I am setting up an alert where CPU utilization of a container is below 0.1 AND CPU requested is above 1. The alert is triggered at the correct condition but the {$container} variable in the alert message shows all the container names even if one of the above conditions matches. I have taken screenshots of all the relevant pages in the dashboard.
For eg. in the following case:
rx cpu is not below 0.1 but it shows in the alert message shows both carbon, rx when captured by {$container} because rx’s CPU is greater than 1 (which is the 2nd condition). My requirement is to only get those container names that meet both conditions.
Please let me know what am I missing here. I am using VictoriaMetric as a datasource.
You are using old deprecated legacy alerting, which has some limitation. Switch to more complex unified alerting and you will be able to alert on particular container (but is more complex = more complicated, but also more powerful).
Is the new alerting system backward compatible with the older one? We have many alerts setup here.