Add support for configurable modes for cloudwatch agent CR to support Prometheus monitoring with Target Allocator #86
Conversation
|
Helm chart integ tests should still pass right? Is the agent running as expected? |
mitali-salvi
changed the title
charts/amazon-cloudwatch-observability/templates/target-allocator-clusterrole.yaml
Outdated
Show resolved
Hide resolved
charts/amazon-cloudwatch-observability/templates/target-allocator-clusterrolebinding.yaml
Outdated
Show resolved
Hide resolved
mitali-salvi
force-pushed
the
mode-support
branch
from
4059532
to
4937c05
Compare
charts/amazon-cloudwatch-observability/templates/linux/cloudwatch-agent-custom-resource.yaml
Show resolved
Hide resolved
mitali-salvi
changed the title
| @@ -1,7 +1,7 @@ | |||
| build | |||
| .tmp | |||
| *.iml | |||
| .DS_Store | |||
| *.DS_Store | |||
| strategy: | ||
| nodeSelector: | ||
| kubernetes.io/os: linux | ||
| image: |
There was a problem hiding this comment.
Just noting that this image does not exist yet*
| @@ -545,8 +545,14 @@ admissionWebhooks: | |||
| ## Secret labels | |||
| secretLabels: { } | |||
|
|
|||
| ## Agent workloads list | |||
There was a problem hiding this comment.
| ## Agent workloads list | |
| ## List of AmazonCloudWatchAgent workloads to install & manage, each representing an independent installation of the AmazonCloudWatchAgent CustomResource. Each entry in this list uses the schema & the defaults from $.agent, so only provide any overrides here. |
| agent: | ||
| name: | ||
| mode: # represents the mode the cloudwatch-agent will run in (deployment, daemonset or statefulset) |
There was a problem hiding this comment.
Shouldnt this default to daemonset?
| agent: | ||
| name: | ||
| mode: # represents the mode the cloudwatch-agent will run in (deployment, daemonset or statefulset) | ||
| replicas: # The total number non-terminated pods targeted by this cloudwatch-agent's deployment or statefulSet. |
There was a problem hiding this comment.
Maybe set this to -1 as the default to make it obvious its not applicable by default and anything using it will fail unless overriden.
| @@ -602,6 +608,37 @@ agent: | |||
| } | |||
| } | |||
| } | |||
| windowsDefaultConfig: | |||
There was a problem hiding this comment.
This content actually now looks identical to the linux default above.
It might actually be cleaner to now have cloudwatch-agent-windows in the agents list above and drop this windowsDefaultConfig field here. We can instead add an os field thats either linux or windows and spin up all other resources based on that flag.
So agents would look like:
agents:
- name: cloudwatch-agent
- name: cloudwatch-agent-windows
os: windows
There was a problem hiding this comment.
Its fine if we do this in a separate PR.
| } | ||
| prometheus: | ||
| config: | ||
| targetAllocator: |
There was a problem hiding this comment.
i think we'll have more TLS related stuff here, but those can come slightly later
| @@ -51,3 +51,6 @@ rules: | |||
| - apiGroups: [ "route.openshift.io" ] | |||
| resources: [ "routes", "routes/custom-host" ] | |||
| verbs: [ "create","delete","get","list","patch","update","watch" ] | |||
| - apiGroups: [ "policy" ] | |||
| resources: [ "poddisruptionbudgets" ] | |||
There was a problem hiding this comment.
Is the latest operator creating PDBs for the deployment mode?
Description of changes:
Add support for prometheusCR monitoring
Testing - manually tested on cluster with the following config
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.