Hi - I am getting repeated EAI_AGAIN errors on a flow that requests data from an external website. This appears to have started in the past 2 weeks - prior to that, this flow had worked flawlessly for over a year.
My understanding is that this error is related to a DNS lookup error but searching here gave me results relating to installation on nodes but not what I am using. Is there anyone else facing this issue?
I have restarted Node-RED, restarted the Pi5 but the error persists. It is sporadic but when it happens, it could happen 5-10 times in a row (my request goes out every 5 minutes) and then stop for some time. There appears to be no pattern to this and browsing the internet from other devices is working normally.
Currently, the fiber modem provided by my ISP does not allow me to override the DNS server and I have to do it at the client level. My internal wifi router is in bridge mode because I have quite a few IP reservations and didn't want to mess with changing my set up.
Would changing the DNS server on the Pi possibly solve this or is there something going on with the http request node? This is the only periodic external call that I am making, so don't have something to compare this to.
Thanks - I don’t think it’s a problem with the server as there are people in the Home Assistant world using that particular server for Tesla integration. I also exchanged messages with the developer who said that I was the only one with this issue. It probably has something to do with my ISP if no one in the NR community has reported anything. I will try specifying the DNS server on the RPi and see if that resolves the issue.
I didn't phrase my reply well. I meant that it will prove that it is a networking issue, not a node red issue. If you also ping, for example, google.com, you will see whether it specific to that server.
What DNS server are you using? Try switching it to 8.8.8.8
The ATT fiber modem will not let me change the dns, so I had to do it at the RPi level. Just did this a few minutes ago (verified that it was using the dns specified on the RPi), so will see what if that fixes it.
It seems that AT&T is having issues with its DNS servers - there is quite a bit of online chatter about this. My brother (who is also in Dallas) who just switched to AT&T fiber also confirmed that he was having issues and that using a VPN made the problem go away.
No need for that, just tell the PC or Pi or whatever to use a different DNS. You can specify two, so there is a fallback. You can use 1.1.1.1 as the primary (which is Cloudflare's) and 8.8.8.8 as fallback (which is Google's).
