Skip to content

Surface the Problem

Surface the Problem

Before we fix anything, let's measure exactly what's wrong with the starting image. After the setup patch, our Dockerfile starts with:

FROM node:18 AS base

node:18 is a full, unoptimised image — and it's where most teams start by default. Let's see what that costs us.

Quick vulnerability overview

docker scout quickview catalog-service --org <YOUR_ORG>

You'll see something like this:

  Target             │  catalog-service:latest  │    2C    26H    25M   122L     4?
    digest           │  360db4f00cbd            │
  Base image         │  node:18                 │    2C    26H    25M   122L     4?
  Updated base image │  node:25-slim            │    0C     1H     2M    24L
                     │                          │    -2    -25    -23    -98     -4

Scout is already pointing at the answer: one FROM line change from node:18 to node:25-slim would eliminate 2 critical and 25 high CVEs immediately.

Policy evaluation

Scout doesn't just count CVEs — it evaluates your image against a set of organisational policies:

docker scout policy catalog-service --org <YOUR_ORG>

Policy status  FAILED  (4/7 policies met)

  Status │                     Policy                     │           Results
─────────┼────────────────────────────────────────────────┼──────────────────────────────
  ✓      │ Default non-root user                          │
  !      │ AGPL v3 licenses found                         │    3 packages
  !      │ Fixable critical or high vulnerabilities found │    2C    26H     0M     0L
  ✓      │ No high-profile vulnerabilities                │
  ✓      │ No outdated base images                        │
  !      │ Unapproved base images found                   │    1 deviation
  ✓      │ Supply chain attestations                      │    0 deviations

4/7 policies failing. This is the reactive "scan and fix" cycle most teams know all too well:

  1. Developers pull a base image, build, ship
  2. CI scanner finds 2 critical and 26 high CVEs
  3. Developers spend 3 days researching fixes
  4. Rebuild — 189 vulnerabilities still remain
  5. Cycle repeats; security blocks deployment

The next 8 sections walk through fixing this proactively, one best practice at a time — and then migrating to DHI to eliminate the cycle altogether.