pieter_rtm
Member
OP
- Joined:
- Jul 2024
- Posts:
- 195
- From:
- Rotterdam, NL
As said, we run ~400 containers across three providers. Backups happen. Restores? To be honest, last manual test was 14 months ago. Found one corrupted pg_dump last week. Lucky discovery.
Current stack:
- Contabo S3 for objects
- OVHcloud internal for configs
- Custom restic scripts, no verification hook
Looking for: tooling that actually restores and checks checksums, not just "backup completed" green dot. Budget: negligible. Pain tolerance: high.
Containers before it was cool
Carl
Member
Rack & Stack
- Joined:
- May 2024
- Posts:
- 227
- From:
- Chicago, US
Honest answer? Most people never test. I have 2U supermicro at Hetzner colo, gen 9 xeon, 380W idle. Backup to tape, LTO-8, ship pallet to Leaseweb secondary. Restore test? Quarterly, full bare metal. Takes 4 hours, 1.2kW power spike, I log watts. No automation. Just calendar reminder and shame.
Your 14 months is average. Worse than average.
visit twice: install and decom