Container IPs
Snapshot taken 2026-07-07 (prod). IPs may change after Incus bridge reconfiguration; the env/*.env profiles are the source of truth.
Since 2026-08-03 there are three physically-isolated Incus fleets — see Test & Dev fleets below. The table immediately under here is prod (Incus project default, bridge 10.53.1.0/24).
See the container-groups diagram on Incus Containers.
Prod fleet — project default, 10.53.1.0/24
| Container | IP |
|---|---|
| redis-queue | 10.53.1.10 |
| claude-orchestrata | 10.53.1.12 |
| janus-test | 10.53.1.15 |
| prolog-node-1 | 10.53.1.16 |
| prolog-node-2 | 10.53.1.17 |
| prolog-node-3 | 10.53.1.18 |
| orchestrata-arch | 10.53.1.20 |
| docker-build | 10.53.1.30 |
| orchgraph-coord-1 | 10.53.1.160 |
| orchgraph-coord-2 | 10.53.1.109 |
| orchgraph-coord-3 | 10.53.1.198 |
| orchgraph-data-1 | 10.53.1.128 |
| orchgraph-data-2 | 10.53.1.123 |
Decommissioned 2026-07-10: the old Memgraph cluster (
memgraph-1,memgraph-2,memgraph-3, HTTP HA controllermemgraph-ha, and standalonememgraph-db) was deleted after data was ported to orchgraph with a verified parity check. Those containers and their IPs no longer exist.Decommissioned 2026-07-27 (migration plan Step 5.1):
neo4j-db(10.53.1.11),neo4j-node-2(10.53.1.13),neo4j-node-3(10.53.1.14), andneo4j-node-4(10.53.1.19) were destroyed after the operational KG migrated to theorchkgdatabase on the orchgraph cluster..11/.13/.14/.19no longer respond (verified unreachable 2026-07-29). See KG migration to orchgraph and theKG_ROUTING_URInote on Environment Variables.Superseded 2026-08-03: the suite's old "dedicated test" Memgraph node
10.53.1.167lived on the prod bridge (only logically separate). It is replaced by the isolated test fleet node10.53.2.10below.
Test & Dev fleets (isolated)
Added 2026-08-03. dev and test are single infra servers (not clusters — prod is the only HA fleet), each in its own Incus project on its own bridge. Cross-env traffic is dropped by Incus network ACLs (orch-no-prod, orch-no-dev, orch-no-test), so from test you cannot reach prod or dev; the internet still works (package updates succeed). Endpoint source of truth: env/test.env, env/dev.env.
| Fleet | Incus project | Subnet | Redis | Memgraph (single) | Prolog |
|---|---|---|---|---|---|
| test | orchestrata-test |
10.53.2.0/24 |
10.53.2.11:6379 |
10.53.2.10:7687 (bolt) |
10.53.2.12 |
| dev | orchestrata-dev |
10.53.3.0/24 |
10.53.3.21:6379 |
10.53.3.20:7687 (bolt) |
10.53.3.22 |
The pytest suite is HARD-PINNED to the test fleet. tests/conftest.py loads env/test.env over .env and refuses to run if any resolved infra endpoint is on the prod subnet 10.53.1.x (override only with ORCHESTRATA_ALLOW_PROD_INFRA=1). Promote code through the fleets with the gated CLI scripts/promote_env.py (validate → test → promote --to prod, prod gated on a GREEN test-env run). Full detail: repo INFRASTRUCTURE.md § "Environment fleets".