DevOpsDays Amsterdam 2026 brought together platform engineers, SREs, and DevOps practitioners for two days of ignite talks and open spaces. UUID v7 appeared in sessions on observability, infrastructure-as-code, and pipeline tooling. See UUID in Terraform Infrastructure and UUID in Kubernetes for implementation details.
Session: UUID v7 as CI/CD correlation IDs
# GitHub Actions — inject UUID v7 correlation ID at workflow start
name: Deploy
on: [push]
jobs:
build:
runs-on: ubuntu-latest
env:
PIPELINE_ID: "" # set in first step
steps:
- name: Generate pipeline correlation ID
run: |
# Node.js one-liner: UUID v7 with timestamp prefix
PIPELINE_ID=$(node -e "const {v7}=require('uuid');console.log(v7())")
echo "PIPELINE_ID=$PIPELINE_ID" >> $GITHUB_ENV
echo "Pipeline ID: $PIPELINE_ID"
- name: Build
run: |
echo "Building with correlation ID: $PIPELINE_ID"
# Pass to build tools, test runners, deploy scripts
- name: Deploy
run: |
curl -X POST https://api.example.com/deploy \
-H "X-Correlation-ID: $PIPELINE_ID" \
-d '{"ref":"${{ github.sha }}"}'
The UUID v7 PIPELINE_ID encodes pipeline start time — parseInt(PIPELINE_ID.replace(/-/g,'').slice(0,12), 16) gives the exact millisecond the pipeline was triggered.
Session: Structured logging with UUID v7 trace IDs
// Structured log format with UUID v7 trace ID
import { v7 as uuidv7 } from 'uuid';
import pino from 'pino';
const logger = pino({
base: { service: 'order-service', env: process.env.NODE_ENV },
});
// Inject at request boundary
app.use((req, res, next) => {
req.traceId = req.headers['x-trace-id'] as string ?? uuidv7();
req.log = logger.child({ traceId: req.traceId });
res.setHeader('x-trace-id', req.traceId);
next();
});
// Every log line includes traceId — searchable by time range using UUID v7 prefix
req.log.info({ orderId: '...' }, 'Order created');
// → {"traceId":"018fbe3a-4c5d-7b12-...","orderId":"...","msg":"Order created"}
See UUID v7 in Microservices for the full correlation middleware pattern.
Session: Terraform reproducible infrastructure IDs
# Terraform — UUID v5 for deterministic infrastructure IDs
# Same name + namespace = same UUID across plan/apply runs
resource "null_resource" "workspace" {
triggers = {
# uuid v5 equivalent: deterministic from workspace name
workspace_id = uuidv5("dns", var.workspace_name)
}
}
# random_uuid for truly random infrastructure IDs (UUID v4)
resource "random_uuid" "deployment_id" {}
# Tag all resources with deployment UUID for cost attribution
locals {
common_tags = {
DeploymentId = random_uuid.deployment_id.result
Environment = var.environment
}
}
See UUID in Terraform Infrastructure for the full pattern including UUID v5 deterministic naming.
Further reading
- UUID v7 in Microservices — correlation IDs
- UUID in Kubernetes
- UUID in Terraform Infrastructure
- UUID in GCP Cloud Run
- UUID in AWS Lambda
External references
Frequently asked questions
How is UUID v7 used in CI/CD pipelines?
UUID v7 works as a build correlation ID injected at pipeline start and propagated through every step — build, test, deploy, verification. Because UUID v7 embeds a timestamp, build duration can be calculated from the pipeline ID and completion time without storing a separate startedAt field. Talks recommended injecting the UUID v7 as an environment variable (PIPELINE_ID) at pipeline trigger time and including it in all log output. See UUID v7 in Microservices — trace IDs.
Can UUID v7 replace UUID v4 for Kubernetes resource UIDs?
Kubernetes assigns its own resource UIDs using UUID v4 — you cannot change this. Application-level resource IDs (database primary keys for workload records, correlation IDs in labels/annotations) can use UUID v7. The talk recommended storing a UUID v7 correlation-id annotation on Pods so that log aggregators can join Pod logs to application trace spans by ID. See UUID in Kubernetes.