Commit Graph

5256 Commits (6d84f3409bfb36ceed207563fac01e7b68423c58)
 

Author SHA1 Message Date
Brad Fitzpatrick f710d1cb20 cmd/tailscale/cli: add set --unattended on Windows
Fixes #6567

Change-Id: I8cb57196c601466401f8602eb50456e7cf7c31ef
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
License Updater d2a51f03ce licenses: update win/apple licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Denton Gentry a200a23f97 Revert "cmd/tailscale: access QNAP via localhost"
When running `tailscale web` as a standalone process,
it was necessary to send auth requests to QTS using
localhost to avoid hitting the proxy recursively.

However running `tailscale web` as a process means it is
consuming RAM all the time even when it isn't actively
doing anything.

After switching back to the `tailscale web` CGI mode, we
don't need to specifically use localhost for QNAP auth.

This reverts commit e0cadc5496.

Signed-off-by: Denton Gentry <dgentry@tailscale.com>
2 years ago
Maisem Ali 82ad585b5b ipn/ipnlocal: account for ControlURL when merging profiles
We merge/dedupe profiles based on UserID and NodeID, however we were not accounting for ControlURLs.

Updates #713

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Maisem Ali adc302f428 all: use named pipes on windows
Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Tom DNetto 45042a76cd cmd/tailscale,ipn: store disallowed TKA's in prefs, lock local-disable
Take 2 of https://github.com/tailscale/tailscale/pull/6546

Builds on https://github.com/tailscale/tailscale/pull/6560

Signed-off-by: Tom DNetto <tom@tailscale.com>
2 years ago
Tom DNetto c4980f33f7 ipn,types/persist: add DisallowedTKAStateIDs, refactor as view type
Supercedes https://github.com/tailscale/tailscale/pull/6557, precursor to trying https://github.com/tailscale/tailscale/pull/6546 again

Signed-off-by: Tom DNetto <tom@tailscale.com>
2 years ago
Brad Fitzpatrick 6d012547b6 ipn/ipnlocal: use double dash flag style
The Go style weirds people out so we try to stick to the more
well-known double hyphen style in docs.

Change-Id: Iad6db5c82cda37f6b7687eed7ecd9276f8fd94d6
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 390d1bb871 Revert "ipn,types/persist: store disallowed TKA's in prefs, lock local-disable"
This reverts commit f1130421f0.

It was submitted with failing tests (go generate checks)

Requires a lot of API changes to fix so rolling back instead of
forward.

Change-Id: I024e8885c0ed44675d3028a662f386dda811f2ad
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Tom DNetto f1130421f0 ipn,types/persist: store disallowed TKA's in prefs, lock local-disable
Signed-off-by: Tom DNetto <tom@tailscale.com>
2 years ago
Aaron Klotz 659e7837c6 health, ipn/ipnlocal: when -no-logs-no-support is enabled, deny access to tailnets that have network logging enabled
We want users to have the freedom to start tailscaled with `-no-logs-no-support`,
but that is obviously in direct conflict with tailnets that have network logging
enabled.

When we detect that condition, we record the issue in health, notify the client,
set WantRunning=false, and bail.

We clear the item in health when a profile switch occurs, since it is a
per-tailnet condition that should not propagate across profiles.

Signed-off-by: Aaron Klotz <aaron@tailscale.com>
2 years ago
Maisem Ali ad41cbd9d5 ipn/ipnlocal: sanitize prefs before sending over IPN Bus
Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Brad Fitzpatrick 0a10a5632b cmd/tailscaled: add TS_DEBUG_BACKEND_DELAY_SEC for testing async startup
This adds an envknob to make testing async startup more reproducible.
We want the Windows GUI to behave well when wintun is not (or it's
doing its initial slow driver installation), but during testing it's often
too fast to see that it's working. This lets it be slowed down.

Updates #6522

Change-Id: I6ae19f46e270ea679cbaea32a53888efcf2943a7
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 256ba62e00 README.md: add commit message style bit
Change-Id: I0c9423b76f773e20f0216d826620e169f82ee37c
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 0cb2ccce7f safesocket: remove the IPN protocol support
Updates #6417

Change-Id: I78908633de842d83b2cc8b10a864a0f88ab1b113
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 06c4c47d46 ipn: remove unused Backend interface
Only the macOS/iOS clients care about it still, so we'll move it
to their repo.

But keep a test that makes sure that LocalBackend continues to
implement it so we get an early warning sign before we break
macOS/iOS.

Change-Id: I56392b740fe55b4d28468b77124c821b5c46c22b
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Joe Tsai 2e5d08ec4f
net/connstats: invert network logging data flow (#6272)
Previously, tstun.Wrapper and magicsock.Conn managed their
own statistics data structure and relied on an external call to
Extract to extract (and reset) the statistics.
This makes it difficult to ensure a maximum size on the statistics
as the caller has no introspection into whether the number
of unique connections is getting too large.

Invert the control flow such that a *connstats.Statistics
is registered with tstun.Wrapper and magicsock.Conn.
Methods on non-nil *connstats.Statistics are called for every packet.
This allows the implementation of connstats.Statistics (in the future)
to better control when it needs to flush to ensure
bounds on maximum sizes.

The value registered into tstun.Wrapper and magicsock.Conn could
be an interface, but that has two performance detriments:

1. Method calls on interface values are more expensive since
they must go through a virtual method dispatch.

2. The implementation would need a sync.Mutex to protect the
statistics value instead of using an atomic.Pointer.

Given that methods on constats.Statistics are called for every packet,
we want reduce the CPU cost on this hot path.

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
2 years ago
Joe Tsai 35c10373b5
types/logid: move logtail ID types here (#6336)
Many packages reference the logtail ID types,
but unfortunately pull in the transitive dependencies of logtail.
Fix this problem by putting the log ID types in its own package
with minimal dependencies.

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
2 years ago
Anton Tolchanov 6cc6c70d70 derp: prevent concurrent access to multiForwarder map
Instead of iterating over the map to determine the preferred forwarder
on every packet (which could happen concurrently with map mutations),
store it separately in an atomic variable.

Fixes #6445

Signed-off-by: Anton Tolchanov <anton@tailscale.com>
2 years ago
Aaron Klotz 6e33d2da2b ipn/ipnauth, util/winutil: add temporary LookupPseudoUser workaround to address os/user.LookupId errors on Windows
I added util/winutil/LookupPseudoUser, which essentially consists of the bits
that I am in the process of adding to Go's standard library.

We check the provided SID for "S-1-5-x" where 17 <= x <= 20 (which are the
known pseudo-users) and then manually populate a os/user.User struct with
the correct information.

Fixes https://github.com/tailscale/tailscale/issues/869
Fixes https://github.com/tailscale/tailscale/issues/2894

Signed-off-by: Aaron Klotz <aaron@tailscale.com>
2 years ago
Brad Fitzpatrick 3b0de97e07 cmd/tailscaled: unify the two Windows paths + separate IPN server path
tailscaled on Windows had two entirely separate start-up paths for running
as a service vs in the foreground. It's been causing problems for ages.
This unifies the two paths, making them be the same as the path used
for every other platform.

Also, it uses the new async LocalBackend support in ipnserver.Server
so the Server can start serving HTTP immediately, even if tun takes
awhile to come up.

Updates #6535

Change-Id: Icc8c4f96d4887b54a024d7ac15ad11096b5a58cf
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick ea25ef8236 util/set: add new set package for SetHandle type
We use this pattern in a number of places (in this repo and elsewhere)
and I was about to add a fourth to this repo which was crossing the line.
Add this type instead so they're all the same.

Also, we have another Set type (SliceSet, which tracks its keys in
order) in another repo we can move to this package later.

Change-Id: Ibbdcdba5443fae9b6956f63990bdb9e9443cefa9
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Tom DNetto 5c8d2fa695 cmd/tailscale,ipn: improve UX of lock init command, cosmetic changes
Signed-off-by: Tom DNetto <tom@tailscale.com>
2 years ago
Brad Fitzpatrick e8cc78b1af ipn/ipnserver: change Server to let LocalBackend be supplied async
This is step 1 of de-special-casing of Windows and letting the
LocalAPI HTTP server start serving immediately, even while the rest of
the world (notably the Engine and its TUN device) are being created,
which can take a few to dozens of seconds on Windows.

With this change, the ipnserver.New function changes to not take an
Engine and to return immediately, not returning an error, and let its
Run run immediately. If its ServeHTTP is called when it doesn't yet
have a LocalBackend, it returns an error. A TODO in there shows where
a future handler will serve status before an engine is available.

Future changes will:

* delete a bunch of tailscaled_windows.go code and use this new API
* add the ipnserver.Server ServerHTTP handler to await the engine
  being available
* use that handler in the Windows GUI client

Updates #6522

Change-Id: Iae94e68c235e850b112a72ea24ad0e0959b568ee
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 8049053f86 ipn/*: make new WindowsUserID type to consolidate docs
The "userID is empty everywhere but Windows" docs on lots of places
but not everywhere while using just a string type was getting
confusing. This makes a new type to wrap up those rules, however
weird/historical they might be.

Change-Id: I142e85a8e38760988d6c0c91d0efecedade81b9b
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 3b73727e39 cmd/tailscale: de-punycode hostnames in status display
Still show original, but show de-punycode version in parens,
similar to how we show DNS-less hostnames.

Change-Id: I7e57da5e4029c5b49e8cd3014c350eddd2b3c338
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 5676d201d6 ipn: add a WatchIPNBus option bit to subscribe to EngineStatus changes
So GUI clients don't need to poll for it.

We still poll internally (for now!) but that's still cheaper. And will
get much cheaper later, without having to modify clients once they
start sending this bit.

Change-Id: I36647b701c8d1fe197677e5eb76f6894e8ff79f7
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick f45106d47c ipn/ipnserver: move Windows-specific code to tailscaled_windows.go
We'll eventually remove it entirely, but for now move get it out of ipnserver
where it's distracting and move it to its sole caller.

Updates #6522

Change-Id: I9c6f6a91bf9a8e3c5ea997952b7c08c81723d447
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 109aa3b2fb cmd/tailscale: add start of "debug derp" subcommand
Updates #6526

Change-Id: I84e440a8bd837c383000ce0cec4ff36b24249e8b
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick b0545873e5 ipn/ipnserver: remove protoSwitchConn shenanigans; just use http.Server early
Now that everything's just HTTP, there's no longer a need to have a
header-sniffing net.Conn wraper that dispatches which route to
take. Refactor to just use an http.Server earlier instead.

Updates #6417

Change-Id: I12a2054db4e56f48660c46f81233db224fdc77cb
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
shayne e567902aa9
gitignore: ignore direnv nix-shell environment cache (#6520)
Also, updated ignore desc for .vscode/.

Signed-off-by: Shayne Sweeney <shayne@tailscale.com>
2 years ago
Brad Fitzpatrick f3ba268a96 ipn/ipnserver: move BabysitProc to tailscaled_windows.go
It's only used by Windows. No need for it to be in ipn/ipnserver,
which we're trying to trim down.

Change-Id: Idf923ac8b6cdae8b5338ec26c16fb8b5ea548071
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Maisem Ali 699b39dec1 ipn/ipnlocal: drop LocalBackend.inServerMode
Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Brad Fitzpatrick 7e016c1d90 ipn/ipnserver: remove IPN protocol server
Unused in this repo as of the earlier #6450 (300aba61a6)
and unused in the Windows GUI as of tailscale/corp#8065.

With this ipn.BackendServer is no longer used and could also be
removed from this repo. The macOS and iOS clients still temporarily
depend on it, but I can move it to that repo instead while and let its
migration proceed on its own schedule while we clean this repo up.

Updates #6417
Updates tailscale/corp#8051

Change-Id: Ie13f82af3eb9f96b3a21c56cdda51be31ddebdcf
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
License Updater 624d9c759b licenses: update win/apple licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Denton Gentry b8fe89d15f net/portmapper: add test for Huawei router
Updates https://github.com/tailscale/tailscale/issues/6320

Signed-off-by: Denton Gentry <dgentry@tailscale.com>
2 years ago
Brad Fitzpatrick 1fdfb0dd08 ipn/localapi: add "enginestatus" debug command to LocalAPI
To force an EngineStatus update to the IPN bus.

This is a temporary measure while migrating the Windows GUI entirely
to the LocalAPI and off the old IPN protocol. The old IPN protocol
had RequestEngineStatus and LocalAPI didn't.

Updates #6417

Change-Id: I8ff525fc3dd82bdd9d92c2bdad6db5b75609eacd
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Maisem Ali c258015165 ipn/ipnlocal,ipnserver: rename {,Set}CurrentUser to {,Set}CurrentUserID
Address comments from https://github.com/tailscale/tailscale/pull/6506#discussion_r1032454064

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Brad Fitzpatrick 0a842f353c ipn/ipnserver: move more connection acceptance logic to LocalBackend
Follow-up to #6467 and #6506.

LocalBackend knows the server-mode state, so move more auth checking
there, removing some bookkeeping from ipnserver.Server.

Updates #6417
Updates tailscale/corp#8051

Change-Id: Ic5d14a077bf0dccc92a3621bd2646bab2cc5b837
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 5ea7c7d603 ipn/{ipnlocal,ipnserver}: add some comments
Change-Id: Ieb5917edaf572342b755caa458693512c7aece81
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Maisem Ali 732c3d2ed0 .github/workflows: use ubuntu-22.04 for qemu tests
Fixes #6507

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Maisem Ali a3cd171773 ipn/ipnserver: remove Server.serverModeUser
We can just rely on LocalBackend.CurrentUser

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Maisem Ali d321b0ea4f ipn/ipnlocal: add docs to LocalBackend.SetCurrentUserID
Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Maisem Ali 992749c44c tsnet/example/tshello: use the correct LocalClient for certs
Fixes #6485

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
shayne 0c4c66948b
cmd/tailscale/cli: Improve messaging when Funnel is unavailable. (#6502)
There are three specific requirements for Funnel to work:
1) They must accept an invite.
2) They must enable HTTPS.
3) The "funnel" node attribute must be appropriately set up in the ACLs.

Signed-off-by: Shayne Sweeney <shayne@tailscale.com>
2 years ago
License Updater 344abaf3d3 licenses: update android licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Brad Fitzpatrick 250edeb3da ipn/ipnserver: only permit the pre-HTTP LocalAPI protocol on Windows
Updates #6417

Change-Id: I1c9dbee3f72969f703b3ff2dbbaa145a17db868b
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Aaron Klotz 033bd94d4c cmd/tailscaled, wgengine/router: use wingoes/com for COM initialization instead of go-ole
This patch removes the crappy, half-backed COM initialization used by `go-ole`
and replaces that with the `StartRuntime` function from `wingoes`, a library I
have started which, among other things, initializes COM properly.

In particular, we should always be initializing COM to use the multithreaded
apartment. Every single OS thread in the process becomes implicitly initialized
as part of the MTA, so we do not need to concern ourselves as to whether or not
any particular OS thread has initialized COM. Furthermore, we no longer need to
lock the OS thread when calling methods on COM interfaces.

Single-threaded apartments are designed solely for working with Win32 threads
that have a message pump; any other use of the STA is invalid.

Fixes https://github.com/tailscale/tailscale/issues/3137

Signed-off-by: Aaron Klotz <aaron@tailscale.com>
2 years ago
Charlotte Brandhorst-Satzkorn d6021ae71c
words: hybrid theory (#6404)
The animal, not the car kind.

Signed-off-by: Charlotte Brandhorst-Satzkorn <charlotte@tailscale.com>
2 years ago
Brad Fitzpatrick b68d008fee envknob: add CanTaildrop (TS_DISABLE_TAILDROP) to disable taildrop on device
This matches CanSSHD (TS_DISABLE_SSH_SERVER) for administratively
disabling the code on a node, regardless of local or server configs.

This can be configured in /etc/default/tailscaled on Linux,
%ProgramData%\Tailscale\tailscaled-env.txt on Windows,
or /etc/tailscale/tailscaled-env.txt on Synology. (see getPlatformEnvFile)

Also delete some dead code and tidy up some docs.

Change-Id: I79a87c03e33209619466ea8aeb0f6651afcb8789
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago