This enables us to mark nodes as relay capable or not. We don't actually
do that yet, as we haven't established a relay CapVer.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
It might complete, interrupting it reduces the chances of establishing a
relay path.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
This is simply for consistency with relayManagerInputEvent(), which
should be the sole launcher of runLoop().
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
relayManager can now hand endpoint a relay epAddr for it to consider
as bestAddr.
endpoint and Conn disco ping/pong handling are now VNI-aware.
Updates tailscale/corp#27502
Updates tailscale/corp#29422
Signed-off-by: Jordan Whited <jordan@tailscale.com>
This commit adds a new type to magicsock, epAddr, which largely ends up
replacing netip.AddrPort in packet I/O paths throughout, enabling
Geneve encapsulation over UDP awareness.
The conn.ReceiveFunc for UDP has been revamped to fix and more clearly
distinguish the different classes of packets we expect to receive: naked
STUN binding messages, naked disco, naked WireGuard, Geneve-encapsulated
disco, and Geneve-encapsulated WireGuard.
Prior to this commit, STUN matching logic in the RX path could swallow
a naked WireGuard packet if the keypair index, which is randomly
generated, happened to overlap with a subset of the STUN magic cookie.
Updates tailscale/corp#27502
Updates tailscale/corp#29326
Signed-off-by: Jordan Whited <jordan@tailscale.com>
Fragmented datagrams would be processed instead of being dumped right
away. In reality, thse datagrams would be dropped anyway later so there
should functionally not be any change. Additionally, the feature is off
by default.
Closes#16203
Signed-off-by: Claus Lensbøl <claus@tailscale.com>
Our conn.Bind implementation is updated to make Send() offset-aware for
future VXLAN/Geneve encapsulation support.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
relayManager is responsible for disco ping/pong probing of relay
endpoints once a handshake is complete.
Future work will enable relayManager to set a relay endpoint as the best
UDP path on an endpoint if appropriate.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
This type improves code clarity and reduces the chance of heap alloc as
we pass it as a non-pointer. VNI being a 3-byte value enables us to
track set vs unset via the reserved/unused byte.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
CallMeMaybeVia reception and endpoint allocation have been collapsed to
a single event channel. discoInfo caching for active relay handshakes
is now implemented.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
Add new rules to update DNAT rules for Kubernetes operator's
HA ingress where it's expected that rules will be added/removed
frequently (so we don't want to keep old rules around or rewrite
existing rules unnecessarily):
- allow deleting DNAT rules using metadata lookup
- allow inserting DNAT rules if they don't already
exist (using metadata lookup)
Updates tailscale/tailscale#15895
Signed-off-by: Irbe Krumina <irbe@tailscale.com>
Co-authored-by: chaosinthecrd <tom@tmlabs.co.uk>
The event loop removes the need for growing locking complexities and
synchronization. Now we simply use channels. The event loop only runs
while there is active work to do.
relayManager remains no-op inside magicsock for the time being.
endpoints are never 'relayCapable' and therefore endpoint & Conn will
not feed CallMeMaybeVia or allocation events into it.
A number of relayManager events remain unimplemented, e.g.
CallMeMaybeVia reception and relay handshaking.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
relayManager will eventually be responsible for handling the allocation
and handshaking of UDP relay server endpoints.
relay servers are endpoint-independent, and Conn must already maintain
handshake state for all endpoints. This justifies a new data structure
to fill these roles.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
Conn.handleDiscoMessage() now makes a distinction between relay
handshake disco messages and peer disco messages.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
Conn.sendDiscoMessage() now verifies if the destination disco key is
associated with any known peer(s) in a thread-safe manner.
Updates #15844
Signed-off-by: Jordan Whited <jordan@tailscale.com>
When an event bus is plumbed in, use it to subscribe and react to port mapping
updates instead of using the client's callback mechanism. For now, the callback
remains available as a fallback when an event bus is not provided.
Updates #15160
Change-Id: I026adca44bf6187692ee87ae8ec02641c12f7774
Signed-off-by: M. J. Fromberger <fromberger@tailscale.com>
In preparation for adding more parameters (and later, moving some away), rework
the portmapper constructor to accept its arguments on a Config struct rather
than positionally.
This is a breaking change to the function signature, but one that is very easy
to update, and a search of GitHub reveals only six instances of usage outside
clones and forks of Tailscale itself, that are not direct copies of the code
fixed up here.
While we could stub in another constructor, I think it is safe to let those
folks do the update in-place, since their usage is already affected by other
changes we can't test for anyway.
Updates #15160
Change-Id: I9f8a5e12b38885074c98894b7376039261b43f43
Signed-off-by: M. J. Fromberger <fromberger@tailscale.com>
Although, at the moment, we do not yet require an event bus to be present, as
we start to add more pieces we will want to ensure it is always available. Add
a new constructor and replace existing uses of new(tsd.System) throughout.
Update generated files for import changes.
Updates #15160
Change-Id: Ie5460985571ade87b8eac8b416948c7f49f0f64b
Signed-off-by: M. J. Fromberger <fromberger@tailscale.com>
This adds netx.DialFunc, unifying a type we have a bazillion other
places, giving it now a nice short name that's clickable in
editors, etc.
That highlighted that my earlier move (03b47a55c7) of stuff from
nettest into netx moved too much: it also dragged along the memnet
impl, meaning all users of netx.DialFunc who just wanted netx for the
type definition were instead also pulling in all of memnet.
So move the memnet implementation netx.Network into memnet, a package
we already had.
Then use netx.DialFunc in a bunch of places. I'm sure I missed some.
And plenty remain in other repos, to be updated later.
Updates tailscale/corp#27636
Change-Id: I7296cd4591218e8624e214f8c70dab05fb884e95
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
The regular android app constructs its own wgengine with
additional FFI shims, so this default codepath only affects
other handcrafted buids like tsnet, which do not let the
caller customize the innards of wgengine.
Android >=14 forbids the use of netlink sockets, which makes
the standard linux router fail to initialize.
Fixes#9836
Signed-off-by: David Anderson <dave@tailscale.com>
fixestailscale/corp#27506
The source address link selection on sandboxed macOS doesn't deal
with loopback addresses correctly. This adds an explicit check to ensure
we return the loopback interface for loopback addresses instead of the
default empty interface.
Specifically, this allows the dns resolver to route queries to a loopback
IP which is a common tactic for local DNS proxies.
Tested on both macos, macsys and tailscaled. Forwarded requests to
127/8 all bound to lo0.
Signed-off-by: Jonathan Nobels <jonathan@tailscale.com>
This adds a new helper to the netmon package that allows us to
rate-limit log messages, so that they only print once per (major)
LinkChange event. We then use this when constructing the portmapper, so
that we don't keep spamming logs forever on the same network.
Updates #13145
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I6e7162509148abea674f96efd76be9dffb373ae4
Ensure that the src address for a connection is one of the primary
addresses assigned by Tailscale. Not, for example, a virtual IP address.
Updates #14667
Signed-off-by: Fran Bull <fran@tailscale.com>
If we get a packet in over some DERP and don't otherwise know how to
reply (no known DERP home or UDP endpoint), this makes us use the
DERP connection on which we received the packet to reply. This will
almost always be our own home DERP region.
This is particularly useful for large one-way nodes (such as
hello.ts.net) that don't actively reach out to other nodes, so don't
need to be told the DERP home of peers. They can instead learn the
DERP home upon getting the first connection.
This can also help nodes from a slow or misbehaving control plane.
Updates tailscale/corp#26438
Change-Id: I6241ec92828bf45982e0eb83ad5c7404df5968bc
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
When in tun mode on Linux, AllowedIPs are not automatically added to
netstack because the kernel is responsible for handling subnet routes.
This ensures that virtual IPs are always added to netstack.
When in tun mode, pings were also not being handled, so this adds
explicit support for ping as well.
Fixestailscale/corp#26387
Change-Id: I6af02848bf2572701288125f247d1eaa6f661107
Signed-off-by: Adrian Dewhurst <adrian@tailscale.com>
Cubic performs better than Reno in higher BDP scenarios, and enables the
use of the hystart++ implementation contributed by Coder. This improves
throughput on higher BDP links with a much faster ramp.
gVisor is bumped as well for some fixes related to send queue processing
and RTT tracking.
Updates #9707
Updates #10408
Updates #12393
Updates tailscale/corp#24483
Updates tailscale/corp#25169
Signed-off-by: James Tucker <james@tailscale.com>
Originally identified by Coder and documented in their blog post, this
implementation differs slightly as our link endpoint was introduced for
a different purpose, but the behavior is the same: apply backpressure
rather than dropping packets. This reduces the negative impact of large
packet count bursts substantially. An alternative would be to swell the
size of the channel buffer substantially, however that's largely just
moving where buffering occurs and may lead to reduced signalling back to
lower layer or upstream congestion controls.
Updates #9707
Updates #10408
Updates #12393
Updates tailscale/corp#24483
Updates tailscale/corp#25169
Signed-off-by: James Tucker <james@tailscale.com>
The gVisor RACK implementation appears to perfom badly, particularly in
scenarios with higher BDP. This may have gone poorly noticed as a result
of it being gated on SACK, which is not enabled by default in upstream
gVisor, but itself has a higher positive impact on performance. Both the
RACK and DACK implementations (which are now one) have overlapping
non-completion of tasks in their work streams on the public tracker.
Updates #9707
Signed-off-by: James Tucker <james@tailscale.com>
We had the debug packet capture code + Lua dissector in the CLI + the
iOS app. Now we don't, with tests to lock it in.
As a bonus, tailscale.com/net/packet and tailscale.com/net/flowtrack
no longer appear in the CLI's binary either.
A new build tag ts_omit_capture disables the packet capture code and
was added to build_dist.sh's --extra-small mode.
Updates #12614
Change-Id: I79b0628c0d59911bd4d510c732284d97b0160f10
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Some natc instances have been observed with excessive memory growth,
dominant in gvisor buffers. It is likely that the connection buffers are
sticking around for too long due to the default long segment time, and
uptuned buffer size applied by default in wgengine/netstack. Apply
configurations in natc specifically which are a better match for the
natc use case, most notably a 5s maximum segment lifetime.
Updates tailscale/corp#25169
Signed-off-by: James Tucker <james@tailscale.com>
Jordan Whited
Claus Lensbøl
Nick Khyl
Irbe Krumina
M. J. Fromberger
David Anderson
Brad Fitzpatrick
Jonathan Nobels
Andrew Dunham
Fran Bull
Adrian Dewhurst
James Tucker
Kristoffer Dalby