commit c8e4a594a2cc4f9063f57a8de6a3f6ed7138df4e
parent d0a12cdda34300c57f57559cd2038c2a3c4a2363
Author: MTRNord <MTRNord@users.noreply.github.com>
Date: Mon, 4 Aug 2025 14:51:39 +0200
Migrate connectivity-tester
Diffstat:
7 files changed, 582 insertions(+), 0 deletions(-)
diff --git a/apps/talos_cluster/connectivity-tester/config.json b/apps/talos_cluster/connectivity-tester/config.json
@@ -0,0 +1,3 @@
+{
+ "api_server_url": "https://api.connectivity-tester.mtrnord.blog"
+}
diff --git a/apps/talos_cluster/connectivity-tester/deployment.yaml b/apps/talos_cluster/connectivity-tester/deployment.yaml
@@ -0,0 +1,363 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+ name: connectivity-tester
+ namespace: matrix
+spec:
+ progressDeadlineSeconds: 120
+ strategy:
+ rollingUpdate:
+ maxUnavailable: 0
+ type: RollingUpdate
+ selector:
+ matchLabels:
+ app: connectivity-tester
+ template:
+ metadata:
+ labels:
+ app: connectivity-tester
+ spec:
+ affinity:
+ podAntiAffinity:
+ requiredDuringSchedulingIgnoredDuringExecution:
+ - labelSelector:
+ matchExpressions:
+ - key: app
+ operator: In
+ values:
+ - connectivity-tester
+ topologyKey: "kubernetes.io/hostname"
+ imagePullSecrets:
+ - name: ghcr-pull
+ hostNetwork: false
+ containers:
+ - name: connection-checker-ui
+ image: ghcr.io/mtrnord/matrix-connection-tester-ui:v0.2.0
+ imagePullPolicy: Always
+ lifecycle:
+ preStop:
+ exec:
+ command:
+ - sleep
+ - "10"
+ resources:
+ limits: {}
+ requests:
+ memory: "344Mi"
+ cpu: "252m"
+ ports:
+ - containerPort: 3000
+ name: web
+ protocol: TCP
+ volumeMounts:
+ - name: configs
+ mountPath: "/usr/share/nginx/html/config.json"
+ subPath: config.json
+ readOnly: true
+ - mountPath: /tmp
+ name: tmp
+ livenessProbe:
+ httpGet:
+ path: /
+ port: web
+ scheme: HTTP
+ readinessProbe:
+ httpGet:
+ path: /
+ port: web
+ scheme: HTTP
+ startupProbe:
+ httpGet:
+ path: /
+ port: web
+ volumes:
+ - name: configs
+ configMap:
+ name: connectivity-tester-config
+ - name: tmp
+ emptyDir:
+ sizeLimit: 2048Mi
+---
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+ name: connectivity-tester-api
+ namespace: matrix
+spec:
+ progressDeadlineSeconds: 120
+ strategy:
+ rollingUpdate:
+ maxUnavailable: 0
+ type: RollingUpdate
+ selector:
+ matchLabels:
+ app: connectivity-tester-api
+ template:
+ metadata:
+ labels:
+ app: connectivity-tester-api
+ spec:
+ affinity:
+ podAntiAffinity:
+ requiredDuringSchedulingIgnoredDuringExecution:
+ - labelSelector:
+ matchExpressions:
+ - key: app
+ operator: In
+ values:
+ - connectivity-tester-api
+ topologyKey: "kubernetes.io/hostname"
+ imagePullSecrets:
+ - name: ghcr-pull
+ hostNetwork: true
+ dnsPolicy: ClusterFirstWithHostNet
+ containers:
+ - name: federation-tester-api
+ image: ghcr.io/mtrnord/rust-federation-tester:v0.2.2
+ imagePullPolicy: IfNotPresent
+ resources:
+ limits: {}
+ requests:
+ memory: "344Mi"
+ cpu: "252m"
+ volumeMounts:
+ - name: api-config
+ mountPath: /app/config.yaml
+ subPath: config.yaml
+ readOnly: true
+ securityContext:
+ runAsUser: 1000
+ runAsGroup: 1000
+ readOnlyRootFilesystem: true
+ ports:
+ - containerPort: 8080
+ name: api
+ protocol: TCP
+ readinessProbe:
+ httpGet:
+ path: /healthz
+ port: api
+ scheme: HTTP
+ livenessProbe:
+ httpGet:
+ path: /healthz
+ port: api
+ scheme: HTTP
+ startupProbe:
+ httpGet:
+ path: /healthz
+ port: api
+ volumes:
+ - name: api-config
+ secret:
+ secretName: connectivity-tester-config
+---
+apiVersion: flagger.app/v1beta1
+kind: MetricTemplate
+metadata:
+ name: connectivity-tester-latency
+ namespace: matrix
+spec:
+ provider:
+ type: prometheus
+ address: "http://vmselect-vm-victoria-metrics-k8s-stack.monitoring.svc:8481/select/0/prometheus"
+ query: |
+ histogram_quantile(0.99,
+ sum(
+ irate(
+ envoy_cluster_upstream_rq_time_bucket{
+ envoy_cluster_name=~".*{{ namespace }}/{{ target }}.*",
+ }[{{ interval }}]
+ )
+ ) by (le)
+ )/1000
+---
+apiVersion: flagger.app/v1beta1
+kind: MetricTemplate
+metadata:
+ name: connectivity-tester-error-rate
+ namespace: matrix
+spec:
+ provider:
+ type: prometheus
+ address: "http://vmselect-vm-victoria-metrics-k8s-stack.monitoring.svc:8481/select/0/prometheus"
+ query: |
+ sum(
+ rate(
+ envoy_cluster_upstream_rq{
+ envoy_cluster_name=~".*{{ namespace }}/{{ target }}.*",
+ envoy_response_code!~"5.*"
+ }[{{ interval }}]
+ )
+ )
+ /
+ sum(
+ rate(
+ envoy_cluster_upstream_rq_total{
+ envoy_cluster_name=~".*{{ namespace }}/{{ target }}.*",
+ }[{{ interval }}]
+ )
+ ) / 100
+---
+apiVersion: flagger.app/v1beta1
+kind: Canary
+metadata:
+ name: connectivity-tester
+ namespace: matrix
+spec:
+ # deployment reference
+ targetRef:
+ apiVersion: apps/v1
+ kind: Deployment
+ name: connectivity-tester
+ # the maximum time in seconds for the canary deployment
+ # to make progress before it is rollback (default 600s)
+ progressDeadlineSeconds: 60
+ # HPA reference (optional)
+ #autoscalerRef:
+ # apiVersion: autoscaling/v2
+ # kind: HorizontalPodAutoscaler
+ # name: connectivity-tester
+ service:
+ # service port number
+ port: 3000
+ # container port number or name (optional)
+ targetPort: 3000
+ # Gateway API HTTPRoute host names
+ hosts:
+ - connectivity-tester.mtrnord.blog
+ - federationtester.mtrnord.blog
+ # Reference to the Gateway that the generated HTTPRoute would attach to.
+ gatewayRefs:
+ - name: envoy-gateway
+ namespace: envoy-gateway
+ retries:
+ attempts: 10
+ perTryTimeout: 5s
+ retryOn: "gateway-error,connect-failure,refused-stream"
+ analysis:
+ # schedule interval (default 60s)
+ interval: 1m
+ # max number of failed metric checks before rollback
+ threshold: 5
+ # max traffic percentage routed to canary
+ # percentage (0-100)
+ maxWeight: 50
+ # canary increment step
+ # percentage (0-100)
+ stepWeight: 10
+ metrics:
+ - name: error-rate
+ # max error rate (5xx responses)
+ # percentage (0-100)
+ templateRef:
+ name: connectivity-tester-error-rate
+ namespace: matrix
+ thresholdRange:
+ max: 2
+ interval: 1m
+ - name: latency
+ templateRef:
+ name: connectivity-tester-latency
+ namespace: matrix
+ # seconds
+ thresholdRange:
+ max: 1
+ interval: 30s
+ sessionAffinity:
+ cookieName: flagger-cookie
+ # testing (optional)
+ webhooks:
+ - name: smoke-test
+ type: pre-rollout
+ url: http://flagger-loadtester.flagger-system/
+ timeout: 15s
+ metadata:
+ type: bash
+ cmd: "curl -s http://connectivity-tester-canary.matrix:3000"
+ - name: load-test
+ url: http://flagger-loadtester.flagger-system/
+ timeout: 10m
+ metadata:
+ cmd: "hey -z 10m -q 10 -c 2 http://connectivity-tester-canary.matrix/?serverName=mtrnord.blog"
+---
+apiVersion: flagger.app/v1beta1
+kind: Canary
+metadata:
+ name: connectivity-tester-api
+ namespace: matrix
+spec:
+ # deployment reference
+ targetRef:
+ apiVersion: apps/v1
+ kind: Deployment
+ name: connectivity-tester-api
+ # the maximum time in seconds for the canary deployment
+ # to make progress before it is rollback (default 600s)
+ progressDeadlineSeconds: 60
+ # HPA reference (optional)
+ #autoscalerRef:
+ # apiVersion: autoscaling/v2
+ # kind: HorizontalPodAutoscaler
+ # name: connectivity-tester
+ service:
+ # service port number
+ port: 8080
+ # container port number or name (optional)
+ targetPort: 8080
+ # Gateway API HTTPRoute host names
+ hosts:
+ - api.connectivity-tester.mtrnord.blog
+ # Reference to the Gateway that the generated HTTPRoute would attach to.
+ gatewayRefs:
+ - name: envoy-gateway
+ namespace: envoy-gateway
+ retries:
+ attempts: 10
+ perTryTimeout: 5s
+ retryOn: "gateway-error,connect-failure,refused-stream"
+ analysis:
+ # schedule interval (default 60s)
+ interval: 1m
+ # max number of failed metric checks before rollback
+ threshold: 5
+ # max traffic percentage routed to canary
+ # percentage (0-100)
+ maxWeight: 50
+ # canary increment step
+ # percentage (0-100)
+ stepWeight: 10
+ metrics:
+ - name: error-rate
+ # max error rate (5xx responses)
+ # percentage (0-100)
+ templateRef:
+ name: connectivity-tester-error-rate
+ namespace: matrix
+ thresholdRange:
+ max: 2
+ interval: 1m
+ - name: latency
+ templateRef:
+ name: connectivity-tester-latency
+ namespace: matrix
+ # seconds
+ thresholdRange:
+ max: 1
+ interval: 30s
+ sessionAffinity:
+ cookieName: flagger-cookie
+ # testing (optional)
+ webhooks:
+ - name: smoke-test
+ type: pre-rollout
+ url: http://flagger-loadtester.flagger-system/
+ timeout: 15s
+ metadata:
+ type: bash
+ cmd: "curl -s http://connectivity-tester-api-canary.matrix:8080/healthz | grep 'ok'"
+ - name: load-test
+ url: http://flagger-loadtester.flagger-system/
+ timeout: 10m
+ metadata:
+ cmd: "hey -z 10m -q 10 -c 2 http://connectivity-tester-api-canary.matrix/api/federation/federation-ok?serverName=mtrnord.blog&no_cache=true"
diff --git a/apps/talos_cluster/connectivity-tester/kustomization.yaml b/apps/talos_cluster/connectivity-tester/kustomization.yaml
@@ -0,0 +1,16 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+namespace: matrix
+resources:
+ - staging_deployment.yaml
+ - deployment.yaml
+ - service.yaml
+configMapGenerator:
+ - name: connectivity-tester-config
+ namespace: matrix
+ files:
+ - config.json=config.json
+ - name: connectivity-tester-stage-config
+ namespace: matrix
+ files:
+ - config.json=staging_config.json
diff --git a/apps/talos_cluster/connectivity-tester/service.yaml b/apps/talos_cluster/connectivity-tester/service.yaml
@@ -0,0 +1,23 @@
+apiVersion: gateway.networking.k8s.io/v1
+kind: HTTPRoute
+metadata:
+ name: connectivity-tester-api-temp
+ namespace: matrix
+spec:
+ parentRefs:
+ - name: envoy-gateway
+ namespace: envoy-gateway
+ hostnames:
+ - connectivity-tester.mtrnord.blog
+ - federationtester.mtrnord.blog
+ rules:
+ - matches:
+ - path:
+ value: /api
+ type: PathPrefix
+ backendRefs:
+ - name: connectivity-tester-api-primary
+ port: 8080
+ timeouts:
+ request: 240s
+ backendRequest: 0s
diff --git a/apps/talos_cluster/connectivity-tester/staging_config.json b/apps/talos_cluster/connectivity-tester/staging_config.json
@@ -0,0 +1,3 @@
+{
+ "api_server_url": "https://stage.connectivity-tester.mtrnord.blog"
+}
diff --git a/apps/talos_cluster/connectivity-tester/staging_deployment.yaml b/apps/talos_cluster/connectivity-tester/staging_deployment.yaml
@@ -0,0 +1,173 @@
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+ name: connectivity-tester-stage
+ namespace: matrix
+spec:
+ progressDeadlineSeconds: 120
+ strategy:
+ rollingUpdate:
+ maxUnavailable: 0
+ type: RollingUpdate
+ selector:
+ matchLabels:
+ app: connectivity-tester-stage
+ template:
+ metadata:
+ labels:
+ app: connectivity-tester-stage
+ spec:
+ affinity:
+ podAntiAffinity:
+ requiredDuringSchedulingIgnoredDuringExecution:
+ - labelSelector:
+ matchExpressions:
+ - key: app
+ operator: In
+ values:
+ - connectivity-tester-stage
+ topologyKey: "kubernetes.io/hostname"
+ - labelSelector:
+ matchExpressions:
+ - key: app
+ operator: In
+ values:
+ - connectivity-tester-api
+ topologyKey: "kubernetes.io/hostname"
+ imagePullSecrets:
+ - name: ghcr-pull
+ hostNetwork: true
+ dnsPolicy: ClusterFirstWithHostNet
+ containers:
+ - name: federation-tester-api
+ image: ghcr.io/mtrnord/rust-federation-tester:main
+ imagePullPolicy: Always
+ resources:
+ limits: {}
+ requests:
+ memory: "344Mi"
+ cpu: "252m"
+ securityContext:
+ runAsUser: 1000
+ runAsGroup: 1000
+ readOnlyRootFilesystem: true
+ volumeMounts:
+ - name: api-config
+ mountPath: /app/config.yaml
+ subPath: config.yaml
+ readOnly: true
+ ports:
+ - containerPort: 8080
+ name: api
+ protocol: TCP
+ readinessProbe:
+ httpGet:
+ path: /healthz
+ port: api
+ scheme: HTTP
+ livenessProbe:
+ httpGet:
+ path: /healthz
+ port: api
+ scheme: HTTP
+ startupProbe:
+ httpGet:
+ path: /healthz
+ port: api
+ - name: connection-checker-ui
+ image: ghcr.io/mtrnord/matrix-connection-tester-ui:main
+ imagePullPolicy: Always
+ lifecycle:
+ preStop:
+ exec:
+ command:
+ - sleep
+ - "10"
+ resources:
+ limits: {}
+ requests:
+ memory: "344Mi"
+ cpu: "252m"
+ ports:
+ - containerPort: 3000
+ name: web
+ protocol: TCP
+ volumeMounts:
+ - name: configs
+ mountPath: "/usr/share/nginx/html/config.json"
+ subPath: config.json
+ readOnly: true
+ - mountPath: /tmp
+ name: tmp
+ livenessProbe:
+ httpGet:
+ path: /
+ port: web
+ scheme: HTTP
+ readinessProbe:
+ httpGet:
+ path: /
+ port: web
+ scheme: HTTP
+ startupProbe:
+ httpGet:
+ path: /
+ port: web
+ volumes:
+ - name: configs
+ configMap:
+ name: connectivity-tester-stage-config
+ - name: tmp
+ emptyDir:
+ sizeLimit: 2048Mi
+ - name: api-config
+ secret:
+ secretName: connectivity-tester-stage-config
+---
+apiVersion: v1
+kind: Service
+metadata:
+ name: connectivity-tester-stage
+spec:
+ selector:
+ app: connectivity-tester-stage
+ ports:
+ - name: api
+ port: 8080
+ targetPort: 8080
+ - name: web
+ port: 3000
+ targetPort: 3000
+---
+apiVersion: gateway.networking.k8s.io/v1
+kind: HTTPRoute
+metadata:
+ name: connectivity-tester-stage
+ namespace: matrix
+spec:
+ parentRefs:
+ - name: envoy-gateway
+ namespace: envoy-gateway
+ hostnames:
+ - stage.connectivity-tester.mtrnord.blog
+ rules:
+ - matches:
+ - path:
+ value: /api
+ type: PathPrefix
+ backendRefs:
+ - name: connectivity-tester-stage
+ port: 8080
+ timeouts:
+ request: 240s
+ backendRequest: 0s
+ - matches:
+ - path:
+ value: /
+ type: PathPrefix
+ backendRefs:
+ - name: connectivity-tester-stage
+ port: 3000
+ timeouts:
+ request: 240s
+ backendRequest: 0s
diff --git a/apps/talos_cluster/kustomization.yaml b/apps/talos_cluster/kustomization.yaml
@@ -12,3 +12,4 @@ resources:
- ./collabora
- ./flagger
- ./gotosocial
+ - ./connectivity-tester
