On darwin an unconnected sendto runs the address and MAC policy checks
and the EndpointSecurity send check for every datagram, and while
xprotectd throttles it each send takes about 185us, which caps a
tunnel's upload and, through its ACKs, its download. A connected socket
runs those checks once, at connect, and it is the only UDP send that
reports the interface queue's flow control, as ENOBUFS.
Only the current direct remote of an established tunnel is eligible.
udp.EstablishedPeerConn is an optional Conn interface, implemented on
darwin; establishedRemotes refcounts remotes over the main hostmap and
tells each writer on the first and last reference: handshake completion
adds one, deleting or retiring a hostinfo removes it, SetRemote (roaming,
SetRemoteIfPreferred, control, ssh) moves it. Pending and relayed
hostinfos hold none, and a rehandshake's two hostinfos share one.
An eligible peer that Nebula sends NEBULA_CONNSOCK_RUN (64) datagrams to
within NEBULA_CONNSOCK_WINDOW (1s) gets a connected socket: SO_REUSEPORT,
bound to the source address the listener's route to the peer picks on
the listener's port, carrying the listener's IP_BOUND_IF. The peer sees
the same address and port as from the listener, and the socket's reader
delivers through the listener's reader. The listener itself never sets
SO_REUSEPORT (listen routines still do), so no other uid can bind the
port; with a specific listen.host and no routines connected sockets stay
off, since sharing an exact address needs SO_REUSEPORT on both.
All routines listening on one address share one set of connected
sockets: at most NEBULA_CONNSOCKS (8), one per peer, with opens from a
bucket that refills one per NEBULA_CONNSOCK_OPEN_EVERY (1s). A socket
closes when its peer stops being eligible, after NEBULA_CONNSOCK_IDLE
(30s) without traffic, on Rebind (of every routine) and Close, and on an
error that means the connected path is gone. Closing never runs under
the hostmap's or the eligibility lock; the peer can't reopen until the
close finishes. A send that gets ENOBUFS or EAGAIN waits within one
10ms budget shared by every writer since the last send that went
through, then drops; nothing waits in the netpoller without a deadline.
When nothing is eligible or open, a send costs one atomic load before
the listener path.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The goroutines Main starts stop only when its context does, and only a caller
holding the Control can arrange that. Cancelling solely on error left a config
test, which returns no Control and no error, with the lighthouse query worker
and a hostname resolver per dns named static host running for the life of the
process. Every embedder that validates a config in process rather than by
exec'ing pays for that on each config load, dnclient and the apple clients
included.
Co-authored-by: Nate Brown <nbrown.us@gmail.com>
* populate default Build version if missing
Use the Go module information built into the binary if the Build var
wasn't set during the build.
This means if you install via a specific tag, you get:
go install github.com/slackhq/nebula/cmd/nebula@v1.9.5
$ nebula -version
Version: 1.9.5
And if you install master, you get:
go install github.com/slackhq/nebula/cmd/nebula@master
$ nebula -version
Version: 1.9.5-0.20250408154034-18279ed17b10
* also default in the library
* cleanup
* upgrade to yaml.v3
The main nice fix here is that maps unmarshal into `map[string]any`
instead of `map[any]any`, so it cleans things up a bit.
* add config.AsBool
Since yaml.v3 doesn't automatically convert yes to bool now, for
backwards compat
* use type aliases for m
* more cleanup
* more cleanup
* more cleanup
* go mod cleanup
This used to work correctly because when the multiple routines work was
first added in #382, but an important part to discover the listen port
before opening the other listeners on the same socket was lost in this
PR: #653.
This change should fix the regression and allow multiple routines to
work correctly when listen.port is set to `0`.
Thanks to @rawdigits for tracking down and discovering this regression.
* Added firewall.rules.hash metric
Added a FNV-1 hash of the firewall rules as a Prometheus value.
* Switch FNV has to int64, include both hashes in log messages
* Use a uint32 for the FNV hash
Let go-metrics cast the uint32 to a int64, so it won't be lossy
when it eventually emits a float64 Prometheus metric.
By default, Nebula replies to packets it has no tunnel for with a `recv_error` packet. This packet helps speed up re-connection
in the case that Nebula on either side did not shut down cleanly. This response can be abused as a way to discover if Nebula is running
on a host though. This option lets you configure if you want to send `recv_error` packets always, never, or only to private network remotes.
valid values: always, never, private
This setting is reloadable with SIGHUP.
When running in `-test` mode, `tun` is set to nil. So we should move the
defer into the `!configTest` if block.
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x28 pc=0x54855c]
goroutine 1 [running]:
github.com/slackhq/nebula.Main.func3(0x4000135e80, {0x0, 0x0})
github.com/slackhq/nebula/main.go:176 +0x2c
github.com/slackhq/nebula.Main(0x400022e060, 0x1, {0x76faa0, 0x5}, 0x4000230000, 0x0)
github.com/slackhq/nebula/main.go:316 +0x2414
main.main()
github.com/slackhq/nebula/cmd/nebula/main.go:54 +0x540
