible for ingress and pods.

Is your feature request related to a problem? Please describe.
I want to use an environment variable with valueFrom.fieldRef.fieldPath. Because you are passing it to the config map, this is not an option at the moment.

Describe the solution you'd like
Have another value that renders to the containers environment variable

Describe alternatives you've considered
None right now

Is this a support request?
No

Describe the bug
Pods don't restart when configmap changes via .Values.relay.environment. This can be combatted with an annotation such as

spec:
  template:
    metadata:
      annotations:
        checksum/config: {{ include ("chart-name.configmap") . | sha256sum }}

To reproduce
Install the chart, change the environment variables via the .Values.relay.environment, and update the deployment. Notice that the pods don't automatically restart to account for the new environment

Expected behavior
Pods should restart

Logs
If applicable, add any log output related to your problem.

Helm chart version
1.0.0

Additional context
Add any other context about the problem here.

I'd be happy to make a PR if preffered?

Is your feature request related to a problem? Please describe.
No

Describe the solution you'd like
We should be able to add additional labels to the pods. For us this is required for log collections.

Describe alternatives you've considered
No alternatives, as this is a de-facto way.

Additional context
N/A

Is your feature request related to a problem? Please describe.
Currently there is no way to mount a secret into the pod, just environment variables. This means there is no way to pass in a TLS cert into the pod, something that is supported by the relay proxy.

Describe the solution you'd like
I want to be able to mount a secret inside the pod

Describe the bug
References to older chart versions no longer resolve

I noticed the index.yaml was recently completely re-built (deleted, then re-added), and older releases now have a new URL like:

https://launchdarkly.github.io/ld-relay-helm/helm-repo/ld-relay-3.1.0.tgz

note the /helm-repo/ component, which doesn't seem to exist in the repo itself. These URLs 404 when fetched

but the newest release 3.3.0 has a URL like:

https://launchdarkly.github.io/ld-relay-helm/ld-relay-3.3.0.tgz

which works.

To reproduce

# Chart.yaml
---
apiVersion: v2
name: launchdarkly-relay-proxy
description: |
  The LaunchDarkly Relay Proxy is a small Go application that runs on your own infrastructure. It connects to the
  LaunchDarkly streaming API and proxies that connection to clients within your organization's network.
type: application
version: 1.0.0
appVersion: '1.0.0'
dependencies:
  - name: ld-relay
    version: 3.1.0
    repository: https://launchdarkly.github.io/ld-relay-helm

$ helm repo add launch-darkly-relay https://launchdarkly.github.io/ld-relay-helm
...

$ helm dependency build .                                                       
Hang tight while we grab the latest from your chart repositories...
...Successfully got an update from the "launch-darkly-relay" chart repository
Update Complete. ⎈Happy Helming!⎈
Saving 1 charts
Downloading ld-relay from repo https://launchdarkly.github.io/ld-relay-helm
Save error occurred:  could not download https://launchdarkly.github.io/ld-relay-helm/helm-repo/ld-relay-3.1.0.tgz: failed to fetch https://launchdarkly.github.io/ld-relay-helm/helm-repo/ld-relay-3.1.0.tgz : 404 Not Found
Error: could not download https://launchdarkly.github.io/ld-relay-helm/helm-repo/ld-relay-3.1.0.tgz: failed to fetch https://launchdarkly.github.io/ld-relay-helm/helm-repo/ld-relay-3.1.0.tgz : 404 Not Found

Expected behavior
Expected old chart versions to be able to be resolved when listed as a dependency

Logs
n/a

Helm chart version
I've verified that all versions < 3.3.0 have the broken URL (/helm-repo/ component)

Relay proxy configuration, developer tools
n/a

Additional context
n/a

Is your feature request related to a problem? Please describe.
With v8.0.0 being released 2 weeks ago we would like to update the proxy chart accordingly. Preventing us from upgrading our PHP SDK to v6.0

Describe the solution you'd like
Update the version of the relay proxy to match up-to-date documentation.

Describe alternatives you've considered
Manually bumping the image version on the helm chart on our side but would like to be able to use the stock helm chart.

Additional context
Add any other context about the feature request here.

Describe the bug
LD relay is a critical component in our stack and we cannot allow it to be down even if there is a zone failure. We would like to define something like this in a deployment:

      topologySpreadConstraints:
      - labelSelector:
          matchLabels:
          {{- include "launchdarkly-relay.selectorLabels" . | nindent 12 }}
        maxSkew: 1
        topologyKey: topology.kubernetes.io/zone
        whenUnsatisfiable: ScheduleAnyway

Describe the bug
Adding more than one secrets to mount certificate and keys causes below error during helm installation

Deployment.apps "relay" is invalid: spec.template.spec.containers[0].volumeMounts[1].mountPath: Invalid value: "/mnt/secrets/": must be unique

To reproduce
Create values.yaml with below

  relay:
    environment:
      TLS_ENABLED: "true"
      TLS_CERT: "/mnt/secrets/tls.crt"
      TLS_KEY: "/mnt/secrets/tls.key"
    secrets:
    - volumePath: tls.key
      volumeName: tlskey
      secretName: tls
      secretKey: tls.key
    - volumePath: tls.crt
      volumeName: tlscrt
      secretName: tls
      secretKey: tls.crt

Use below helm template command

helm install relay --set relay.environment.LD_ENV_YourEnvironment=your-sdk-key -f values.yaml launchdarkly-ld-relay/ld-relay

Expected behavior
The generated mountPath should be uniq

Logs
If applicable, add any log output related to your problem.

Helm chart version

NAME                          	CHART VERSION	APP VERSION	DESCRIPTION
launchdarkly-ld-relay/ld-relay	2.2.0        	7.2.1      	LaunchDarkly Relay Proxy

Relay proxy configuration, developer tools

Helm : v3.11.2
LD Version : launchdarkly/ld-relay:7.2.1

Describe the bug
LD relay is a critical component in our stack and we cannot allow it to be down even when rolling nodes.