for chain infra teams · one cron line

Encrypted offsite backups for chain indexers and validators.

Your indexer database took weeks to build. Back it up in one cron line: encrypted before it leaves your machine, proven retrievable continuously, paid in USDC.

Start free · 12 GB, no card Full architecture

# nightly indexer backup

$ pg_dump $DB | zstd | age -r $BACKUP_KEY | rclone rcat obsideo:backups/indexer-$(date +%F).sql.zst.age

this exact pipeline verified end-to-end against the production gateway · 2026-07-09

what happens to your backup live network numbers · 2026-07-18

01 encrypt in your pipeline (age/gpg) or SDK AES-256-GCM, before upload 02 replicate 3 replicas per object (network target) 03 challenge merkle proof on every replica · every 4 h 04 settle providers paid in USDC only for proofs passed network, 2026-07-18: 181,769 merkle proofs passed · 49,490 in one cycle

you find out your backup is retrievable now, not during a disaster

why teams like yours use it

●Encrypted before it leaves your machine.

Encrypt in your own pipeline (age, gpg, your E2EE layer); the S3 gateway is a passthrough that never holds your keys. Or use the SDK and its client-side AES-256-GCM. Encrypt before upload and there is no plaintext on our infrastructure to hand over.

●Proof, not promises.

Every object is challenged with cryptographic proofs on a continuous cycle, targeting roughly every 4 hours. Proofs do fail and the pass rate varies by node, which is why replication is 3x and providers are paid per passed proof. You find out your backup is retrievable now, not during a disaster.

●Fast when it matters.

Hybrid network: datacenter-grade hot tier for quick uploads and restores, distinct distributed provider nodes for redundancy (operator and facility independence is not guaranteed). Restore speed of a cloud, survivability of a network.

pricing

$0.50 per TB-day · egress included · billed on logical (not replicated) size.

50 GB compressed daily dump × 30-day retention ≈ 1.5 TB stored ≈ 1.5 TB-days/day ≈ $22.50/month

Settlement in USDC (Noble). No card, no KYC-wall, no cloud account.

restores

Restoring is the same S3 call in reverse.

$ rclone cat obsideo:backups/indexer-2026-07-08.sql.zst.age | age -d -i key.txt | zstd -d > indexer.sql

restore pipeline verified end-to-end against the production gateway · 2026-07-09

Coordinator-outage evacuation: your recovery bundle (pre-fetched download authorizations plus a cached provider directory) lets the SDK pull your ciphertext straight from providers while those authorizations remain valid. Durable coordinator-independent recovery is in active development.

live today

Chain-infrastructure backup workloads in production, including a full chain indexer's PostgreSQL database, compressed and encrypted by the operator, backed up daily to Obsideo.

get started

Start now, talk later.

Free tier is self-serve: 12 GB, no card, credentials in your terminal in about a minute (quickstart). For production indexer workloads, USDC invoicing, or anything bespoke: email regan@obsideo.io or DM on Telegram.

Start free · 12 GB Full architecture