I deployed the prometheus-community/helm-charts/prometheus Helm chart in agent mode to my Kubernetes cluster. However, the prometheus-server pod is stuck in "CrashLoopBackOff" with the following error: "Error loading config (--config.file=/etc/config/prometheus.yml)" file=/etc/config/prometheus.yml err="field rule_files is not allowed in agent mode".
To work around this issue, I removed the "rule_files" field from the prometheus-server ConfigMap, and the pod started successfully. However, I'd like to configure the Helm chart in a way that avoids having to remove this field manually.
How can I set the Helm values or configuration to ensure that the "rule_files" field is not present from the beginning when deploying the Prometheus Helm chart in agent mode?
Here's my current Helm values.yaml file:
alertmanager:
enabled: false
prometheus-pushgateway:
enabled: false
server:
configPath: /etc/config/prometheus.yml
defaultFlagsOverride:
- --enable-feature=agent
- --config.file=/etc/config/prometheus.yml
extraFlags:
- web.enable-lifecycle
global:
evaluation_interval: 30s
scrape_interval: 30s
scrape_timeout: 10s
remoteWrite:
- basic_auth:
password: xxxx
username: yyyy
url: https://my-external-prometheus.com/api/v1/write
I fixed this by adding the following to my values, this ensures that no rule_files are processed.
prometheus:
serverFiles:
prometheus.yml:
rule_files: []