monitor kestra with prometheus alerts
This commit is contained in:
@@ -168,6 +168,33 @@ in {
|
||||
labels.severity = "warning";
|
||||
annotations.description = "One or more systemd units are failed on the host.";
|
||||
})
|
||||
(mkAlertRule {
|
||||
uid = "az_kestra_down";
|
||||
title = "Kestra metrics unreachable";
|
||||
expr = ''up{job="kestra"}'';
|
||||
for = "2m";
|
||||
evaluatorType = "lt";
|
||||
evaluatorParams = [1 0];
|
||||
labels = {
|
||||
service = "kestra";
|
||||
severity = "critical";
|
||||
};
|
||||
annotations.description = "Prometheus cannot scrape Kestra metrics.";
|
||||
})
|
||||
(mkAlertRule {
|
||||
uid = "az_kestra_execution_failed";
|
||||
title = "Kestra execution failed";
|
||||
expr = ''sum by(namespace_id, flow_id, state) (increase(kestra_executor_execution_end_count_total{job="kestra",state=~"FAILED|WARNING|KILLED"}[5m]))'';
|
||||
for = "1m";
|
||||
noDataState = "OK";
|
||||
evaluatorType = "gt";
|
||||
evaluatorParams = [0 0];
|
||||
labels = {
|
||||
service = "kestra";
|
||||
severity = "critical";
|
||||
};
|
||||
annotations.description = "A Kestra flow execution ended with FAILED, WARNING, or KILLED.";
|
||||
})
|
||||
(mkAlertRule {
|
||||
uid = "az_http_probe_failed";
|
||||
title = "HTTP probe failed";
|
||||
|
||||
Reference in New Issue
Block a user