Arkhivio Backup Framework

1. Scan

2. Upload

3. S3 Storage

4. Audit

5. Restore

Files scanned

Uploaded

0 B

Bytes transferred

—

Integrity

Pipeline

🔍

Scanner

scanner.py — CRC32 + mtime/size

config.yaml

app.log

data/records.db

reports/q1.pdf

reports/q2.pdf

media/logo.png

static/style.css

keys/cert.pem

↑

Uploader

uploader.py — 4 workers · token-bucket throttle

config.yaml

app.log

data/records.db

reports/q1.pdf

reports/q2.pdf

media/logo.png

static/style.css

keys/cert.pem

s3 PUT

s3 GET

☕

S3 Bucket — prod-backups

webserver-01/{absolute/path}

x-amz-meta-crc32 · x-amz-meta-compressed

↓

Restore

restore_onlys3.py — no MongoDB required

✓ /restore/config.yaml — CRC32 match

✓ /restore/app.log — CRC32 match

✓ /restore/data/records.db — CRC32 match

✓ /restore/reports/q1.pdf — CRC32 match

✓ /restore/reports/q2.pdf — CRC32 match

✓ /restore/media/logo.png — CRC32 match

✓ /restore/static/style.css — CRC32 match

✓ /restore/keys/cert.pem — CRC32 match

Audit — 3-check integrity verification

📄

latest_file

At least one snapshot document exists in MongoDB

pending

🔑

s3_metadata

x-amz-meta-crc32 on the latest S3 object matches MongoDB

pending

💾

restore & verify

Download a random file, recompute CRC32, compare with stored checksum

pending

Structured JSON log output

backup-framework — structured logs