We have been using Contabo for our main VPS provider since 2020, and we used to recommend them for their affordable pricing on Asia region servers, However, we have recently switched to a new provider, and we wanted to share our experience and the reasons behind our decision.

Why We Chose Contabo

Back in 2020, where Teapot.py was growing, we needed a cheap VPS provider to kickstart our projects. Contabo offered a great deal with their VPS plans, which allowed us to get a decent server at that price point.

We originally started with a small VPS in the Europe region, then moved to the US region in 2021, a few months later they announced their new data centre in Singapore, we were excited to see how they would perform in the region with us paying a little extra for the Asia region server (still, it was cheaper than other providers). We had a good experience with them for a while and we were happy with the service, although we have small outages very rarely, we were able to get support from them relatively quickly via calls.

Downfall in 2024

In 2024, we started to notice some issues with Contabo’s service. The first sign was the increase in downtime, which started to become more frequent. Then, the support response times were getting longer, and the quality of support was declining. We had a few incidents where we had to wait for days to get a response from their support team, and when we did get a response, it was too generic.

One of the most significant incidents wes on May 24, 2024, we have noticed a significant latency in our Singapore VPS, a ssh connection took about 7 minutes to establish, and the server is just not responding to any request, it was down for 4 days. perform reboot via their panel just “There has been an error while processing your request.”

The only information we got from contacting support just tell me that there are “Network instability in Singapore and Japan Data centers”

Network instability in Singapore and Japan Data centers

Thanks for reaching out to Contabo customer support.

We apologize for the delayed response. Due to the popularity and high demand of our products and the related high number of incoming requests, we unfortunately could not get back to you earlier.

We have experienced some disruption in the Data Center where your servers are located. We are sorry for the inconvenience.

The issue has been resolved and all servers are now back online. 

If you have any further questions or need help, please do not hesitate to contact us.

Still, we decided to give them another chance, but the situation only got worse, the downtime of our VPS was getting longer, our VPS keep getting disconnected from network, random reboots every week or so.

htop downtime

We also encounter weird IO issues along with network problem which never happened in any other VPS provider, which made us believe that it was a virtualization or hardware issues. (Notes: the database is hosted on the same server)

IO issues and network disconnect

Even contabo itself admitted that their server are sometime freezing or required a reboot, However even after that, we still see the same issues.

Even with so many issues, contabo eventually provided us with a price change, with a 20% price increase.

Contabo price increase

finally, we decided to move our VPS to another provider.

uptime graph of our VPS

We eventually moved our VPS to netcup which is a german VPS provider, although they do not have data center in Asia, we still decided to get a root server in their german data center, and so far we are happy with the performance and support, they have similar pricing to Contabo, but they give lower CPU Steal%Steal% is a metric that indicates the percentage of time a virtual CPU is waiting for a real CPU while the hypervisor is servicing another virtual CPU, the lower the better. and better support response time.


Post-Migration

At 9th June 2025, we have cancelled all of our contabo services (objects storage and VPS), therefore vmi585337 is removed from our infrastructure, 90% of our services are now running on netcup with the remaining servics shutdown.