22 - Incident Playbooks

Dokumen ini berisi SOP teknis untuk insiden runtime utama.

1. Refresh Token Replay / Fingerprint Mismatch

Gejala:

banyak 401/403 di auth runtime

activity menunjukkan replay/fingerprint mismatch

Langkah:

cek logs dengan request_id terkait

cek endpoint /auth/activities/suspicious

verifikasi token rotation path di auth flow

force logout session terdampak jika perlu

dokumentasikan root cause

2. Tenant Context Mismatch

Gejala:

403 pada endpoint tenant

header tenant tidak match token tenant

Langkah:

verifikasi token tenant_id

verifikasi X-Tenant-ID pada client

cek middleware SetTenantContext dan ValidateTenantToken

instruksikan FE reset tenant context dan select ulang

3. Billing Webhook Failure

Gejala:

invoice/subscription state tidak update

dunning stuck

Langkah:

cek ingress webhook provider endpoint

cek forensic records (BillingWebhookForensicService)

pastikan signature/provider mapping valid

rerun reconciliation/report command jika dibutuhkan

4. Cluster Connection Issue

Gejala:

tenant runtime query error connection

provisioning taxonomies gagal

Langkah:

cek cluster registry (ListClustersCommand)

probe cluster (ProbeClusterCommand)

cek ClusterConnectionManager config mapping

re-run migration cluster jika perlu

Post-Incident Checklist

incident summary + root cause

preventive action item

test case tambahan bila ada gap

update dokumentasi playbook jika pattern baru ditemukan

22 - Incident Playbooks

1. Refresh Token Replay / Fingerprint Mismatch#

2. Tenant Context Mismatch#

3. Billing Webhook Failure#

4. Cluster Connection Issue#

Post-Incident Checklist#

1. Refresh Token Replay / Fingerprint Mismatch

2. Tenant Context Mismatch

3. Billing Webhook Failure

4. Cluster Connection Issue

Post-Incident Checklist