Skip to content

How to harden a NATS container (and is nats:2.10-alpine safe for your messages?)

NATS is the nervous system of an event-driven stack: every service publishes to it and subscribes from it, and with JetStream it also persists the stream. A stock docker run nats:2.10-alpine keeps that broker behind a boundary weaker than the traffic deserves. Graded on IronClaw's seven-dimension containment scale, the default configuration scores 48 of 100, grade D (porous). Higher is safer. A broker exists to be connected to, so it cannot take --network=none the way a co-located database can. That sets an honest ceiling of 89 of 100, grade B, and the flags below reach it. Here are the exact gaps and fixes from the scan data.

Every number here comes from a read-only docker inspect of nats:2.10-alpine, the same data behind its isolation scorecard. No workload is executed. How scoring works →

Where the default configuration leaks

ironctl scan grades seven independent containment boundaries. On a default docker run nats:2.10-alpine, three fail and one warns:

Dimension Verdict Score What the scan found
Non-root user (uid != 0) ❌ FAIL 0/15 runs as root (uid 0); a container escape starts with host-uid 0
Dropped capabilities ❌ FAIL 4/20 default capability set retained (CAP_NET_RAW, CAP_MKNOD, and more)
Seccomp profile ✅ PASS 15/15 seccomp profile active
Network isolation / egress ⚠️ WARN 4/15 network=bridge: outbound egress is possible
Read-only root filesystem ❌ FAIL 0/10 root filesystem is writable
No docker.sock exposure ✅ PASS 15/15 no control socket mounted
No shared host namespaces ✅ PASS 10/10 no host PID/IPC/network sharing

The one that should worry you most is root. A NATS process that escapes as root escapes as root on the host, right next to every message flowing through it and, with JetStream, the persisted stream on disk. The default capability set and writable rootfs widen and entrench that foothold. The network dimension stays a WARN by design here, because a broker has to accept connections.

Harden it: the exact --fix remediation

ironctl scan my-nats --fix prints one remediation per failed dimension, then one hardened run. For nats:2.10-alpine:

  • --user 1000:1000 (Non-root user, +15): pin a non-root uid so an escape does not begin as host uid 0. Point any JetStream store directory at a volume this uid owns.
  • --cap-drop=ALL (Dropped capabilities, +16): drop every Linux capability; NATS needs none of the default set to serve its client and monitoring ports.
  • --read-only --tmpfs /tmp (Read-only rootfs, +10): make the root filesystem read-only and mount the JetStream data directory as an explicit writable volume. Removes the persistence surface.
  • Scoped private network (Network, held at WARN by design): a broker exists to be connected to, so --network=none would break it. Put NATS on a user-defined network scoped to just its publishers, subscribers, and cluster peers, with no default route out. The network dimension holds at a WARN (4 of 15). That is the honest ceiling.

Before and after

# Before: 48/100, grade D
docker run -d --name nats nats:2.10-alpine

# After: 89/100, grade B (scoped private network for clients and peers)
docker run -d --name nats-hardened \
  --user 1000:1000 \
  --cap-drop=ALL \
  --security-opt=no-new-privileges \
  --read-only --tmpfs /tmp \
  -v nats-data:/data \
  --network=messaging-internal \
  nats:2.10-alpine

Rescan: ironctl scan nats-hardened reports 89/100 grade B. A 41-point swing with no custom image build, just the right flags. The only dimension still short of full marks is the network (4 of 15), because a broker exists to be reached by its publishers and subscribers; network=none would score the last points but leave nothing able to connect. That is the honest ceiling for this role, and it is a long way from the default D.

Verify it on your own NATS

# install (Homebrew)
brew install ironsecco/ironclaw/ironclaw

# grade your running container, then print the fixes
ironctl scan my-nats
ironctl scan my-nats --fix

ironctl scan also reads a docker-compose.yml service or a Kubernetes manifest, so you can grade the NATS in your stack, not just a bare docker run.

Keep going