What's a reasonable uptime for you?

I’m still hesitant to abandon Hetzner cloud in favour of something else because of the excellent uptime I’ve enjoyed wih them. I’ve long since rented a cheaper and more powerful VPS from a LowEnd provider, but there have been several brief outages and I don’t want to move everything there only to find myself regretting the choice when/if a big one comes.

For an online shop uptime is important, but it obviously depends on the use case. Under this aspect, I’ve been dissatisfied with digitalocean, OVH cloud was a nightmare, while Hetzner sports a 99.9865% uptime for me, including my reboots. I’d say 99,99% is ideal, 99,9% is disappointing but still somehow viable, anything less is just not for me.

So, what’s your uptime and what are your expectations?

1 Like

I’d expect at least 99.99% as a customer, but more is better.

On that, we aim for as high as possible (ideally, 99.999) but sometimes there are things that are out of control (upstream network, etc).

Ideally you want to monitor from multiple networks as a single network can cause false results - I’ve got a server that is only seeing 99.76% uptime according to Hetrix but every other monitoring service/own monitors is showing over 99.99% for the same server.

Clouviders is currently top of my “uptime” list with Frankfurt - showing 100% network uptime since monitoring started late last year.


Thanks for the tip about not trusting one single network (albeit with different locations). I usually monitor with both HetrixTools and UptimeRobot so I don’t have to worry in case only one goes down.
At work, one of the bosses unilaterally decided to outsource the website hosting to a company he trusts: uptime is currently 97.66% as opposed to 99.99% for another website that is still on our own servers.
I hear only good things about Clouvider…

97.66 is about 30 minutes a day (according to SLA & Uptime calculator: How much downtime corresponds to 97.66 % uptime) which isn’t great, but if you’re getting 99.98% on Hetzner that sounds like a good bet.

Yeah not sure what happens but when Hetrix things it’s down, it’s because all monitors think it’s down. I’ve checked with the provider/external monitoring/etc and nothing seems up, other than Hetrix. It’s settled down a lot now to be honest but I did consider querying it with Hetrix just incase.

Dom and his team are brilliant. Shameless self plug but if you need something that isn’t a dedi on Clouviders network feel free to get in touch :slight_smile:

As a customer, a few minutes of downtime a month isn’t going to make or break my business. As a provider, obviously you strive to offer months of 100% uptime, but realistically sometimes there is needed planned maintenance, sometimes there is unexpected urgent maintenance, sometimes ‘shit just breaks’ or fails in areas outside your control.

Best to just have a plan in place, regardless if you’re an end-user or a provider. No single person or single company, big or small, is immune to outages. (See: Cloudflare, AWS, Azure, etc outages)

Actually, is there any sort of large company that has any credible and verified proof of having a record for uptime and availability? I’m not sure if things like Facebook and Google outages limited to specific geographic regions would count towards the company as downtime or not. Would be neat to read or learn more about that, though.