This repo contains a simple extension for argos - and so therefore perhaps xbar* - to display active alerts for a configured set of OpenShift clusters.
* not tested
Fill in the cluster details in clusters.yaml and copy to the config directory:
mkdir ~/.config/ocp-alerts
cp clusters.yaml ~/.config/ocp-alerts/
The syntax for the config is as follows:
clusters:
cluster0:
url: https://alertmanager-main-openshift-monitoring.apps.cloud.example.com
token: eyJhbGciOiJSUzI1NiIsImtpZCI6IlRkUnBKcmktV1ZlbkZJOVZCOTNYWlNFcEZrWGI2
severity: ["warning", "critical", "info"]
cluster1:
url: https://alertmanager-main-openshift-monitoring.apps.lab.example.com
token: eyJhbGciOiJSUzI1NiIsImtpZCI6InFOYk0wNHA0YWdzZ3dJRGhNZ0xUWExIY0EtYlV
severity: ["warning", "critical"]
cluster2:
url: https://alertmanager-main-openshift-monitoring.apps.qa.example.com
token: eyJhbGciOiJSUzI1NiIsImtpZCI6ImY5bkRfN3FIQWw5Y3VUalhtd0Rscy1PS2NnWH
severity: ["warning", "critical", "info"]
The url can be determined with:
oc get routes -n openshift-monitoring prometheus-k8s -o json | jq -r .spec.host
Any valid token with sufficient privileges will do, such as:
oc get secret -n openshift-monitoring $(oc get secret -n openshift-monitoring | grep prometheus-k8s-token | awk '{print $1}') -o json | jq -r '.data.token|@base64d'
Copy ocp_alerts.py to ~/.config/argos and rename it to reflect desired update interval in seconds. For example, to run the script and update the status every 15 minutes, the script should be ~/.config/argos/ocp_alerts.1500s+.py.
Or you could just symlink it:
ln -s ~/code/path/to/repo/ocp_alerts.py ~/.config/argos/ocp_alerts.1500s+.py
This is how it should look:
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent