You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
tailscale/derp
Brad Fitzpatrick 3672f29a4e net/netns, net/dns/resolver, etc: make netmon required in most places
The goal is to move more network state accessors to netmon.Monitor
where they can be cheaper/cached. But first (this change and others)
we need to make sure the one netmon.Monitor is plumbed everywhere.

Some notable bits:

* tsdial.NewDialer is added, taking a now-required netmon

* because a tsdial.Dialer always has a netmon, anything taking both
  a Dialer and a NetMon is now redundant; take only the Dialer and
  get the NetMon from that if/when needed.

* netmon.NewStatic is added, primarily for tests

Updates tailscale/corp#10910
Updates tailscale/corp#18960
Updates #7967
Updates #3299

Change-Id: I877f9cb87618c4eb037cee098241d18da9c01691
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
7 months ago
..
derphttp net/netns, net/dns/resolver, etc: make netmon required in most places 7 months ago
testdata
README.md
derp.go
derp_client.go derp: optimize another per client field alignment 10 months ago
derp_server.go cmd/derper, derp, tailcfg: add admission controller URL option 9 months ago
derp_server_default.go
derp_server_linux.go
derp_test.go all: use Go 1.22 range-over-int 7 months ago
dropreason_string.go

README.md

DERP

This directory (and subdirectories) contain the DERP code. The server itself is in ../cmd/derper.

DERP is a packet relay system (client and servers) where peers are addressed using WireGuard public keys instead of IP addresses.

It relays two types of packets:

  • "Disco" discovery messages (see ../disco) as the a side channel during NAT traversal.

  • Encrypted WireGuard packets as the fallback of last resort when UDP is blocked or NAT traversal fails.

DERP Map

Each client receives a "DERP Map" from the coordination server describing the DERP servers the client should try to use.

The client picks its home "DERP home" based on latency. This is done to keep costs low by avoid using cloud load balancers (pricey) or anycast, which would necessarily require server-side routing between DERP regions.

Clients pick their DERP home and report it to the coordination server which shares it to all the peers in the tailnet. When a peer wants to send a packet and it doesn't already have a WireGuard session open, it sends disco messages (some direct, and some over DERP), trying to do the NAT traversal. The client will make connections to multiple DERP regions as needed. Only the DERP home region connection needs to be alive forever.

DERP Regions

Tailscale runs 1 or more DERP nodes (instances of cmd/derper) in various geographic regions to make sure users have low latency to their DERP home.

Regions generally have multiple nodes per region "meshed" (routing to each other) together for redundancy: it allows for cloud failures or upgrades without kicking users out to a higher latency region. Instead, clients will reconnect to the next node in the region. Each node in the region is required to to be meshed with every other node in the region and forward packets to the other nodes in the region. Packets are forwarded only one hop within the region. There is no routing between regions. The assumption is that the mesh TCP connections are over a VPC that's very fast, low latency, and not charged per byte. The coordination server assigns the list of nodes in a region as a function of the tailnet, so all nodes within a tailnet should generally be on the same node and not require forwarding. Only after a failure do clients of a particular tailnet get split between nodes in a region and require inter-node forwarding. But over time it balances back out. There's also an admin-only DERP frame type to force close the TCP connection of a particular client to force them to reconnect to their primary if the operator wants to force things to balance out sooner. (Using the (*derphttp.Client).ClosePeer method, as used by Tailscale's internal rarely-used cmd/derpprune maintenance tool)

We generally run a minimum of three nodes in a region not for quorum reasons (there's no voting) but just because two is too uncomfortably few for cascading failure reasons: if you're running two nodes at 51% load (CPU, memory, etc) and then one fails, that makes the second one fail. With three or more nodes, you can run each node a bit hotter.