From 4ab5bf36e4f878791ae8dd8c7689d0ea3d56f07c Mon Sep 17 00:00:00 2001
From: Sheogorath <sheogorath@shivering-isles.com>
Date: Mon, 30 Oct 2023 03:01:37 +0100
Subject: [PATCH] feat(immich): Add SLO for Immich
---
apps/k8s01/immich/kustomization.yaml | 1 +
apps/k8s01/immich/slo.yaml | 41 ++++++++++++++++++++++++++++
2 files changed, 42 insertions(+)
create mode 100644 apps/k8s01/immich/slo.yaml
diff --git a/apps/k8s01/immich/kustomization.yaml b/apps/k8s01/immich/kustomization.yaml
index 0ce62fbdf..9b88b77a5 100644
--- a/apps/k8s01/immich/kustomization.yaml
+++ b/apps/k8s01/immich/kustomization.yaml
@@ -6,6 +6,7 @@ resources:
- immich-values.yaml
- certificate.yaml
- limitrange.yaml
+ - slo.yaml
- ../../../shared/resourcequotas/default.yaml
patchesStrategicMerge:
- database-override.yaml
diff --git a/apps/k8s01/immich/slo.yaml b/apps/k8s01/immich/slo.yaml
new file mode 100644
index 000000000..8e10c7dcf
--- /dev/null
+++ b/apps/k8s01/immich/slo.yaml
@@ -0,0 +1,41 @@
+apiVersion: sloth.slok.dev/v1
+kind: PrometheusServiceLevel
+metadata:
+ name: requests-immich
+ namespace: immich
+spec:
+ service: "immich"
+ slos:
+ - name: "requests-availability"
+ objective: 98
+ description: "Immich: SLO based on availability for HTTP request responses."
+ sli:
+ events:
+ errorQuery: sum(rate(nginx_ingress_controller_requests{exported_namespace="immich",ingress="immich",status=~"(5..|429)"}[{{.window}}]))
+ totalQuery: sum(rate(nginx_ingress_controller_requests{exported_namespace="immich",ingress="immich"}[{{.window}}])) > 0 OR vector(1)
+ alerting:
+ name: ImmichHighErrorRate
+ labels:
+ category: "availability"
+ annotations:
+ summary: "High error rate on 'immich' requests responses"
+ - name: "requests-latency"
+ objective: 95
+ description: "Immich: SLO based on latency for HTTP request responses. Warns if requests take longer than 250ms. When responses are slower than 200ms they become noticable slow."
+ labels:
+ category: latency
+ sli:
+ events:
+ errorQuery: |
+ (
+ sum(rate(nginx_ingress_controller_request_duration_seconds_count{exported_namespace="immich",ingress="immich",method!="WATCH"}[{{.window}}]))
+ -
+ sum(rate(nginx_ingress_controller_request_duration_seconds_bucket{exported_namespace="immich",ingress="immich",le="0.25",verb!="WATCH"}[{{.window}}]))
+ )
+ totalQuery: sum(rate(nginx_ingress_controller_request_duration_seconds_count{exported_namespace="immich",ingress="immich",method!="WATCH"}[{{.window}}])) > 0 OR vector(1)
+ alerting:
+ name: ImmichLatencyAlert
+ labels:
+ category: "latency"
+ annotations:
+ summary: "Slow responses on 'immich' requests responses. More than 1% take more than 250ms."
\ No newline at end of file
--
GitLab