-
Notifications
You must be signed in to change notification settings - Fork 19
Open
Description
Just seen the Tailscale Services announcement and I wonder what would be the easiest way to integrate TSIDP with this to provide high availability.
My understanding is that we would have to:
- modify
tsidpto add a new flag with service reference and bind to that service on startup; - run a number of the
tsidpinstances across different locations within the same tailnet, using the same service reference.
If using svcid, the FQDN of TSIDP kind of does not make sense because everybody should be using the service FQDN, so accessing the individual instance of tsidp should be redirecting to the service URL automatically or giving error. Also, these tsidp nodes could be ephemeral, not even having any name?
Any thoughts? Is my understanding correct? Should we do something like this?
b-, Yethal, warnspread, swiftsudo, randName and 1 more
Metadata
Metadata
Assignees
Labels
No labels