diff --git a/Documentation/etcd-mixin/mixin.libsonnet b/Documentation/etcd-mixin/mixin.libsonnet index 13e0dcfa4..b6cc471ff 100644 --- a/Documentation/etcd-mixin/mixin.libsonnet +++ b/Documentation/etcd-mixin/mixin.libsonnet @@ -151,7 +151,7 @@ { alert: 'etcdHighNumberOfFailedHTTPRequests', expr: ||| - sum(rate(etcd_http_failed_total{%(etcd_selector)s}[5m])) BY (method) / sum(rate(etcd_http_received_total{%(etcd_selector)s}[5m])) + sum(rate(etcd_http_failed_total{%(etcd_selector)s, code!="404"}[5m])) BY (method) / sum(rate(etcd_http_received_total{%(etcd_selector)s}[5m])) BY (method) > 0.01 ||| % $._config, 'for': '10m', @@ -165,7 +165,7 @@ { alert: 'etcdHighNumberOfFailedHTTPRequests', expr: ||| - sum(rate(etcd_http_failed_total{%(etcd_selector)s}[5m])) BY (method) / sum(rate(etcd_http_received_total{%(etcd_selector)s}[5m])) + sum(rate(etcd_http_failed_total{%(etcd_selector)s, code!="404"}[5m])) BY (method) / sum(rate(etcd_http_received_total{%(etcd_selector)s}[5m])) BY (method) > 0.05 ||| % $._config, 'for': '10m', diff --git a/Documentation/op-guide/etcd3_alert.rules.yml b/Documentation/op-guide/etcd3_alert.rules.yml index deaa89b6f..17287172b 100644 --- a/Documentation/op-guide/etcd3_alert.rules.yml +++ b/Documentation/op-guide/etcd3_alert.rules.yml @@ -107,7 +107,7 @@ groups: message: '{{ $value }}% of requests for {{ $labels.method }} failed on etcd instance {{ $labels.instance }}' expr: | - sum(rate(etcd_http_failed_total{job=~".*etcd.*"}[5m])) BY (method) / sum(rate(etcd_http_received_total{job=~".*etcd.*"}[5m])) + sum(rate(etcd_http_failed_total{job=~".*etcd.*", code!="404"}[5m])) BY (method) / sum(rate(etcd_http_received_total{job=~".*etcd.*"}[5m])) BY (method) > 0.01 for: 10m labels: @@ -117,7 +117,7 @@ groups: message: '{{ $value }}% of requests for {{ $labels.method }} failed on etcd instance {{ $labels.instance }}.' expr: | - sum(rate(etcd_http_failed_total{job=~".*etcd.*"}[5m])) BY (method) / sum(rate(etcd_http_received_total{job=~".*etcd.*"}[5m])) + sum(rate(etcd_http_failed_total{job=~".*etcd.*", code!="404"}[5m])) BY (method) / sum(rate(etcd_http_received_total{job=~".*etcd.*"}[5m])) BY (method) > 0.05 for: 10m labels: