How to harden a Cassandra container (and is cassandra:5.0 safe for untrusted workloads?)¶
Short answer: a stock docker run cassandra:5.0 is not a boundary you should trust around
untrusted code or an untrusted network. Graded on IronClaw's seven-dimension containment scale,
the default configuration scores 48 of 100, grade D (porous). Higher is safer. Four runtime
flags take the same image to 100 of 100, grade A. This guide shows the exact gaps and the
exact fixes, straight from the scan data.
Every number here comes from a read-only
docker inspectofcassandra:5.0, 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
cassandra:5.0, four of them fail or warn:
| 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 |
For a wide-column database, the two that should worry you most are egress and root. A Cassandra process that can reach the network is one that can exfiltrate every keyspace it holds the moment it is compromised (a deserialization CVE, a poisoned UDF, a driver bug). And a root process that escapes the container escapes as root on the host.
Harden it: the exact --fix remediation¶
ironctl scan my-cassandra --fix prints one remediation per failed dimension, then assembles a
single copy-pasteable hardened run. For cassandra:5.0 the prescription is:
--cap-drop=ALL(Dropped capabilities, +16): drop every Linux capability; add back only what the workload provably needs. Cassandra itself needs none of the defaults.--user 65532:65532(Non-root user, +15): pin a non-root uid so an escape does not begin as host uid 0. Point the data directory at a volume this uid owns.--network=none(Network isolation, +11): a single-node Cassandra reached only by co-located services can cut host egress entirely; otherwise attach one internal network with no default route, notbridge. See the cluster note below.--read-only --tmpfs /tmp(Read-only rootfs, +10): make the root filesystem read-only and mount/var/lib/cassandraas an explicit writable volume. Removes the persistence surface.
The cluster note¶
A multi-node Cassandra ring uses the gossip protocol between nodes, so those nodes must reach each
other. --network=none fits a single-node instance that only its co-located app talks to. For a
real cluster, hold the network dimension at a WARN instead: attach a user-defined network scoped to
just the ring members and their client apps, with no default route out, so a compromised node cannot
call arbitrary internet addresses. That posture scores the 89/100, grade B ceiling the network
services on this set share; the 100/A run below is the single-node case.
Before and after¶
# Before: 48/100, grade D
docker run -d --name cassandra cassandra:5.0
# After: 100/100, grade A (single-node, private data volume)
docker run -d --name cassandra-hardened \
--user 65532:65532 \
--cap-drop=ALL \
--security-opt=no-new-privileges \
--read-only --tmpfs /tmp \
-v cassandra-data:/var/lib/cassandra \
--network=none \
cassandra:5.0
Rescan and the same seven dimensions all pass: ironctl scan cassandra-hardened reports
100/100 grade A. That is a 52-point swing from four one-line flags, no image rebuild.
Verify it on your own database¶
The grade above is the default image. Your deployment is what matters. Scan it in ten seconds:
# install (Homebrew)
brew install ironsecco/ironclaw/ironclaw
# grade your running container, then print the fixes
ironctl scan my-cassandra
ironctl scan my-cassandra --fix
ironctl scan also reads a docker-compose.yml service or a Kubernetes manifest, so you can grade
the Cassandra in your stack, not just a bare docker run.
Keep going¶
- All hardening guides →: every harden-a-container walkthrough, with grade deltas.
- cassandra:5.0 isolation scorecard →: the full dimension breakdown.
- Databases, ranked by isolation →: how Cassandra compares to Postgres, MongoDB, Redis, and the rest.
- How to harden a ClickHouse container →: the same walkthrough for the analytics side.
- Scan any container in 10 seconds →: the full
ironctl scanreference. - Run untrusted code in a real sandbox →: IronClaw wraps every AI-agent session in a gVisor/Kata boundary with
network=noneby default.