23f814ae29
Spark executors crashed immediately (exit 1) because the driver advertised its pod name as the RPC hostname, which is not DNS-resolvable in-cluster. Added headless Service `spark-live-dashboard-driver` and pinned spark.driver.host/port + spark.blockManager.port so workers can connect back to the driver. Stops the JVM fork storm that saturated all three nodes at ~100% CPU. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
97 lines
2.6 KiB
YAML
Executable File
97 lines
2.6 KiB
YAML
Executable File
apiVersion: v1
|
|
kind: Service
|
|
metadata:
|
|
name: spark-live-dashboard-driver
|
|
labels:
|
|
app.kubernetes.io/name: spark-analytics
|
|
app.kubernetes.io/part-of: nowchess
|
|
spec:
|
|
clusterIP: None
|
|
selector:
|
|
app: spark-live-dashboard
|
|
ports:
|
|
- name: driver
|
|
port: 37007
|
|
targetPort: 37007
|
|
- name: blockmanager
|
|
port: 37777
|
|
targetPort: 37777
|
|
---
|
|
apiVersion: apps/v1
|
|
kind: Deployment
|
|
metadata:
|
|
name: spark-live-dashboard
|
|
labels:
|
|
app.kubernetes.io/name: spark-analytics
|
|
app.kubernetes.io/part-of: nowchess
|
|
spec:
|
|
replicas: 1
|
|
selector:
|
|
matchLabels:
|
|
app: spark-live-dashboard
|
|
template:
|
|
metadata:
|
|
labels:
|
|
app: spark-live-dashboard
|
|
app.kubernetes.io/name: spark-analytics
|
|
app.kubernetes.io/part-of: nowchess
|
|
spec:
|
|
serviceAccountName: spark-analytics
|
|
securityContext:
|
|
fsGroup: 185
|
|
imagePullSecrets:
|
|
- name: ghcr-pull-secret
|
|
containers:
|
|
- name: spark-driver
|
|
image: ghcr.io/now-chess/now-chess-systems/analytics:latest
|
|
command:
|
|
- /opt/spark/bin/spark-submit
|
|
- --master
|
|
- spark://spark-master:7077
|
|
- --conf
|
|
- spark.executor.instances=1
|
|
- --conf
|
|
- spark.executor.memory=512m
|
|
- --conf
|
|
- spark.driver.memory=512m
|
|
- --conf
|
|
- spark.driver.host=spark-live-dashboard-driver
|
|
- --conf
|
|
- spark.driver.port=37007
|
|
- --conf
|
|
- spark.blockManager.port=37777
|
|
- --class
|
|
- de.nowchess.analytics.LiveDashboardJob
|
|
- local:///app/analytics.jar
|
|
- /spark-checkpoint
|
|
env:
|
|
- name: NOWCHESS_JDBC_URL
|
|
valueFrom:
|
|
secretKeyRef:
|
|
name: ncs-db-secrets
|
|
key: STORE_DB_URL
|
|
- name: NOWCHESS_DB_USER
|
|
valueFrom:
|
|
secretKeyRef:
|
|
name: ncs-db-secrets
|
|
key: STORE_DB_USER
|
|
- name: NOWCHESS_DB_PASS
|
|
valueFrom:
|
|
secretKeyRef:
|
|
name: ncs-db-secrets
|
|
key: STORE_DB_PASSWORD
|
|
resources:
|
|
requests:
|
|
cpu: 50m
|
|
memory: 64Mi
|
|
limits:
|
|
cpu: "1"
|
|
memory: 1Gi
|
|
volumeMounts:
|
|
- name: spark-checkpoint
|
|
mountPath: /spark-checkpoint
|
|
volumes:
|
|
- name: spark-checkpoint
|
|
persistentVolumeClaim:
|
|
claimName: spark-streaming-checkpoint
|