Does OrbStack use ports 9200 and 9600 for itself?

[chiperific]

I've had an opensearch container in my docker-compose for a few months, but just today I ran into this error message:

Error response from daemon: driver failed programming external connectivity on endpoint opensearch-node1
(ba11079a968212c4cf95dde82a8371622b1da21ce2264f102ad99089917eb358):
Bind for 0.0.0.0:9200 failed: port is already allocated

With no containers running, here's the port list:

➜  docker-stack git:(main) ✗ lsof -Pi | grep OrbStack
OrbStack  57833 chipkragt    9u  IPv4  0x7f812a0cb791ea6      0t0   ICMP *:*
OrbStack  57833 chipkragt   10u  IPv6 0xc95e65fabfbb34f7      0t0 ICMPV6 *:*
OrbStack  57833 chipkragt   76u  IPv4 0xfe4a5d14d8ff78e9      0t0    TCP localhost:32222 (LISTEN)
OrbStack  57833 chipkragt   77u  IPv6 0x2978d901148d4dea      0t0    TCP localhost:32222 (LISTEN)
OrbStack  57833 chipkragt   78u  IPv4 0x7ca07e35031d91c3      0t0    TCP localhost:50303 (LISTEN)
OrbStack  57833 chipkragt  100u  IPv4 0xfc0b1d6196e5a70b      0t0    TCP localhost:50303->localhost:50304 (ESTABLISHED)
OrbStack  57833 chipkragt  102u  IPv4 0xcf7dcfd2bb4299c1      0t0    TCP *:9200 (LISTEN)
OrbStack  57833 chipkragt  103u  IPv4 0x1389b4e399ac8451      0t0    TCP *:9600 (LISTEN)
OrbStack  57833 chipkragt  104u  IPv6 0x75e8427727c6d780      0t0    TCP *:9200 (LISTEN)
OrbStack  57833 chipkragt  105u  IPv6 0x9e32d658a07271ec      0t0    TCP *:9600 (LISTEN)
OrbStack  57833 chipkragt  108u  IPv4 0x4a45a5ae35ff27d2      0t0    TCP 192.168.4.198:50305->216.24.57.4:443 (ESTABLISHED)

At first I thought OrbStack was "reserving" ports it saw in docker-compose files, but even after deleting my opensearch-node1 container, OrbStack still lists these ports on the lsof.

For context, here's the piece of my docker-compose.yml:

services:
  opensearch-node1:
    container_name: opensearch-node1
    extends:
      file: common-services.yaml
      service: opensearch
    volumes:
      - opensearch-data1:/usr/share/opensearch/data
    environment:
      - node.name=opensearch-node1
      - discovery.seed_hosts=opensearch-node1
      - cluster.initial_cluster_manager_nodes=opensearch-node1
    ports:
      - 9200:9200 # REST API
      - 9600:9600 # Performance Analyzer
    healthcheck:
      test: ["CMD", "curl", "-s", "http://localhost:9240/_cluster/health?wait_for_status=yellow&timeout=30s"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 30s

Opensearch uses ports 9200 and 9600 by default, and as I mentioned, I've had this working for months now with no issues. Just today I started getting this error.

[kdrag0n]

OrbStack doesn't use any fixed ports for itself. This specific error means that you have another running container using the port.

[chiperific]

I had no containers running when I ran lsof -Pi | grep OrbStack
Even after deleting my opensearch-node1 container, OrbStack still lists these ports on the lsof.