Commit Graph

529 Commits (4992aca6ecc632c0722800a4901afe621f0641d0)

Author SHA1 Message Date
Brad Fitzpatrick b104688e04 ipn/ipnlocal, types/netmap: replace hasCapability with set lookup on NetworkMap
When node attributes were super rare, the O(n) slice scans looking for
node attributes was more acceptable. But now more code and more users
are using increasingly more node attributes. Time to make it a map.

Noticed while working on tailscale/corp#17879

Updates #cleanup

Change-Id: Ic17c80341f418421002fbceb47490729048756d2
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
8 months ago
Brad Fitzpatrick f45594d2c9 control/controlclient: free memory on iOS before full netmap work
Updates tailscale/corp#18514

Change-Id: I8d0330334b030ed8692b25549a0ee887ac6d7188
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
8 months ago
Brad Fitzpatrick 8444937c89 control/controlclient: fix panic regression from earlier load balancer hint header
In the recent 20e9f3369 we made HealthChangeRequest machine requests
include a NodeKey, as it was the oddball machine request that didn't
include one. Unfortunately, that code was sometimes being called (at
least in some of our integration tests) without a node key due to its
registration with health.RegisterWatcher(direct.ReportHealthChange).

Fortunately tests in corp caught this before we cut a release. It's
possible this only affects this particular integration test's
environment, but still worth fixing.

Updates tailscale/corp#1297

Change-Id: I84046779955105763dc1be5121c69fec3c138672
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
8 months ago
Joe Tsai 85febda86d
all: use zstdframe where sensible (#11491)
Use the zstdframe package where sensible instead of plumbing
around our own zstd.Encoder just for stateless operations.

This causes logtail to have a dependency on zstd,
but that's arguably okay since zstd support is implicit
to the protocol between a client and the logging service.
Also, virtually every caller to logger.NewLogger was
manually setting up a zstd.Encoder anyways,
meaning that zstd was functionally always a dependency.

Updates #cleanup
Updates tailscale/corp#18514

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
8 months ago
Adrian Dewhurst 2f7e7be2ea control/controlclient: do not alias peer CapMap
Updates #cleanup

Change-Id: I10fd5e04310cdd7894a3caa3045b86eb0a06b6a0
Signed-off-by: Adrian Dewhurst <adrian@tailscale.com>
8 months ago
Brad Fitzpatrick 20e9f3369d control/controlclient: send load balancing hint HTTP request header
Updates tailscale/corp#1297

Change-Id: I0b102081e81dfc1261f4b05521ab248a2e4a1298
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
8 months ago
Claire Wang 221de01745
control/controlclient: fix sending peer capmap changes (#11457)
Instead of just checking if a peer capmap is nil, compare the previous
state peer capmap with the new peer capmap.
Updates tailscale/corp#17516

Signed-off-by: Claire Wang <claire@tailscale.com>
8 months ago
Brad Fitzpatrick e1bd7488d0 all: remove LenIter, use Go 1.22 range-over-int instead
Updates #11058
Updates golang/go#65685

Change-Id: Ibb216b346e511d486271ab3d84e4546c521e4e22
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
9 months ago
Joe Tsai 94a4f701c2
all: use reflect.TypeFor now available in Go 1.22 (#11078)
Updates #cleanup

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
9 months ago
Brad Fitzpatrick 2bd3c1474b util/cmpx: delete now that we're using Go 1.22
Updates #11058

Change-Id: I09dea8e86f03ec148b715efca339eab8b1f0f644
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
9 months ago
Jordan Whited 8b47322acc
wgengine/magicsock: implement probing of UDP path lifetime (#10844)
This commit implements probing of UDP path lifetime on the tail end of
an active direct connection. Probing configuration has two parts -
Cliffs, which are various timeout cliffs of interest, and
CycleCanStartEvery, which limits how often a probing cycle can start,
per-endpoint. Initially a statically defined default configuration will
be used. The default configuration has cliffs of 10s, 30s, and 60s,
with a CycleCanStartEvery of 24h. Probing results are communicated via
clientmetric counters. Probing is off by default, and can be enabled
via control knob. Probing is purely informational and does not yet
drive any magicsock behaviors.

Updates #540

Signed-off-by: Jordan Whited <jordan@tailscale.com>
10 months ago
James Tucker 38a1cf748a control/controlclient,util/execqueue: extract execqueue into a package
This is a useful primitive for asynchronous execution of ordered work I
want to use in another change.

Updates tailscale/corp#16833
Signed-off-by: James Tucker <james@tailscale.com>
10 months ago
James 'zofrex' Sanderson 124dc10261
controlclient,tailcfg,types: expose MaxKeyDuration via localapi (#10401)
Updates tailscale/corp#16016

Signed-off-by: James Sanderson <jsanderson@tailscale.com>
10 months ago
James 'zofrex' Sanderson 10c595d962
ipn/ipnlocal: refresh node key without blocking if cap enabled (#10529)
Updates tailscale/corp#16016

Signed-off-by: James Sanderson <jsanderson@tailscale.com>
Co-authored-by: Maisem Ali <maisem@tailscale.com>
10 months ago
Andrew Lytvynov 2716250ee8
all: cleanup unused code, part 2 (#10670)
And enable U1000 check in staticcheck.

Updates #cleanup

Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
11 months ago
Andrew Lytvynov 945cf836ee
ipn: apply tailnet-wide default for auto-updates (#10508)
When auto-update setting in local Prefs is unset, apply the tailnet
default value from control. This only happens once, when we apply the
default (or when the user manually overrides it), tailnet default no
longer affects the node.

Updates #16244

Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
11 months ago
Naman Sood 0a59754eda linuxfw,wgengine/route,ipn: add c2n and nodeattrs to control linux netfilter
Updates tailscale/corp#14029.

Signed-off-by: Naman Sood <mail@nsood.in>
11 months ago
Matt Layher a217f1fccf all: fix nilness issues
Signed-off-by: Matt Layher <mdlayher@gmail.com>
11 months ago
Brad Fitzpatrick fb829ea7f1 control/controlclient: support incremental packet filter updates [capver 81]
Updates #10299

Change-Id: I87e4235c668a1db7de7ef1abc743f0beecb86d3d
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
12 months ago
Jordan Whited e848736927
control/controlknobs,wgengine/magicsock: implement SilentDisco toggle (#10195)
This change exposes SilentDisco as a control knob, and plumbs it down to
magicsock.endpoint. No changes are being made to magicsock.endpoint
disco behavior, yet.

Updates #540

Signed-off-by: Jordan Whited <jordan@tailscale.com>
Co-authored-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 44c6909c92 control/controlclient: move watchdog out of mapSession
In prep for making mapSession's lifetime not be 1:1 with a single HTTP
response's lifetime, this moves the inactivity timer watchdog out of
mapSession and into the caller that owns the streaming HTTP response.

(This is admittedly closer to how it was prior to the mapSession type
existing, but that was before we connected some dots which were
impossible to even see before the mapSession type broke the code up.)

Updates #7175

Change-Id: Ia108dac84a4953db41cbd30e73b1de4a2a676c11
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick c87d58063a control/controlclient: move lastPrintMap field from Direct to mapSession
It was a really a mutable field owned by mapSession that we didn't move
in earlier commits.

Once moved, it's then possible to de-func-ify the code and turn it into
a regular method rather than an installed optional hook.

Noticed while working to move map session lifetimes out of
Direct.sendMapRequest's single-HTTP-connection scope.

Updates #7175
Updates #cleanup

Change-Id: I6446b15793953d88d1cabf94b5943bb3ccac3ad9
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Adrian Dewhurst 5347e6a292 control/controlclient: support certstore without cgo
We no longer build Windows releases with cgo enabled, which
automatically turned off certstore support. Rather than re-enabling cgo,
we updated our fork of the certstore package to no longer require cgo.
This updates the package, cleans up how the feature is configured, and
removes the cgo build tag requirement.

Fixes tailscale/corp#14797
Fixes tailscale/coral#118

Change-Id: Iaea34340761c0431d759370532c16a48c0913374
Signed-off-by: Adrian Dewhurst <adrian@tailscale.com>
1 year ago
Claire Wang 754fb9a8a8
tailcfg: add tailnet field to register request (#9675)
Updates tailscale/corp#10967

Signed-off-by: Claire Wang <claire@tailscale.com>
1 year ago
Maisem Ali 3655fb3ba0 control/controlclient: fix deadlock in shutdown
Fixes a deadlock observed in a different repo.
Regressed in 5b3f5eabb5.

Updates tailscale/corp#14950
Updates tailscale/corp#14515
Updates tailscale/corp#14139
Updates tailscale/corp#13175

Signed-off-by: Maisem Ali <maisem@tailscale.com>
1 year ago
Brad Fitzpatrick 425cf9aa9d tailcfg, all: use []netip.AddrPort instead of []string for Endpoints
It's JSON wire compatible.

Updates #cleanup

Change-Id: Ifa5c17768fec35b305b06d75eb5f0611c8a135a6
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 246e0ccdca tsnet: add a test for restarting a tsnet server, fix Windows
Thanks to @qur and @eric for debugging!

Fixes #6973

Change-Id: Ib2cf8f030cf595cc73dd061c72e78ac19f5fae5d
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 5b3f5eabb5 control/controlclient: fix leaked http2 goroutines on shutdown
If a noise dial was happening concurrently with shutdown, the
http2 goroutines could leak.

Updates tailscale/corp#14950
Updates tailscale/corp#14515
Updates tailscale/corp#14139
Updates tailscale/corp#13175

Signed-off-by: Brad Fitzpatrick <brad@danga.com>
1 year ago
Claire Wang e3d6236606
winutil: refactor methods to get values from registry to also return (#9536)
errors
Updates tailscale/corp#14879

Signed-off-by: Claire Wang <claire@tailscale.com>
1 year ago
Andrew Dunham 530aaa52f1 net/dns: retry forwarder requests over TCP
We weren't correctly retrying truncated requests to an upstream DNS
server with TCP. Instead, we'd return a truncated request to the user,
even if the user was querying us over TCP and thus able to handle a
large response.

Also, add an envknob and controlknob to allow users/us to disable this
behaviour if it turns out to be buggy ( DNS ).

Updates #9264

Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: Ifb04b563839a9614c0ba03e9c564e8924c1a2bfd
1 year ago
Brad Fitzpatrick 3b32d6c679 wgengine/magicsock, controlclient, net/dns: reduce some logspam
Updates #cleanup

Change-Id: I78b0697a01e94baa33f3de474b591e616fa5e6af
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Val 65dc711c76 control,tailcfg,wgengine/magicsock: add nodeAttr to enable/disable peer MTU
Add a nodeAttr to enable/disable peer path MTU discovery.

Updates #311

Signed-off-by: Val <valerie@tailscale.com>
1 year ago
Tom DNetto c08cf2a9c6 all: declare & plumb IPv6 masquerade address for peer
This PR plumbs through awareness of an IPv6 SNAT/masquerade address from the wire protocol
through to the low-level (tstun / wgengine). This PR is the first in two PRs for implementing
IPv6 NAT support to/from peers.

A subsequent PR will implement the data-plane changes to implement IPv6 NAT - this is just plumbing.

Signed-off-by: Tom DNetto <tom@tailscale.com>
Updates ENG-991
1 year ago
Maisem Ali 19a9d9037f tailcfg: add NodeCapMap
Like PeerCapMap, add a field to `tailcfg.Node` which provides
a map of Capability to raw JSON messages which are deferred to be
parsed later by the application code which cares about the specific
capabilities. This effectively allows us to prototype new behavior
without having to commit to a schema in tailcfg, and it also opens up
the possibilities to develop custom behavior in tsnet applications w/o
having to plumb through application specific data in the MapResponse.

Updates #4217

Signed-off-by: Maisem Ali <maisem@tailscale.com>
1 year ago
Brad Fitzpatrick 0d991249e1 types/netmap: remove NetworkMap.{Addresses,MachineStatus}
And convert all callers over to the methods that check SelfNode.

Now we don't have multiple ways to express things in tests (setting
fields on SelfNode vs NetworkMap, sometimes inconsistently) and don't
have multiple ways to check those two fields (often only checking one
or the other).

Updates #9443

Change-Id: I2d7ba1cf6556142d219fae2be6f484f528756e3c
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Maisem Ali a61caea911 tailcfg: define a type for NodeCapability
Instead of untyped string, add a type to identify these.

Updates #cleanup

Signed-off-by: Maisem Ali <maisem@tailscale.com>
1 year ago
James Tucker e7727db553 tailcfg: add DNS address list for IsWireGuardOnly nodes
Tailscale exit nodes provide DNS service over the peer API, however
IsWireGuardOnly nodes do not have a peer API, and instead need client
DNS parameters passed in their node description.

For Mullvad nodes this will contain the in network 10.64.0.1 address.

Updates #9377

Signed-off-by: James Tucker <james@tailscale.com>
1 year ago
Brad Fitzpatrick 9203916a4a control/controlknobs: move more controlknobs code from controlclient
Updates #cleanup

Change-Id: I2b8b6ac97589270f307bfb20e33674894ce873b5
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 3af051ea27 control/controlclient, types/netmap: start plumbing delta netmap updates
Currently only the top four most popular changes: endpoints, DERP
home, online, and LastSeen.

Updates #1909

Change-Id: I03152da176b2b95232b56acabfb55dcdfaa16b79
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 7c1ed38ab3 ipn/ipnlocal: fix missing controlknobs.Knobs plumbing
I missed connecting some controlknobs.Knobs pieces in 4e91cf20a8
resulting in that breaking control knobs entirely.

Whoops.

The fix in ipn/ipnlocal (where it makes a new controlclient) but to
atone, I also added integration tests. Those integration tests use
a new "tailscale debug control-knobs" which by itself might be useful
for future debugging.

Updates #9351

Change-Id: Id9c89c8637746d879d5da67b9ac4e0d2367a3f0d
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 42072683d6 control/controlknobs: move ForceBackgroundSTUN to controlknobs.Knobs
This is both more efficient (because the knobs' bool is only updated
whenever Node is changed, rarely) and also gets us one step closer to
removing a case of storing a netmap.NetworkMap in
magicsock. (eventually we want to phase out much of the use of that
type internally)

Updates #1909

Change-Id: I37e81789f94133175064fdc09984e4f3a431f1a1
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 4e91cf20a8 control/controlknobs, all: add plumbed Knobs type, not global variables
Previously two tsnet nodes in the same process couldn't have disjoint
sets of controlknob settings from control as both would overwrite each
other's global variables.

This plumbs a new controlknobs.Knobs type around everywhere and hangs
the knobs sent by control on that instead.

Updates #9351

Change-Id: I75338646d36813ed971b4ffad6f9a8b41ec91560
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 9a86aa5732 all: depend on zstd unconditionally, remove plumbing to make it optional
All platforms use it at this point, including iOS which was the
original hold out for memory reasons. No more reason to make it
optional.

Updates #9332

Change-Id: I743fbc2f370921a852fbcebf4eb9821e2bdd3086
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Maisem Ali f6a203fe23 control/controlclient: check c.closed in waitUnpause
We would only check if the client was paused, but not
if the client was closed. This meant that a call to
Shutdown may block forever/leak goroutines

Updates #cleanup

Signed-off-by: Maisem Ali <maisem@tailscale.com>
1 year ago
Maisem Ali d06a75dcd0 ipn/ipnlocal: fix deadlock in resetControlClientLocked
resetControlClientLocked is called while b.mu was held and
would call cc.Shutdown which would wait for the observer queue
to drain.
However, there may be active callbacks from cc already waiting for
b.mu resulting in a deadlock.

This makes it so that resetControlClientLocked does not call
Shutdown, and instead just returns the value.
It also makes it so that any status received from previous cc
are ignored.

Updates tailscale/corp#12827

Signed-off-by: Maisem Ali <maisem@tailscale.com>
1 year ago
Joe Tsai c6fadd6d71
all: implement AppendText alongside MarshalText (#9207)
This eventually allows encoding packages that may respect
the proposed encoding.TextAppender interface.
The performance gains from this is between 10-30%.

Updates tailscale/corp#14379

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
1 year ago
Brad Fitzpatrick 003e4aff71 control/controlclient: clean up various things in prep for state overhaul
We want the overall state (used only for tests) to be computed from
the individual states of each component, rather than moving the state
around by hand in dozens of places.

In working towards that, we found a lot of things to clean up.

Updates #cleanup

Change-Id: Ieaaae5355dfae789a8ec7a56ce212f1d7e3a92db
Co-authored-by: Maisem Ali <maisem@tailscale.com>
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 9cbec4519b control/controlclient: serialize Observer calls
Don't just start goroutines and hope for them to be ordered.

Fixes potential regression from earlier 7074a40c0.

Updates #cleanup

Change-Id: I501a6f3e4e8e6306b958bccdc1e47869991c31f7
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick f7b7ccf835 control/controlclient, ipn/ipnlocal: unplumb a bool true literal opt
Updates #cleanup

Change-Id: I664f280a2e06b9875942458afcaf6be42a5e462a
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Maisem Ali 959362a1f4 ipn/ipnlocal,control/controlclient: make Logout more sync
We already removed the async API, make it more sync and remove
the FinishLogout state too.

This also makes the callback be synchronous again as the previous
attempt was trying to work around the logout callback resulting
in a client shutdown getting blocked forever.

Updates #3833

Signed-off-by: Maisem Ali <maisem@tailscale.com>
1 year ago