The solution on the OpenHab platform was to increase the timeout from 10 to 20 seconds. No idea if that will help on HE or if it can be implemented.
tagging @bobbyD
1 Like
I'm going to go out on the limb and say that api.ecobee.com was unavailable when hub software started, which resulted in an "unresolved" cached DNS entry. In upcoming version 2.2.4, DNS cache has a timeout, which enables hub software to properly handle NTP over unstable network. Same setting should help with Ecobee and other similar drivers as well, with no development effort required.
For current versions, the only way I can think of to address this is a hub reboot from diagnostics tool. That will clear DNS cache. If a reboot doesn't help, then I'm wrong about the root cause.
1 Like
This occurred before and after rebooting the hub for another reason.
I'm looking forward to 2.2.4 
1 Like
Now on 2.2.4.145 is this something that can be set? Can't find anything.
The DNS resolution timeout is on, and there's no way to turn it off.
Ahh this is DNS only, thought it was also possible to adjust for slow http requests. However on closer inspection of my original post, it does appear to be a DNS error. I recently switched the default DNS in the ORBI router and have not seen this for a while.
Unfortunatley i just got this on the ecobee app, but signin looks OK
app:22802020-11-20 17:41:27.895 errorException getEcobeeDevices: 500, e:groovyx.net.http.HttpResponseException: Internal Server Error, data:[status:[code:14, message:Authentication token has expired. Refresh your tokens. ]]
Thank you for all or your efforts getting 2.2.4 done. I waited until the dashboard links were fixed in 145. I'm almost all zigbee on a C-4 and it's running OK .
1 Like
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.
