kenji3
Member
- Joined:
- Jul 2024
- Posts:
- 138
- From:
- Osaka, JP
My experience with OVHcloud in Singapore and Sydney has been solid. I run a small Kubernetes cluster for personal projects.
Latency to home: 45ms Singapore, 95ms Sydney. Uptime last 18 months: 99.97% by my own monitoring. Support tickets answered in 4-6 hours, though I have not needed many.
I have not used Contabo or RackNerd, so I cannot compare directly. I hope this helps somewhat.
conbini > datacenter snacks
mediaaustin
Member
Deliverability Nerd
- Joined:
- Jul 2024
- Posts:
- 276
- From:
- Austin, US
For what it's worth, most of you are evaluating the wrong thing entirely.
I run mail infrastructure for a mid-size SaaS. Every "budget" provider I've tested—Contabo, OVHcloud, RackNerd, Hetzner, the rest—ships with IP ranges so polluted that your SPF/DKIM/DMARC stack is meaningless. You're on blocklists before you send hello. I've documented seventeen IP ranges across these providers with SenderScore below 70. The "dedicated IP" upsell is often a recycled address that was spamming six hours ago.
If you're not checking MXToolbox and MultiRBL (https://mxtoolbox.com/blacklists.aspx) before you even apt update, you're already burned. The hosting quality barely matters when your deliverability is zero.
I know this isn't what PacketpunkUma asked. But I see "budget cloud" threads ignore email reputation constantly, and it costs people their businesses.
SPF, DKIM, DMARC — holy trinity ✉️
kenji3
Member
- Joined:
- Jul 2024
- Posts:
- 138
- From:
- Osaka, JP
99.97% over 18 months? What tool you using to measure that
Uptime Kuma running from my apartment on a Raspberry Pi 4, pinging every 60 seconds via ICMP. The 0.03% was a planned maintenance window in Singapore last March that took 23 minutes. I do not count that against them.
I also have Blackbox Exporter on a secondary monitor in a friend's flat in Tokyo. Both agree within 0.01%.
Not enterprise-grade monitoring, but honest.
conbini > datacenter snacks