Alexander Wong

@bradfitz @Tailscale Nvm I RTFMed the issue. Glad to see an upcoming resolution, thanks!

@bradfitz @Tailscale Oh the ts client maintains per host routes for every other machine on the tailnet? Not just a cidr block?

@Tailscale is it possible that an api call to a webservice which in turn calls the ts api to remove a machine from a tailnet would cause a net::ERR_NETWORK_CHANGED error to occur on other machines on that tailnet, causing that original api call’s server response to be interrupted

@maisem_ali @Tailscale I guess definition of key here is concatenation of prefix-id-secret

@maisem_ali @Tailscale Works! Sorry, RTFM error. Still might be nice to include expiry as a resp hdr so it can be managed w middleware. Is the :keyid always guaranteed to be split(‘-‘)[1] parseable from the key itself to match which key is used? Would have assumed keys should be treated as opaque.

Would be real nice if the @Tailscale API had an endpoint that would tell you when the API key you’re using is due to expire so clients could help drive necessary renewal processes

@maisem_ali @Tailscale Thanks for the reply. I mean API keys, not individual machine/device keys

@ablewhiskey @Tailscale We do have an API to look up expiry for a specific key #get-apiv2tailnettailnetkeyskeyid---get-information-for-a-specific-key" target="_blank" rel="nofollow noopener">github.com/tailscale/tail…

@bradfitz Where “drive necessary renewal processes” is like hitting a Slack webhook to remind #tailscale-admins to get off their butt and renew the API key ahead of the expiry 😉

@ablewhiskey Seems doable.

Pandemic poetry cuts deep

Mune

Moisture

J̶e̶a̶n̶o̶m̶i̶c̶s̶ ̶m̶a̶c̶r̶o̶s̶c̶o̶p̶y̶

Lune

Moon