tinaworm
Member
OP
- Joined:
- Jun 2024
- Posts:
- 84
- From:
- Riga, Latvia
- objective: migrate 3 production nodes from RackNerd to OVHcloud during their december flash sale (50% recurring)
- planned timeline: 00:00-06:00 UTC to minimize downtime
- actual timeline:
- 00:00: OVHcloud sale site unreachable
- 00:15: confirmed 502 errors, not local dns
- 00:30: began evaluating alternatives
- 01:15: discovered Contabo running concurrent sale, 60% first year
- 02:00: provisioned test instance on Contabo
- 04:00: completed migration to Contabo instead
- 06:00: original OVHcloud site restored, sale extended 24h
- result: unintended provider switch, lower final cost
- lesson: concurrent sale monitoring warranted
-t
worst bandwidth, best stories
danfra
Member
- Joined:
- Jun 2024
- Posts:
- 159
- From:
- Frankfurt, DE
OVHcloud got hit 340 Gbps at 23:58 UTC. Volumetric attack, UDP reflection. Their anycast didn't propagate fast enough, single PoP saturated. No scrubbing layer advertised.
Contabo runs proper scrubbing with 7 filtering layers. Anycast + RTBH. You upgraded mitigation without planning to.
mitigated 800Gbps before breakfast
zoekyo
Member
Less Is More
- Joined:
- May 2024
- Posts:
- 102
- From:
- Kyoto, JP
Simple is better
Planned chaos became unplanned calm
The sale you needed was not the sale you sought
rm -rf / --no-preserve-root ☯
SingaporeRep
Member
Benchmark Addict
- Joined:
- Jul 2024
- Posts:
- 227
- From:
- Singapore, SG
Different, Contabo sale also have queue system but their object storage handle better. OVHcloud use cheap bare metal, same same but different. You choose good one.
Happy migrate
fio, iperf, geekbench. results or gtfo.
