Files
AutopilotMonitor/scripts/Backup/setup-backup-container.ps1
T
Oliver Kieselbach d4034842ee feat(backend/backup): PR1 — critical-table backup feature (daily timer + manual GA trigger)
Adds the full backup pipeline that snapshots 15 critical tables (Admin/Tenant
config, rules, identities, audit, preview) into a logically-isolated blob
container, daily at 04:00 UTC and on operator demand. Restore endpoints follow
in PR2/PR3.

Architecture:
* Daily TimerTrigger + manual POST /api/global/backups/trigger (GA-only,
  fail-hard producer). HTTP returns 202 + jobId on enqueue, 500 if the queue
  write fails (and the BackupJobs row is rolled to Failed).
* BackgroundService queue worker (not QueueTrigger — matches the Analyze/
  IndexReconcile pattern that avoids the Functions-host connection-binding
  app-setting). Worker holds the maintenance lease for the whole run; service
  is intentionally lease-unaware so it stays unit-testable.
* BlobBackupStore wraps the new 'critical-table-backups' container with a
  shared maintenance-lease sentinel at '_lock/maintenance.lock' that
  serialises every backup or restore op across worker/timer/single-row paths.
* CriticalTableBackupService streams each table through CryptoStream(SHA256)
  into per-table NDJSON blobs, then writes manifest.json LAST as the
  durability anchor. Per-table failures degrade Outcome to Partial without
  aborting; per-run budget (50 min) → Skipped status.
* BackupJobsRepository persists the six-state machine
  (Queued/Running/Completed/Failed/Skipped/BlockedTerminal) with ETag-CAS
  updates. Enum→string mapping rejects unknown values on read so refactors
  surface loudly instead of silently shifting state.
* BackupJobWatchdog (30 min timer) heals stuck rows via a probe lease on the
  shared sentinel: Queued > 60 min → Failed, Running with stale heartbeat
  AND free lease → Failed (lease-busy means a live worker — leave alone).
* OpsEvents: distinct CriticalTableBackupPartial (Warning) vs Failed (Error)
  so operators can wire softer Telegram rules for partial captures.
* MaintenanceService.MonitoredPoisonQueues + OPS_EVENT_TYPES catalog
  registered (dual-register memory).

Test coverage (15 new tests, 1952/1952 backend xUnit green):
* BackupManifestJsonTests: enum-as-string (allowIntegerValues=false rejects
  numeric values), DictionaryKeyPolicy unset (Azure column names preserved
  case-faithfully), manifest field round-trip.
* BackupJobsRepositoryMappingTests: full entity↔domain round-trip, unknown
  Kind/State/BackupOutcome rejected, legacy BackupOutcome="Failed" rejected.
* CriticalTableBackupServiceTests: happy/partial/empty/budget-exceeded paths
  exercise the per-table loop with in-memory table + blob fakes.

Setup: scripts/Backup/setup-backup-container.ps1 — idempotent az-CLI bootstrap
for container + lifecycle rule (90 d delete on prefix). Soft-Delete and
Versioning are NOT applied automatically (storage-account-wide blast radius
— operator must validate per separate doc note).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-26 15:19:07 +02:00

93 lines
3.4 KiB
PowerShell

# setup-backup-container.ps1
#
# Idempotent setup for the critical-table backup feature (plan §PR1).
#
# Creates the 'critical-table-backups' container in the backend storage account
# and applies a lifecycle rule that deletes blobs older than 90 days under the
# 'critical-table-backups/' prefix. Run once per environment, then again any
# time the storage account is rebuilt.
#
# WARNING (plan §Wave15 Azure-Hinweis): Blob Versioning and Blob Soft-Delete
# are STORAGE-ACCOUNT settings (not container-scoped). Enabling them affects
# every container in the account (e.g. 'diagnostics', 'offboarding-state').
# This script does NOT enable them automatically — operator must validate the
# blast radius separately and run the corresponding `az storage account
# blob-service-properties update` command if desired.
[CmdletBinding()]
param(
[Parameter(Mandatory = $true)] [string] $StorageAccountName,
[Parameter(Mandatory = $true)] [string] $ResourceGroupName,
[string] $ContainerName = 'critical-table-backups',
[int] $RetentionDays = 90
)
$ErrorActionPreference = 'Stop'
Write-Host "==> Verifying Azure CLI login"
az account show --only-show-errors > $null
Write-Host "==> Ensuring container '$ContainerName' in account '$StorageAccountName'"
az storage container create `
--account-name $StorageAccountName `
--name $ContainerName `
--auth-mode login `
--only-show-errors `
--fail-on-exist false | Out-Null
Write-Host "==> Building lifecycle rule (delete > $RetentionDays days, prefix '$ContainerName/')"
$policyJson = @"
{
"rules": [
{
"enabled": true,
"name": "delete-critical-table-backups",
"type": "Lifecycle",
"definition": {
"actions": {
"baseBlob": {
"delete": { "daysAfterModificationGreaterThan": $RetentionDays }
}
},
"filters": {
"blobTypes": [ "blockBlob" ],
"prefixMatch": [ "$ContainerName/" ]
}
}
}
]
}
"@
$tmpFile = [System.IO.Path]::GetTempFileName()
try {
Set-Content -Path $tmpFile -Value $policyJson -Encoding UTF8
Write-Host "==> Applying lifecycle policy to storage account '$StorageAccountName'"
Write-Host " NOTE: this REPLACES the account's lifecycle policy. If other rules"
Write-Host " exist (e.g. for 'deletion-manifests'), merge them in manually."
az storage account management-policy create `
--account-name $StorageAccountName `
--resource-group $ResourceGroupName `
--policy "@$tmpFile" `
--only-show-errors | Out-Null
}
finally {
Remove-Item -Path $tmpFile -ErrorAction SilentlyContinue
}
Write-Host ""
Write-Host "==> Done."
Write-Host " Container ready: $ContainerName"
Write-Host " Lifecycle: delete blobs > $RetentionDays days under '$ContainerName/' prefix"
Write-Host ""
Write-Host "Operator follow-ups (NOT applied by this script):"
Write-Host " * If you also want Blob Versioning + Soft-Delete (defence-in-depth against"
Write-Host " operator mistakes), run separately and confirm the blast radius first:"
Write-Host " az storage account blob-service-properties update \\"
Write-Host " --account-name $StorageAccountName \\"
Write-Host " --resource-group $ResourceGroupName \\"
Write-Host " --enable-versioning true \\"
Write-Host " --enable-delete-retention true \\"
Write-Host " --delete-retention-days 14"
Write-Host " Both settings affect ALL containers in the account."