Auth times out every few days.

[turbo2ltr]

Which node are you reporting an issue on?

Google Calendar

What are the steps to reproduce?

Set up account and authenticate. Have "Include ongoing events" checked.

What happens?

Every few days (maybe 5-7 days) it stops working.

When it stops working, I get the following in the debug repeated frequently:

1/15/2022, 6:42:20 AMnode: 1e6abc4d626a18a9
msg : string[21]
"refresh token error: "
1/15/2022, 6:42:20 AMnode: 1e6abc4d626a18a9
msg : string[36]
"too many refresh attempts, giving up"
1/15/2022, 6:42:20 AMnode: 6bdb574f68bd7c34
msg : string[50]
"Error: Error: too many refresh attempts, giving up"

I then have to reenter my client ID/secret and auth again and it starts working for another few days.

My only guess is that because it's running in a docker instance, it requires some other setup I have not done?
The docker host is 10.0.0.252 so I set up a entry in my local windows hosts file (on the machine I access the nodeRed UI from) to map node-red.example.com to 10.0.0.252.
Seems to work for manual auth. Maybe there are some additional steps for docker instances?

What do you expect to happen?

I expect not to have to reauth the node every 5 days.

Please tell us about your environment:

• Node Red: 2.0.3
• Node-JS: 14.17.3
• NMP: 6.14.13
• OS/Platform: Node Red Docker image. https://registry.hub.docker.com/r/nodered/node-red/ (Alpine Linux v3.11 running in docker on a Synology DS1618)
• Browser: Brave 1.33.106 / Chromium: 96.0.4664.110

Thank you.

[turbo2ltr]

Additionally, when I update my credentials, the "Update" button at the top is still greyed out so there doesn't seem to be a way to save them. But if I hit cancel, then done (on the calendar node config screen) then force a change and redeploy, it works (for a few days)