Member OP
Amsterdam, Los Angeles, Singapore. The $5 plan lets you pick any location, it's just storage and bandwidth that scale with price. Singapore latency to my Mumbai origin is ~80ms, good enough for HTTP check.
Actually got the Singapore one during their last stock drop, was €4.50 equivalent at the time.
Following this thread because I need to stop living like Carl3. I run a bakery in Portland and somehow also a WooCommerce site on a Leaseweb dedicated box in San Francisco.
Last month I found out my SSL expired because a customer called. A customer! Called my bakery! To tell me my website was broken!
I need whatever the opposite of "just dont check" is.
This is the exact arc I saw hundreds of times in support. The "it works until it doesn't" crowd always had the worst incidents because they had zero telemetry when things finally broke for real.
For you: Uptime Robot free tier is better than nothing, but honestly just do what RAJ did. $5 external watcher, Telegram to your phone. You will thank yourself.
HostHatch Singapore is in Digital Realty's SIN10, if I recall correctly. Not their own floor. This matters for GDPR purposes if your status page logs there, because Singapore's PDPA does not provide adequacy decision from the European Commission.
My static status page is hosted on a €3.50 Hetzner box in Nuremberg. Overkill for a few kilobytes, but the legal jurisdiction is predictable.
Member
I administrate servers for my association in Lyon since 2018. We use a simple solution: Nagios for monitoring, and for status page we have a WordPress with a green/red banner. It is not beautiful but it functions.
The GDPR question of Fritz48 is interesting. We use Matomo self-hosted for analytics, so we control the data. But I never thought about the status page itself as a legal problem.
This thread convinced me to finally set up proper monitoring. Been running bare metal at Leaseweb Frankfurt for 2 years, check it manually like a caveman.
One question: does Uptime Kuma do proper certificate expiry alerts? That's my actual nightmare scenario.
Member OP
Yes, built-in. You can set days-before threshold per monitor. I have mine at 14 days for Let's Encrypt, 30 for the wildcard from Sectigo.
Also does push monitors, which I use for the backup server that has no open ports. Game changer.
Member
Hermano I have Leaseweb in Amsterdam, same thing. Certificate expiry got me last year, customer in Mexico City couldn't pay for three hours on a Sunday. Never again.
Now I use acme.sh with DNS challenge and Uptime Kuma watches the cert. But the real fix was adding a second watcher in case the first dies. Two $5 HostHatch boxes, different cities.
I was productive before Kubernetes and I'm productive after, but these threads always make me tired. You people have 47 monitoring systems to monitor your monitoring systems.
My status page is a 15-line shell script that rsyncs an HTML file to a different provider if the first provider fails. Has worked since 2019. No JavaScript, no cookies, no database, no problem.
This is actually elegant. What provider do you use for the backup?
