Simple Domains Like Google.com are Getting `getaddrinfo EAI_AGAIN` Errors

Describe the bug I have about 10 domains I am watching, 7 of my own domains and 3 common domains, I recently installed piHole and I noticed 3 of my domains, 2 common and 1 of mine, started failing all the time. I looked into it and seems as these all are getting getaddrinfo EAI_AGAIN errors. I would understand if it was only my own domains but the common ones like Google and Reddit are weird. I looked at the logs within Portainer and I am not seeing any errors being thrown only Heartbeat <number>. The two common domains that are failing are www.google.com and www.reddit.com. If I click on the links within Uptime-Kuma I successfully get to the sites.

Expected behavior Better Logging?

Desktop (please complete the following information):

• Uptime Kuma Version: 1.0.6
• Using Docker?: Yes
• OS: Ubuntu 20.04 LTS
• Browser: Firefox/Chrome

Additional context Uptime-Kuma is in same Docker-Compose as piHole and have house router running through piHole now.

[antoine2tt]

Yes sure ! It could be network related.

Anyway, with the “retries” I setup, I can avoid theses kind of errors 👌

[rezzorix]

DuckDNS as DynDNS? That would be a simple DNS lookup… there is (should be) no throttling… most hiccups come from propagation taking time.

So basically you tell DuckDNS to change the A record for domain x every time your IP at home changes. Depending on how often your home IP changes this can cause hiccups with a DNS lookup because your A records need to propagate first.

So some DNS servers may still have your old home IP e.g. 1.2.3.4 listed, while the record is already updated to the new one which is e.g. 11.12.13.14. So if the DNS server you are using is taking a bit longer to update and still shows the old record - it will result in timeouts.