Bug 2009352: IR-253: add prometheus rules for image registry operations metrics

also document new rules.

Author: flavianmissi

docs/README.md

@@ -5,3 +5,4 @@ These documents are intended to be used by the Image Registry Operator developer
 * [How the Operator gets credentials for the image registry](credentials-flow.md)
 * [How to deploy a development build of the Image Registry Operator](development-build.md)
 * [The image registry life cycle](life-cycle.md)
+* [Exported metrics](metrics.md)

docs/metrics.md

@@ -0,0 +1,15 @@
+# Exported metrics
+
+## `imageregistry:operations_count:sum`
+
+| Operation | Resource type      |
+| --------- | ------------------ |
+| `get`     | `blob`, `manifest` |
+| `create`  | `blob`, `manifest` |
+
+## `imageregistry:imagestreamtags_count:sum`
+
+| Source     | Location             | Description                                                   |
+| ---------- | -------------------- | ------------------------------------------------------------- |
+| `imported` | `openshift`, `other` | Image Stream Tags imported in 'openshift' or other namespaces |
+| `pushed`   | `openshift`, `other` | Image Stream Tags pushed to 'openshift' or other namespaces   |

manifests/09-prometheus-rules-imagestreams.yaml

@@ -3,6 +3,10 @@ kind: PrometheusRule
 metadata:
   name: imagestreams-rules
   namespace: openshift-image-registry
+  annotations:
+    include.release.openshift.io/ibm-cloud-managed: "true"
+    include.release.openshift.io/self-managed-high-availability: "true"
+    include.release.openshift.io/single-node-developer: "true"
 spec:
   groups:
   - name: imagestreams.rules

manifests/09-prometheus-rules-registry-operations.yaml

@@ -0,0 +1,44 @@
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  name: image-registry-rules
+  namespace: openshift-image-registry
+  annotations:
+    include.release.openshift.io/ibm-cloud-managed: "true"
+    include.release.openshift.io/self-managed-high-availability: "true"
+    include.release.openshift.io/single-node-developer: "true"
+spec:
+  groups:
+  - name: imageregistry.operations.rules
+    rules:
+    - expr: |
+        label_replace(
+          label_replace(
+            sum by (operation) (imageregistry_request_duration_seconds_count{operation="BlobStore.ServeBlob"}), "operation", "get", "operation", "(.+)"
+          ), "resource_type", "blob", "resource_type", ""
+        )
+      record: imageregistry:operations_count:sum
+
+    - expr: |
+        label_replace(
+          label_replace(
+            sum by (operation) (imageregistry_request_duration_seconds_count{operation="BlobStore.Create"}), "operation", "create", "operation", "(.+)"
+          ), "resource_type", "blob", "resource_type", ""
+        )
+      record: imageregistry:operations_count:sum
+
+    - expr: |
+        label_replace(
+          label_replace(
+            sum by (operation) (imageregistry_request_duration_seconds_count{operation="ManifestService.Get"}), "operation", "get", "operation", "(.+)"
+          ), "resource_type", "manifest", "resource_type", ""
+        )
+      record: imageregistry:operations_count:sum
+
+    - expr: |
+        label_replace(
+          label_replace(
+            sum by (operation) (imageregistry_request_duration_seconds_count{operation="ManifestService.Put"}), "operation", "create", "operation", "(.+)"
+          ), "resource_type", "manifest", "resource_type", ""
+        )
+      record: imageregistry:operations_count:sum