How does remote access work

There is no true “Guest” account in Plex.
(What I mean by that: access per URL or per one-time use password for persons who are not a plex.tv account holder or something similar is not possible atm.)

So:

  1. the invitation email is only directing invited users to the plex.tv website where they are required to create an account for themselves. The server owner then shares his media (or a part of it) to these newly created plex.tv user accounts
  2. they type plex.tv or app.plex.tv and sign in with their personal user name and password
    (There is a different way available for Plex Pass owners, who can create “sub-accounts” of their own, for close family members who don’t need a full plex.tv account [‘managed users’].)
  3. They can use whichever Plex client app they prefer. (I am referring to “real” Plex clients. A generic DLNA-compatible client is not a plex client.)
  4. plex.tv serves several roles:
  • authentication of users
  • DynDNS for servers
    (there is no need to set up a domain name for your server yourself)
  • handing out cryptographic certificates to servers for secure connections between servers and clients
    (There is no need to get a certificate elsewhere. Fully trusted certs are included in all plex accounts, automatically.)
  • in some (rare) cases, when there is no way for the client to reach the server directly, plex.tv will serve as a ‘Relay’. This mode is restricted in bandwith and works only encrypted
  • in all other cases, communication between clients and servers is direct and plex.tv is only helping entitled clients to find the server(s) they have access to
  1. There is no fixed external address necessary. A periodically updated public IP is sufficient (which is used for ~ 80% of all domestic internet connections). Since plex.tv plays the role of DynDNS, it gets periodically notified by the server, which external IP it currently has and then updates the FQDN of the server accordingly. More details here
    There is currently a restriction that the server needs a publicly reachable IPv4 address in order to be reachable by all clients: If the server is behind a carrier-grade-NAT, it cannot be reached ‘directly’ (often used by mobile network operators or public/campus networks). Thus a direct access is not possible and the only way to use remote access is per ‘Relay’ connection. If the server only has a public IPv6 adddress, it can only be reached by clients which also have a fully working IPv6 connection (which is sadly still not the usual case).

  2. Yes. It works because the server is directly accessible and has a FQDN, which the mobile app is using to find the server. Unless the network operator interferes (some have found a way to identify Plex traffic and ‘throttle’ it), this is working pretty well, provided there is sufficient mobile bandwidth available.
    You simply ‘sign in’ the mobile plex app to your own plex.tv user account.

1 Like