Opscraft
DevSecOps & secrets management · Lesson 6 of 7 · 3 min read

Compliance, vulnerability management and the bigger picture

Vulnerability management lifecycle

Discover (scanners across code, images, hosts, cloud) → prioritise (severity, exploitability, exposure, asset criticality) → remediate (patch, upgrade, rebuild, mitigate with config/WAF) → verify (re-scan) → report (SLA compliance, trends). Running workloads must be re-scanned continuously — an image clean at build time can have new CVEs next week.

Enlarge diagramDiscoverscan code,images, hostsPrioritiseexploitable?exposed?Remediatepatch, rebuild,mitigateVerifyre-scanReporttrends, SLAs met
Figure 1. The vulnerability management loop.
Swipe sideways to see all of it, or tap to open it full screen.

Example: Log4Shell-style response

  1. Identify exposure fast: query SBOMs / image scans for the vulnerable library across all running images.
  2. Mitigate immediately where possible: WAF rules, configuration flags, network egress restrictions.
  3. Patch: bump the dependency, rebuild, redeploy via the normal pipeline (fast because builds are automated).
  4. Verify with scans; hunt for exploitation in logs (GuardDuty, WAF, app logs).
  5. Postmortem: how fast could we answer "where are we vulnerable?" — improve SBOM coverage.
Enlarge diagramFind exposurequery SBOMsand scansMitigate nowWAF rules, configflags, egressPatchbump, rebuild,redeployVerify + huntre-scan; look forexploitationLearnimprove SBOMcoverage12345
Figure 2. Responding to a Log4Shell-style vulnerability. SBOMs turn "are we affected?" from days into minutes.
Swipe sideways to see all of it, or tap to open it full screen.

Compliance as code

Frameworks you'll hear in India-based product companies and BFSI: ISO 27001, SOC 2, PCI DSS (payments), RBI guidelines for regulated entities, India's DPDP Act for personal data. DevOps contributes evidence automatically: IaC with policy checks, CloudTrail/Config history, access reviews, change approvals in Git, vulnerability SLAs, encryption everywhere, audit logs retained immutably.

Threat modelling (lightweight)

For a new service: draw the data flow, list trust boundaries, use STRIDE (Spoofing, Tampering, Repudiation, Information disclosure, Denial of service, Elevation of privilege) to find threats, and map mitigations (authN/Z, TLS/mTLS, input validation, rate limits, logging, least privilege).

Enlarge diagramSpoofingpretending to be someoneTamperingchanging data or codeRepudiationdenying an actionInformation disclosureleaking dataDenial of servicemaking it unavailableElevation of privilegegaining more rights
Figure 3. STRIDE: six kinds of threat to check at every trust boundary.
Swipe sideways to see all of it, or tap to open it full screen.