AWS 502 safe restart decision response

When this happens

Use this when a restart might fix the symptom but could also interrupt users or jobs. Do not stop at the screen message; validate whether reload, restart, queue drain, or rollback is safer first.

Symptom checklist

• 502 appears repeatedly in the AWS UI or logs.
• CloudTrail, IAM policy, region, VPC, target health differs between successful and failed requests.
• The issue appears only after separating AWS API failure from application failure.
• It often follows deploys, permission changes, configuration edits, or data refreshes.

Likely causes

• 502 specifically changes the investigation surface for AWS: verify the exact failing object, route, user, and timestamp before applying the broader pattern.
• The proxy can connect to the edge but not to the intended backend target.
• Health checks use a different path, port, host header, or protocol than real traffic.
• FastCGI, app server, or container workers are down while the front server is alive.
• Timeouts are configured inconsistently across proxy, load balancer, and app layers.
• A deploy changed backend port or protocol without updating the routing layer.
• For the safe restart decision case, the first useful clue is whether reload, restart, queue drain, or rollback is safer.

First 1-minute checks

1. Write down the first failure time, latest change, affected user, path, and object ID.
2. Compare CloudTrail, IAM policy, region, VPC, target health for success and failure in the same window.
3. Test the hypothesis: upstream failure caused by unhealthy target, wrong backend port, proxy timeout, FastCGI handoff, or load balancer health check mismatch.
4. Classify this as safe restart decision: whether reload, restart, queue drain, or rollback is safer.
5. Capture current values before changing configuration.

First evidence

Treat 502 as a safe restart decision case. First collect evidence for whether reload, restart, queue drain, or rollback is safer.

Output examples

Normal output

Config test passes and active requests or jobs can be drained safely.

Failing output

The service is crash-looping, has invalid config, or still has in-flight critical jobs.

Output-to-action branches

• A restart might fix the symptom but could also interrupt users or jobs.
  Run config tests, inspect active jobs, choose reload when possible, and prepare rollback.
• The working and failing outputs differ.
  Act on the differing layer first: For 502, apply the fix only after reproducing the same condition and saving the before/after evidence for this exact code.
• Command output is normal but users still fail.
  Separate browser cache, cookies, permissions, and network location before declaring it fixed.

Do not do this

• Do not restart blindly before saving logs and validating configuration.
• Do not change multiple layers before identifying the failing layer.
• Do not delete production data, grant broad permissions, or disable security controls as a first response.

Evidence quality

Auto-generated operator draft: includes issue-specific causes, commands, output branches, and unsafe-action warnings. Official-source links and real incident validation are queued for enrichment.

Commands to run first

grep -R "502" ./logs

aws sts get-caller-identity

aws cloudtrail lookup-events --max-results 5

aws elbv2 describe-target-health --target-group-arn ARN

aws iam simulate-principal-policy --policy-source-arn ARN --action-names ACTION

curl -Iv http://127.0.0.1:3000/health

grep -R "upstream\|502\|503\|504\|proxy_fcgi" ./logs /var/log/nginx

systemctl status nginx php-fpm

systemctl status SERVICE --no-pager && journalctl -u SERVICE -n 100 --no-pager

Fix order

1. Record the full 502 message, failing URL, user, object ID, and latest change.
2. Collect issue-specific evidence for upstream failure caused by unhealthy target, wrong backend port, proxy timeout, FastCGI handoff, or load balancer health check mismatch.
3. Compare the failing case with a successful case before editing settings.
4. If this is the safe restart decision branch, Run config tests, inspect active jobs, choose reload when possible, and prepare rollback.
5. Re-check with the same command and URL, then record the normal output.

Actions by cause

• For 502, apply the fix only after reproducing the same condition and saving the before/after evidence for this exact code.
• Test the backend directly before changing the public proxy.
• Align health check path, host header, port, and expected status.
• Restart or scale the failing worker pool after checking logs.
• Tune proxy and app timeouts together.
• Record the working direct-backend command in the runbook.
• For the safe restart decision branch, Run config tests, inspect active jobs, choose reload when possible, and prepare rollback.

Evidence links

• AWS Elastic Load Balancing troubleshooting official
• AWS CloudFront troubleshooting official

Verification metadata

• operator-draft
• official-reference-linked
• 2026-07-23

Update queue

• Review cadence
  weekly-source-review
• Next enrichment
  Add one official-source check and one real output example for AWS 502.

Environment-specific checks

• Shared hosting, proxies, VPNs, or CDN layers can change AWS API failure from application failure results.
• Do not trust only the AWS UI; compare command output.
• Japanese hosting panels may show completion before DNS or SSL fully propagates.
• Test from both office and external networks.

Prevent it next time

• Store normal examples for CloudTrail, IAM policy, region, VPC, target health.
• Add IAM least privilege, region match, security groups, target groups to the release checklist.
• Keep recurring errors in the same note format.
• Split alerts by error rate, latency, certificates, disk, and permission changes.

Related notes

• AWS ALB 502 target response error
• S3 AccessDenied fix note
• CloudFront 403 fix note
• EC2 Connection refused fix note
• S3 AccessDenied fix note 2
• CloudFront 403 fix note 2
• EC2 Connection refused fix note 2
• S3 AccessDenied fix note 3