backend redesign

This commit is contained in:
2026-03-11 18:47:11 -04:00
parent 8ff277c8c6
commit e99ef5d2dd
210 changed files with 12147 additions and 155 deletions

View File

@@ -0,0 +1,40 @@
# Flink Job Configuration
# ZeroMQ bind address and ports
zmq_bind_address: "tcp://*"
zmq_ingestor_work_queue_port: 5555
zmq_ingestor_response_port: 5556
zmq_ingestor_control_port: 5557
zmq_market_data_pub_port: 5558
zmq_client_request_port: 5559
zmq_cep_webhook_port: 5560
# Notification endpoints
# Task managers PUSH to job manager PULL socket at this address
notification_publish_endpoint: "tcp://flink-jobmanager:5561"
# Job manager binds PULL socket on this port to receive from task managers
notification_pull_port: 5561
# Kafka configuration
kafka_bootstrap_servers: "kafka:9092"
kafka_tick_topic: "market-tick"
kafka_ohlc_topic: "market-ohlc"
kafka_topics_file: "/topics-dev.yaml" # Use topics-dev.yaml for single broker dev environment
# Iceberg catalog
iceberg_catalog_uri: "http://iceberg-catalog:8181"
iceberg_warehouse: "s3://trading-warehouse/"
iceberg_namespace: "trading"
iceberg_table_prefix: "market"
hadoop_conf_dir: "/etc/hadoop/conf"
# Flink configuration
flink_parallelism: 1
flink_checkpoint_interval_ms: 60000
# Flink memory configuration (required)
jobmanager.memory.process.size: 1600m
taskmanager.memory.process.size: 1728m
taskmanager.numberOfTaskSlots: 2
jobmanager.rpc.address: flink-jobmanager
jobmanager.rpc.port: 6123

View File

@@ -0,0 +1,24 @@
# CCXT Ingestor Configuration
# Relay ZMQ endpoints (relay is the well-known gateway)
flink_hostname: relay
ingestor_work_port: 5555 # SUB - receives DataRequest with exchange prefix
# Note: No response port needed - async architecture via Kafka!
# Supported exchanges (subscribe to these prefixes)
supported_exchanges:
- BINANCE
- COINBASE
- KRAKEN
# Kafka configuration
kafka_brokers:
- kafka:9092
kafka_topic: market-ohlc
# Worker configuration
max_concurrent: 10
poll_interval_ms: 10000
# Logging
log_level: info

View File

@@ -0,0 +1,19 @@
# ZMQ Relay Configuration
# Bind address for all relay sockets
bind_address: "tcp://*"
# Client-facing ports
client_request_port: 5559 # ROUTER - Client historical data requests
market_data_pub_port: 5558 # XPUB - Market data fanout to clients
# Ingestor-facing ports
ingestor_work_port: 5555 # PUB - Distribute work with exchange prefix
ingestor_response_port: 5556 # ROUTER - Receive responses from ingestors
# Flink connection
flink_market_data_endpoint: "tcp://flink-jobmanager:5558" # XSUB - Subscribe to Flink market data (MARKET_DATA_PUB)
# Timeouts and limits
request_timeout_secs: 30 # Timeout for pending client requests
high_water_mark: 10000 # ZMQ high water mark for all sockets

View File

@@ -0,0 +1,519 @@
---
# Kafka (KRaft mode - no Zookeeper needed)
# Using apache/kafka:3.9.0 instead of confluentinc/cp-kafka because:
# - cp-kafka's entrypoint script has issues with KRaft configuration
# - apache/kafka allows explicit command configuration
# - For production, use Strimzi operator (see kafka/ directory)
apiVersion: v1
kind: Service
metadata:
name: kafka
spec:
selector:
app: kafka
ports:
- name: broker
protocol: TCP
port: 9092
targetPort: 9092
- name: controller
protocol: TCP
port: 9093
targetPort: 9093
type: ClusterIP
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: kafka
spec:
serviceName: kafka
replicas: 1
selector:
matchLabels:
app: kafka
template:
metadata:
labels:
app: kafka
spec:
containers:
- name: kafka
image: apache/kafka:3.9.0
ports:
- containerPort: 9092
name: broker
- containerPort: 9093
name: controller
command:
- sh
- -c
- |
CLUSTER_ID="dexorder-dev-cluster"
if [ ! -f /var/lib/kafka/data/meta.properties ]; then
/opt/kafka/bin/kafka-storage.sh format -t $CLUSTER_ID -c /opt/kafka/config/kraft/server.properties
fi
/opt/kafka/bin/kafka-server-start.sh /opt/kafka/config/kraft/server.properties \
--override node.id=1 \
--override process.roles=broker,controller \
--override listeners=PLAINTEXT://:9092,CONTROLLER://:9093 \
--override advertised.listeners=PLAINTEXT://kafka:9092 \
--override controller.quorum.voters=1@kafka:9093 \
--override controller.listener.names=CONTROLLER \
--override listener.security.protocol.map=CONTROLLER:PLAINTEXT,PLAINTEXT:PLAINTEXT \
--override log.dirs=/var/lib/kafka/data \
--override offsets.topic.replication.factor=1 \
--override transaction.state.log.replication.factor=1 \
--override transaction.state.log.min.isr=1
env: []
volumeMounts:
- name: kafka-data
mountPath: /var/lib/kafka/data
volumeClaimTemplates:
- metadata:
name: kafka-data
spec:
accessModes: ["ReadWriteOnce"]
resources:
requests:
storage: 5Gi
---
# PostgreSQL (for Iceberg catalog metadata)
apiVersion: v1
kind: Service
metadata:
name: postgres
spec:
selector:
app: postgres
ports:
- protocol: TCP
port: 5432
targetPort: 5432
type: ClusterIP
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: postgres
spec:
serviceName: postgres
replicas: 1
selector:
matchLabels:
app: postgres
template:
metadata:
labels:
app: postgres
spec:
containers:
- name: postgres
image: postgres:15
ports:
- containerPort: 5432
env:
- name: POSTGRES_USER
value: postgres
- name: POSTGRES_PASSWORD
valueFrom:
secretKeyRef:
name: postgres-secret
key: password
- name: POSTGRES_DB
value: iceberg
volumeMounts:
- name: postgres-data
mountPath: /var/lib/postgresql/data
volumeClaimTemplates:
- metadata:
name: postgres-data
spec:
accessModes: ["ReadWriteOnce"]
resources:
requests:
storage: 2Gi
---
# MinIO (S3-compatible object storage)
apiVersion: v1
kind: Service
metadata:
name: minio
spec:
selector:
app: minio
ports:
- name: api
protocol: TCP
port: 9000
targetPort: 9000
- name: console
protocol: TCP
port: 9001
targetPort: 9001
type: ClusterIP
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
name: minio
spec:
serviceName: minio
replicas: 1
selector:
matchLabels:
app: minio
template:
metadata:
labels:
app: minio
spec:
containers:
- name: minio
image: minio/minio:latest
args:
- server
- /data
- --console-address
- ":9001"
ports:
- containerPort: 9000
name: api
- containerPort: 9001
name: console
env:
- name: MINIO_ROOT_USER
valueFrom:
secretKeyRef:
name: minio-secret
key: root-user
- name: MINIO_ROOT_PASSWORD
valueFrom:
secretKeyRef:
name: minio-secret
key: root-password
volumeMounts:
- name: minio-data
mountPath: /data
volumeClaimTemplates:
- metadata:
name: minio-data
spec:
accessModes: ["ReadWriteOnce"]
resources:
requests:
storage: 10Gi
---
# Iceberg REST Catalog
apiVersion: v1
kind: Service
metadata:
name: iceberg-catalog
spec:
selector:
app: iceberg-catalog
ports:
- protocol: TCP
port: 8181
targetPort: 8181
type: ClusterIP
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: iceberg-catalog
spec:
replicas: 1
selector:
matchLabels:
app: iceberg-catalog
template:
metadata:
labels:
app: iceberg-catalog
spec:
initContainers:
- name: wait-for-postgres
image: busybox:1.36
command: ['sh', '-c', 'until nc -z postgres 5432; do echo waiting for postgres; sleep 2; done;']
- name: wait-for-minio
image: busybox:1.36
command: ['sh', '-c', 'until nc -z minio 9000; do echo waiting for minio; sleep 2; done;']
containers:
- name: iceberg-catalog
image: tabulario/iceberg-rest:latest
ports:
- containerPort: 8181
env:
- name: CATALOG_WAREHOUSE
value: "s3://warehouse/"
- name: CATALOG_IO__IMPL
value: "org.apache.iceberg.aws.s3.S3FileIO"
- name: CATALOG_S3_ENDPOINT
value: "http://minio:9000"
- name: CATALOG_S3_ACCESS__KEY__ID
valueFrom:
secretKeyRef:
name: minio-secret
key: root-user
- name: CATALOG_S3_SECRET__ACCESS__KEY
valueFrom:
secretKeyRef:
name: minio-secret
key: root-password
- name: CATALOG_S3_PATH__STYLE__ACCESS
value: "true"
- name: AWS_REGION
value: "us-east-1"
---
# Flink JobManager
apiVersion: v1
kind: Service
metadata:
name: flink-jobmanager
spec:
selector:
app: flink-jobmanager
ports:
- name: rpc
protocol: TCP
port: 6123
targetPort: 6123
- name: ui
protocol: TCP
port: 8081
targetPort: 8081
- name: zmq-market-data
protocol: TCP
port: 5558
targetPort: 5558
- name: zmq-notif-pull
protocol: TCP
port: 5561
targetPort: 5561
type: ClusterIP
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: flink-jobmanager
spec:
replicas: 1
selector:
matchLabels:
app: flink-jobmanager
template:
metadata:
labels:
app: flink-jobmanager
spec:
initContainers:
- name: wait-for-kafka
image: busybox:1.36
command: ['sh', '-c', 'until nc -z kafka 9092; do echo waiting for kafka; sleep 2; done;']
containers:
- name: flink-jobmanager
image: dexorder/flink:latest
imagePullPolicy: Never
args: ["standalone-job", "--job-classname", "com.dexorder.flink.TradingFlinkApp"]
ports:
- containerPort: 6123
name: rpc
- containerPort: 8081
name: ui
- containerPort: 5558
name: zmq-market-data
- containerPort: 5561
name: zmq-notif-pull
env:
- name: JOB_MANAGER_RPC_ADDRESS
value: flink-jobmanager
- name: AWS_REGION
value: us-east-1
- name: AWS_ACCESS_KEY_ID
valueFrom:
secretKeyRef:
name: flink-secrets
key: minio-access-key
- name: AWS_SECRET_ACCESS_KEY
valueFrom:
secretKeyRef:
name: flink-secrets
key: minio-secret-key
volumeMounts:
- name: flink-config
mountPath: /etc/config/config.yaml
subPath: config.yaml
- name: flink-secrets
mountPath: /etc/secrets
volumes:
- name: flink-config
configMap:
name: flink-config
- name: flink-secrets
secret:
secretName: flink-secrets
---
# Flink TaskManager
apiVersion: apps/v1
kind: Deployment
metadata:
name: flink-taskmanager
spec:
replicas: 1
selector:
matchLabels:
app: flink-taskmanager
template:
metadata:
labels:
app: flink-taskmanager
spec:
initContainers:
- name: wait-for-jobmanager
image: busybox:1.36
command: ['sh', '-c', 'until nc -z flink-jobmanager 6123; do echo waiting for jobmanager; sleep 2; done;']
containers:
- name: flink-taskmanager
image: dexorder/flink:latest
imagePullPolicy: Never
args: ["taskmanager"]
env:
- name: JOB_MANAGER_RPC_ADDRESS
value: flink-jobmanager
- name: AWS_REGION
value: us-east-1
- name: AWS_ACCESS_KEY_ID
valueFrom:
secretKeyRef:
name: flink-secrets
key: minio-access-key
- name: AWS_SECRET_ACCESS_KEY
valueFrom:
secretKeyRef:
name: flink-secrets
key: minio-secret-key
volumeMounts:
- name: flink-config
mountPath: /etc/config/config.yaml
subPath: config.yaml
- name: flink-secrets
mountPath: /etc/secrets
volumes:
- name: flink-config
configMap:
name: flink-config
- name: flink-secrets
secret:
secretName: flink-secrets
---
# Relay (ZMQ router)
apiVersion: v1
kind: Service
metadata:
name: relay
spec:
selector:
app: relay
ports:
- name: work-queue
protocol: TCP
port: 5555
targetPort: 5555
- name: responses
protocol: TCP
port: 5556
targetPort: 5556
- name: market-data
protocol: TCP
port: 5558
targetPort: 5558
- name: client-requests
protocol: TCP
port: 5559
targetPort: 5559
type: ClusterIP
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: relay
spec:
replicas: 1
selector:
matchLabels:
app: relay
template:
metadata:
labels:
app: relay
spec:
containers:
- name: relay
image: dexorder/relay:latest
imagePullPolicy: Never
ports:
- containerPort: 5555
name: work-queue
- containerPort: 5556
name: responses
- containerPort: 5558
name: market-data
- containerPort: 5559
name: client-requests
env:
- name: RUST_LOG
value: relay=info
- name: CONFIG_PATH
value: /config/config.yaml
volumeMounts:
- name: relay-config
mountPath: /config
volumes:
- name: relay-config
configMap:
name: relay-config
---
# Ingestor (CCXT data fetcher)
apiVersion: apps/v1
kind: Deployment
metadata:
name: ingestor
spec:
replicas: 1
selector:
matchLabels:
app: ingestor
template:
metadata:
labels:
app: ingestor
spec:
initContainers:
- name: wait-for-relay
image: busybox:1.36
command: ['sh', '-c', 'until nc -z relay 5555; do echo waiting for relay; sleep 2; done;']
- name: wait-for-kafka
image: busybox:1.36
command: ['sh', '-c', 'until nc -z kafka 9092; do echo waiting for kafka; sleep 2; done;']
containers:
- name: ingestor
image: dexorder/ingestor:latest
imagePullPolicy: Never
env:
- name: LOG_LEVEL
value: info
- name: CONFIG_PATH
value: /config/config.yaml
volumeMounts:
- name: ingestor-config
mountPath: /config
- name: ingestor-secrets
mountPath: /secrets
volumes:
- name: ingestor-config
configMap:
name: ingestor-config
- name: ingestor-secrets
secret:
secretName: ingestor-secrets

View File

@@ -0,0 +1,11 @@
---
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: ai-ingress
spec:
ingressClassName: nginx
rules:
- host: dexorder.local
http:
paths: []

View File

@@ -0,0 +1,32 @@
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization
namespace: default
# Base resources
resources:
- ../base
- infrastructure.yaml
# No patches needed currently
patches: []
# ingress-dev.yaml - removed until we have services to expose
# ConfigMaps for service configs
configMapGenerator:
- name: relay-config
files:
- config.yaml=configs/relay-config.yaml
- name: ingestor-config
files:
- config.yaml=configs/ingestor-config.yaml
- name: flink-config
files:
- config.yaml=configs/flink-config.yaml
# Secrets (managed via kubectl, not committed)
# These are created by bin/secret-update
secretGenerator: []
generatorOptions:
disableNameSuffixHash: true

View File

@@ -0,0 +1,7 @@
apiVersion: v1
kind: Secret
metadata:
name: ai-secrets
type: Opaque
stringData:
anthropic-api-key: "sk-ant-YOUR_KEY_HERE"

View File

@@ -0,0 +1,9 @@
apiVersion: v1
kind: Secret
metadata:
name: flink-secrets
type: Opaque
stringData:
# MinIO/S3 credentials for Iceberg S3FileIO
minio-access-key: "minio"
minio-secret-key: "minio123"

View File

@@ -0,0 +1,13 @@
apiVersion: v1
kind: Secret
metadata:
name: ingestor-secrets
type: Opaque
stringData:
# Exchange API keys (if needed for authenticated endpoints)
binance-api-key: ""
binance-api-secret: ""
coinbase-api-key: ""
coinbase-api-secret: ""
kraken-api-key: ""
kraken-api-secret: ""

View File

@@ -0,0 +1,8 @@
apiVersion: v1
kind: Secret
metadata:
name: minio-secret
type: Opaque
stringData:
root-user: "minio"
root-password: "minio123"

View File

@@ -0,0 +1,7 @@
apiVersion: v1
kind: Secret
metadata:
name: postgres-secret
type: Opaque
stringData:
password: "password"