-
Notifications
You must be signed in to change notification settings - Fork 86
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
RulesGroup doesn't follow Prometheus spec #932
Comments
This means we cannot copy rules 1:1 into the ClusterRules CRD. |
Hi, thanks for reporting! This is intentional as there is no global interval setting for rule evaluations (aka global.evaluation_interval), because that option only exists in Prometheus. In GMP it's a separate component that evaluates rules.
That's true, this means you have to manually/via sed/script add interval explicitly. Just to understand the consequences of this decision -- this is one-off annoyance though, right? Or is it causing more problems for you? |
This is mostly a one-off annoyance, but causes confusion as we cannot point people directly to the Prometheus docs. Given all the components are managed by GCP, it feels like it should just use the value of |
Got it, thanks. I assume you can point people directly to the Prometheus docs, but you have to mention the nuance to specify that one field explicitly, right? The point is that evaluation interval is also useful to specify explicit. It matters, depending on the rule. Do you think a global ruleEvaluator flag (e.g. in OperatorConfig) would be good enough here?
There is no such setting in AM, probably you mean Prometheus - which is not relevant in our stack as we don't use Prometheus for rule evals? Default to 1m is an option, yes, maybe not too bad. |
That's exactly what we do, but not everyone reads the nuance, especially if they have used prometheus/alertmanager before.
Ah got it.
This would be my preference, as it's the Prom default |
Thanks. Moving this to prioritization queue (OSS help wanted too). Acceptance Criteria
|
According to the docs:
However in Prometheus
rule_group
's,interval
is optional. For theRuleGroup
,interval
is required, which is different from what's expected.The text was updated successfully, but these errors were encountered: