Sometimes workload is undefined in Notification Grouping #1681
Description
Describe the bug
Here is slack sink configuration:
- slack_sink:
name: main_slack_sink
slack_channel: alerts-xxxx
api_key: "{{ env.MAIN_SLACK_SINK }}"
grouping:
group_by:
- workload
- severity
interval: 86400
notification_mode:
summary:
threaded: true
by:
- workload
- severity
Nofifications are sent but if pod was killed by OOM or crush reason is CrashLoopBackOff. worload is undefined.
Matching criteria: severity: HIGH, workload: (undefined)
To Reproduce
See above
Expected behavior
As far as I understand the worload name is the name of a parent object ["Deployment", "ReplicaSet", "DaemonSet", "StatefulSet", "Pod", "Job"]:
event.operation in [K8sOperationType.CREATE, K8sOperationType.UPDATE]
and event.obj.kind in ["Deployment", "ReplicaSet", "DaemonSet", "StatefulSet", "Pod", "Job"]
I expect the name of the parent object has to be assigned to the workload name.
Screenshots
Desktop (please complete the following information):
- OS: Linux
- Browser: Slack client
- Version [e.g. 22]
Smartphone (please complete the following information):
Not applicable.
Additional context
Add any other context about the problem here.