Commit Graph

7049 Commits

Author SHA1 Message Date
Will Norris
9843e922b8 README: update docs for building web client
Move into the main "building" section and add the missing `yarn install`
step.

Updates #9312

Signed-off-by: Will Norris <will@tailscale.com>
2023-09-13 09:38:40 -07:00
Tyler Smalley
82c1dd8732
cmd/tailscale: funnel wip cleanup and additional test coverage (#9316)
General cleanup and additional test coverage of WIP code.

* use enum for serveType
* combine instances of ServeConfig access within unset
* cleanMountPoint rewritten into cleanURLPath as it only handles URL paths
* refactor and test expandProxyTargetDev

> **Note**
> Behind the `TAILSCALE_USE_WIP_CODE` flag

updates #8489

Signed-off-by: Tyler Smalley <tyler@tailscale.com>
2023-09-13 10:41:30 -05:00
Brad Fitzpatrick
3c276d7de2 wgengine: remove SetDERPMap method from Engine interface
(continuing the mission of removing rando methods from the Engine
 interface that we don't need anymore)

Updates #cleanup

Change-Id: Id5190917596bf04d7185c3b331a852724a3f5a16
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-12 21:16:56 -07:00
Brad Fitzpatrick
67396d716b ipn/ipnlocal: remove defensiveness around not having a magicsock.Conn
We always have one. Stop pretending we might not.

Instead, add one early panic in NewLocalBackend if we actually don't.

Updates #cleanup

Change-Id: Iba4b78ed22cb6248e59c2b01a79355ca7a200ec8
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-12 20:52:08 -07:00
Brad Fitzpatrick
b8a4c96c53 wgengine: remove LinkChange method from Engine interface
It was only used by Android, until
https://github.com/tailscale/tailscale-android/pull/131
which does the call to the netMon directly instead.

Updates #cleanup

Change-Id: Iab8a1d8f1e63250705835c75f40e2cd8c1c4d5b8
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-12 17:25:17 -07:00
Brad Fitzpatrick
727b1432a8 wgengine: remove SetNetInfoCallback method from Engine
LocalBackend can talk to magicsock on its own to do this without
the "Engine" being involved.

(Continuing a little side quest of cleaning up the Engine
interface...)

Updates #cleanup

Change-Id: I8654acdca2b883b1bd557fdc0cfb90cd3a418a62
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-12 15:14:14 -07:00
Andrew Dunham
ad4c11aca1 net/netmon: log when the gateway/self IP changes
This logs that the gateway/self IP address has changed if one of the new
values differs.

Updates #8992

Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I0919424b68ad97fbe1204dd36317ed6f5915411f
2023-09-12 17:48:29 -04:00
License Updater
45eafe1b06 licenses: update win/apple licenses
Signed-off-by: License Updater <noreply+license-updater@tailscale.com>
2023-09-12 14:11:00 -07:00
Brad Fitzpatrick
eb9f1db269 cmd/tsconnect/wasm: register netstack.Impl with tsd.System
I missed this in 343c0f1031 and I guess we don't have integration
tests for wasm. But it compiled! :)

Updates #fixup to a #cleanup

Change-Id: If147b90bab254d144ec851a392e8db10ab97f98e
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-12 14:10:06 -07:00
Brad Fitzpatrick
343c0f1031 wgengine{,/netstack}: remove AddNetworkMapCallback from Engine interface
It had exactly one user: netstack. Just have LocalBackend notify
netstack when here's a new netmap instead, simplifying the bloated
Engine interface that has grown a bunch of non-Engine-y things.
(plenty of rando stuff remains after this, but it's a start)

Updates #cleanup

Change-Id: I45e10ab48119e962fc4967a95167656e35b141d8
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-12 13:55:57 -07:00
David Crawshaw
47ffbffa97
clientupdate: add root key (#9364)
Signed-off-by: David Crawshaw <crawshaw@tailscale.com>
2023-09-12 15:24:01 -05:00
Brad Fitzpatrick
39ade4d0d4 tstest/integration: add start of integration tests for incremental map updates
This adds a new integration test with two nodes where the first gets a
incremental MapResponse (with only PeersRemoved set) saying that the
second node disappeared.

This extends the testcontrol package to support sending raw
MapResponses to nodes.

Updates #1909

Change-Id: Iea0c25c19cf0d72b52dba5a46d01b5cc87b9b39d
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-12 12:47:38 -07:00
Brad Fitzpatrick
9203916a4a control/controlknobs: move more controlknobs code from controlclient
Updates #cleanup

Change-Id: I2b8b6ac97589270f307bfb20e33674894ce873b5
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-12 12:44:35 -07:00
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>
2023-09-12 12:23:24 -07:00
Andrew Lytvynov
c0ade132e6
clientupdate: restart tailscale after install on DSM6 (#9363)
DSM6 does not automatically restart packages on install, we have to do
it explicitly.

Also, DSM6 has a filter for publishers in Package Center. Make the error
message more helpful when update fails because of this filter not
allowing our package.

Fixes #9361

Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
2023-09-12 13:08:00 -05:00
Brad Fitzpatrick
668a0dd5ab cmd/tailscale/cli: fix panic in netcheck print when no DERP home
Fixes #8016

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-12 08:47:45 -07:00
Andrew Dunham
9ee173c256 net/portmapper: fall back to permanent UPnP leases if necessary
Some routers don't support lease times for UPnP portmapping; let's fall
back to adding a permanent lease in these cases. Additionally, add a
proper end-to-end test case for the UPnP portmapping behaviour.

Updates #9343

Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I17dec600b0595a5bfc9b4d530aff6ee3109a8b12
2023-09-12 11:16:45 -04:00
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>
2023-09-12 06:17:14 -07:00
Marwan Sulaiman
12d4685328 ipn/localapi, ipn/ipnlocal: add etag support for SetServeConfig
This PR adds optimistic concurrency control in the local client and
api in order to ensure multiple writes of the ServeConfig do not
conflict with each other.

Updates #9273

Signed-off-by: Marwan Sulaiman <marwan@tailscale.com>
2023-09-12 04:41:10 -04:00
Brad Fitzpatrick
ff6fadddb6 wgengine/magicsock: stop retaining *netmap.NetworkMap
We're trying to start using that monster type less and eventually get
rid of it.

Updates #1909

Change-Id: I8e1e725bce5324fb820a9be6c7952767863e6542
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-11 20:07:30 -07:00
Brad Fitzpatrick
f06e64c562 wgengine: use set.HandleSet in another place
I guess we missed this one earlier when we unified the various
copies into set.HandleSet.

Updates #cleanup

Change-Id: I7e6de9ce16e8fc4846abf384dfcc8eaec4d99e60
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-11 20:07:05 -07:00
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>
2023-09-11 18:11:09 -07:00
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>
2023-09-11 12:44:03 -07:00
Brad Fitzpatrick
d050700a3b wgengine/magicsock: make peerMap also keyed by NodeID
In prep for incremental netmap update plumbing (#1909), make peerMap
also keyed by NodeID, as all the netmap node mutations passed around
later will be keyed by NodeID.

In the process, also:

* add envknob.InDevMode, as a signal that we can panic more aggressively
  in unexpected cases.
* pull two moderately large blocks of code in Conn.SetNetworkMap out
  into their own methods
* convert a few more sets from maps to set.Set

Updates #1909

Change-Id: I7acdd64452ba58e9d554140ee7a8760f9043f961
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-11 12:43:47 -07:00
Paul Scott
683ba62f3e
cmd/testwrapper: fix exit deflake (#9342)
Sometimes `go test` would exit and close its stdout before we started reading
it, and we would return that "file closed" error then forget to os.Exit(1).
Fixed to prefer the go test subprocess error and exit regardless of the type of
error.

Fixes #9334

Signed-off-by: Paul Scott <paul@tailscale.com>
2023-09-11 19:06:11 +01:00
Brad Fitzpatrick
0396366aae cmd/testwrapper/flakytest: don't spam stderr in Mark when not under wrapper
If the user's running "go test" by hand, no need to spam stderr with
the sentinel marker. It already calls t.Logf (which only gets output
on actual failure, or verbose mode) which is enough to tell users it's
known flaky. Stderr OTOH always prints out and is distracting to
manual "go test" users.

Updates #cleanup

Change-Id: Ie5e6881bae291787c30f75924fa132f4a28abbb2
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-11 11:04:18 -07:00
Brad Fitzpatrick
70ea073478 tailcfg: flesh out some docs on MapResponse, clarify slices w/ omitempty
Updates #cleanup

Change-Id: If4caf9d00529edc09ae7af9cc70f6ba0ade38378
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-10 09:22:41 -07:00
Brad Fitzpatrick
a5ffd5e7c3 tailcfg: remove unused MapRequest.IncludeIPv6 field
It's been implicitly enabled (based on capver) for years.

Updates #cleanup

Change-Id: I8ff1ab844f9ed75c97e866e778dfc0b56cfa98a2
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-10 09:21:07 -07:00
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>
2023-09-10 08:36:05 -07:00
Marwan Sulaiman
f12c71e71c cmd/tailscale: reduce duplicate calls to LocalBackend
This PR ensures calls to the LocalBackend are not happening
multiples times and ensures the set/unset methods are
only manipulating the serve config

Updates #8489

Signed-off-by: Marwan Sulaiman <marwan@tailscale.com>
2023-09-10 11:31:57 -04:00
Brad Fitzpatrick
dc7aa98b76 all: use set.Set consistently instead of map[T]struct{}
I didn't clean up the more idiomatic map[T]bool with true values, at
least yet.  I just converted the relatively awkward struct{}-valued
maps.

Updates #cleanup

Change-Id: I758abebd2bb1f64bc7a9d0f25c32298f4679c14f
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-09 10:59:19 -07:00
Brad Fitzpatrick
d506a55c8a ipn/ipnstate: address TODO about garbage during peer sorting
Updates #cleanup

Change-Id: I34938bca70a95571cc62ce1f76eaab5db8c2c3ef
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-09 09:12:23 -07:00
Brad Fitzpatrick
60e9bd6047 ipn/ipnstate: add some missing docs
Updates #cleanup

Change-Id: I689f8124a5986a98b8eb3891727d39c96408f0a7
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-09 08:55:43 -07:00
Brad Fitzpatrick
db307d35e1 types/netmap: delete a copy of views.SliceEqual
Updates #cleanup

Change-Id: Ibdfa6c5dc9211f5c97c763ba323802a1c1d80c9e
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-08 20:49:10 -07:00
David Anderson
95082a8dde util/lru, util/limiter: add debug helper to dump state as HTML
For use in tsweb debug handlers, so that we can easily inspect cache
and limiter state when troubleshooting.

Updates tailscale/corp#3601

Signed-off-by: David Anderson <danderson@tailscale.com>
2023-09-08 14:47:03 -07:00
Andrew Lytvynov
d23b8ffb13
cmd/tailscale/cli,ipn: mention available update in "tailscale status" (#9205)
Cache the last `ClientVersion` value that was received from coordination
server and pass it in the localapi `/status` response.
When running `tailscale status`, print a message if `RunningAsLatest` is
`false`.

Updates #6907

Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
2023-09-08 14:27:49 -07:00
Andrew Lytvynov
1073b56e18
ipn/ipnlocal: add logging and locking to c2n /update (#9290)
Log some progress info to make updates more debuggable. Also, track
whether an active update is already started and return an error if
a concurrent update is attempted.

Some planned future PRs:
* add JSON output to `tailscale update`
* use JSON output from `tailscale update` to provide a more detailed
  status of in-progress update (stage, download progress, etc)

Updates #6907

Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
2023-09-08 14:26:55 -07:00
Sonia Appasamy
1eadb2b608 client/web: clean up assets handling
A #cleanup that moves all frontend asset handling into assets.go
(formerly dev.go), and stores a single assetsHandler field back
to web.Server that manages when to serve the dev vite proxy versus
static files itself.

Signed-off-by: Sonia Appasamy <sonia@tailscale.com>
2023-09-08 16:05:11 -04:00
License Updater
4a38d8d372 licenses: update win/apple licenses
Signed-off-by: License Updater <noreply+license-updater@tailscale.com>
2023-09-08 08:44:43 -07:00
License Updater
0dc65b2e47 licenses: update tailscale{,d} licenses
Signed-off-by: License Updater <noreply+license-updater@tailscale.com>
2023-09-08 08:42:58 -07:00
License Updater
1383fc57ad licenses: update android licenses
Signed-off-by: License Updater <noreply+license-updater@tailscale.com>
2023-09-08 08:41:56 -07:00
Joe Tsai
0a0adb68ad
ssh/tailssh: log when recording starts and finishes (#9294)
Updates tailscale/corp#14579

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
2023-09-07 18:47:04 -07:00
Marwan Sulaiman
a1d4144b18 cmd/tailscale: combine foreground and background serve logic
Previously, foreground mode only worked in the simple case of `tailscale funnel <port>`.
This PR ensures that whatever you can do in the background can also be
done in the foreground such as setting mount paths or tcp forwarding.

Updates #8489

Signed-off-by: Marwan Sulaiman <marwan@tailscale.com>
2023-09-07 20:48:58 -04:00
Craig Rodrigues
8452d273e3 util/linuxfw: Fix comment which lists supported linux arches
Only arm64 and amd64 are supported

Signed-off-by: Craig Rodrigues <rodrigc@crodrigues.org>
2023-09-07 16:49:50 -07:00
David Anderson
0909e90890 util/lru: replace container/list with a custom ring implementation
pre-generics container/list is quite unpleasant to use, and the pointer
manipulation operations for an LRU are simple enough to implement directly
now that we have generic types.

With this change, the LRU uses a ring (aka circularly linked list) rather
than a simple doubly-linked list as its internals, because the ring makes
list manipulation edge cases more regular: the only remaining edge case is
the transition between 0 and 1 elements, rather than also having to deal
specially with manipulating the first and last members of the list.

While the primary purpose was improved readability of the code, as it
turns out removing the indirection through an interface box also speeds
up the LRU:

       │ before.txt  │              after.txt              │
       │   sec/op    │   sec/op     vs base                │
LRU-32   67.05n ± 2%   59.73n ± 2%  -10.90% (p=0.000 n=20)

       │ before.txt │             after.txt              │
       │    B/op    │    B/op     vs base                │
LRU-32   21.00 ± 0%   10.00 ± 0%  -52.38% (p=0.000 n=20)

       │ before.txt │           after.txt            │
       │ allocs/op  │ allocs/op   vs base            │
LRU-32   0.000 ± 0%   0.000 ± 0%  ~ (p=1.000 n=20) ¹

Updates #cleanup

Signed-off-by: David Anderson <danderson@tailscale.com>
2023-09-07 16:04:39 -07:00
David Anderson
472eb6f6f5 util/lru: add a microbenchmark
The benchmark simulates an LRU being queries with uniformly random
inputs, in a set that's too large for the LRU, which should stress
the eviction codepath.

Signed-off-by: David Anderson <danderson@tailscale.com>
2023-09-07 16:04:39 -07:00
Maisem Ali
18b2638b07 metrics: add missing comma in histogram JSON export
Updates tailscale/corp#8641

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2023-09-07 15:28:12 -07:00
Tyler Smalley
70a9854b39
cmd/tailscale: add background mode to serve/funnel wip (#9202)
> **Note**
> Behind the `TAILSCALE_FUNNEL_DEV` flag

* Expose additional listeners through flags
* Add a --bg flag to run in the background
* --set-path to set a path for a specific target (assumes running in background)

See the parent issue for more context.

Updates #8489

Signed-off-by: Tyler Smalley <tyler@tailscale.com>
2023-09-07 15:07:53 -07:00
Maisem Ali
5ee349e075 tsweb/varz: fix exporting histograms
Updates tailscale/corp#8641

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2023-09-07 14:52:59 -07:00
Brad Fitzpatrick
1bd3edbb46 release/dist/unixpkgs: demote deb iptables+iproute2 packages to recommended
Fixes #9236

Change-Id: Idbad2edb0262ef842afd6b40ae47f46e685b112d
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2023-09-07 14:19:39 -07:00