In this PR, we refactor the LocalBackend extension system, moving from direct callbacks to a more organized extension host model.
Specifically, we:
- Extract interface and callback types used by packages extending LocalBackend functionality into a new ipn/ipnext package.
- Define ipnext.Host as a new interface that bridges extensions with LocalBackend.
It enables extensions to register callbacks and interact with LocalBackend in a concurrency-safe, well-defined, and controlled way.
- Move existing callback registration and invocation code from ipnlocal.LocalBackend into a new type called ipnlocal.ExtensionHost,
implementing ipnext.Host.
- Improve docs for existing types and methods while adding docs for the new interfaces.
- Add test coverage for both the extracted and the new code.
- Remove ipn/desktop.SessionManager from tsd.System since ipn/desktop is now self-contained.
- Update existing extensions (e.g., ipn/auditlog and ipn/desktop) to use the new interfaces where appropriate.
We're not introducing new callback and hook types (e.g., for ipn.Prefs changes) just yet, nor are we enhancing current callbacks,
such as by improving conflict resolution when more than one extension tries to influence profile selection via a background profile resolver.
These further improvements will be submitted separately.
Updates #12614
Updates tailscale/corp#27645
Updates tailscale/corp#26435
Updates tailscale/corp#18342
Signed-off-by: Nick Khyl <nickk@tailscale.com>
This reverts commit 476a4c6ff174d46ce3b125c018c07c43713e1c10.
Reason: redundant with `tailscale status --json | jq '.Self.Relay'`
which we all forgot about. Whoops.
Updates #15625
Adds a new diagram for ProxyGroups running in Ingress mode.
Documentation is currently not publicly available, but a link needs
adding once it is.
Updates tailscale/corp#24795
Change-Id: I0d5dd6bf6f0e1b8b0becae848dc97d8b4bfb9ccb
Signed-off-by: Tom Proctor <tomhjp@users.noreply.github.com>
Ultimately, we'd like to get rid of the concept of the "current user". It is only used on Windows,
but even then it doesn't work well in multi-user and enterprise/managed Windows environments.
In this PR, we update LocalBackend and profileManager to decouple them a bit more from this obsolete concept.
This is done in a preparation for extracting ipnlocal.Extension-related interfaces and types, and using them
to implement optional features like tailscale/corp#27645, instead of continuing growing the core ipnlocal logic.
Notably, we rename (*profileManager).SetCurrentUserAndProfile() to SwitchToProfile() and change its signature
to accept an ipn.LoginProfileView instead of an ipn.ProfileID and ipn.WindowsUserID. Since we're not removing
the "current user" completely just yet, the method sets the current user to the owner of the target profile.
We also update the profileResolver callback type, which is typically implemented by LocalBackend extensions,
to return an ipn.LoginProfileView instead of ipn.ProfileID and ipn.WindowsUserID.
Updates tailscale/corp#27645
Updates tailscale/corp#18342
Signed-off-by: Nick Khyl <nickk@tailscale.com>
ResourceCheck was previously using cmp.Diff on multiline goroutine stacks
The produced output was difficult to read for a number of reasons:
- the goroutines were sorted by count, and a changing count caused them to
jump around
- diffs would be in the middle of stacks
Instead, we now parse the pprof/goroutines?debug=1 format goroutines and
only diff whole stacks.
Updates #1253
Signed-off-by: Paul Scott <paul@tailscale.com>
Default tags to `$TAGS` if set, so that people can choose arbitrary
subsets of features.
Updates #12614
Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
Fix the index out of bound panic when a request is made to the local
fileserver mux with a valid secret-token, but missing share name.
Example error:
http: panic serving 127.0.0.1:40974: runtime error: slice bounds out of range [2:1]
Additionally, we document the edge case behavior of utilities that
this fileserver mux depends on.
Signed-off-by: Craig Hesling <craig@hesling.com>
This makes sure that the log target override is respected even if a
custom HTTP client is passed to logpolicy.
Updates tailscale/maple#29
Signed-off-by: Anton Tolchanov <anton@tailscale.com>
The http.StatusMethodNotAllowed status code was being erroneously
set instead of http.StatusBadRequest in multiple places.
Updates #cleanup
Signed-off-by: Jordan Whited <jordan@tailscale.com>
In this PR, we update the Windows client updater to:
- Run msiexec with logging enabled and preserve the log file in %ProgramData%\Tailscale\Logs;
- Preserve the updater's own log file in the same location;
- Properly handle ERROR_SUCCESS_REBOOT_REQUIRED, ERROR_SUCCESS_REBOOT_INITIATED, and ERROR_INSTALL_ALREADY_RUNNING exit codes. The first two values indicate that installation
completed successfully and no further retries are needed. The last one means the Windows Installer
service is busy. Retrying immediately is likely to fail and may be risky; it could uninstall the current version
without successfully installing the new one, potentially leaving the user without Tailscale.
Updates tailscale/corp#27496
Updates tailscale#15554
Signed-off-by: Nick Khyl <nickk@tailscale.com>
Installer tests only run when changes are made to pkgserve. This PR schedules
these tests to be run daily and report any failures to Slack.
Fixestailscale/corp#19103
Signed-off-by: Jason O'Donnell <2160810+jasonodonnell@users.noreply.github.com>
This flag is currently no-op and hidden. The flag does round trip
through the related pref. Subsequent commits will tie them to
net/udprelay.Server. There is no corresponding "tailscale up" flag,
enabling/disabling of the relay server will only be supported via
"tailscale set".
This is a string flag in order to support disablement via empty string
as a port value of 0 means "enable the server and listen on a random
unused port". Disablement via empty string also follows existing flag
convention, e.g. advertise-routes.
Early internal discussions settled on "tailscale set --relay="<port>",
but the author felt this was too ambiguous around client vs server, and
may cause confusion in the future if we add related flags.
Updates tailscale/corp#27502
Signed-off-by: Jordan Whited <jordan@tailscale.com>
`tailscale set` was created to set preferences, which used to be
overloaded into `tailscale up`. To move people over to the new
command, `up` was supposed to be frozen and no new preference flags
would be added. But people forgot, there was no test to warn them, and
so new flags were added anyway.
TestUpFlagSetIsFrozen complains when new flags are added to
`tailscale up`. It doesn’t try all combinations of GOOS, but since
the CI builds in every OS, the pull-request tests should cover this.
Updates #15460
Signed-off-by: Simon Law <sfllaw@sfllaw.ca>
Ensure no services are advertised as part of shutting down tailscaled.
Prefs are only edited if services are currently advertised, and they're
edited we wait for control's ~15s (+ buffer) delay to failover.
Note that editing prefs will trigger a synchronous write to the state
Secret, so it may fail to persist state if the ProxyGroup is getting
scaled down and therefore has its RBAC deleted at the same time, but that
failure doesn't stop prefs being updated within the local backend,
doesn't affect connectivity to control, and the state Secret is
about to get deleted anyway, so the only negative side effect is a harmless
error log during shutdown. Control still learns that the node is no
longer advertising the service and triggers the failover.
Note that the first version of this used a PreStop lifecycle hook, but
that only supports GET methods and we need the shutdown to trigger side
effects (updating prefs) so it didn't seem appropriate to expose that
functionality on a GET endpoint that's accessible on the k8s network.
Updates tailscale/corp#24795
Change-Id: I0a9a4fe7a5395ca76135ceead05cbc3ee32b3d3c
Signed-off-by: Tom Proctor <tomhjp@users.noreply.github.com>
As IPv4 and IPv6 end up with different MSS and different congestion
control strategies, proxying between them can really amplify TCP
meltdown style conditions in many real world network conditions, such as
with higher latency, some loss, etc.
Attempt to match up the protocols, otherwise pick a destination address
arbitrarily. Also shuffle the target address to spread load across
upstream load balancers.
Updates #15367
Signed-off-by: James Tucker <james@tailscale.com>
When we have an old cert that is being rotated, include it in the order.
If we're in the ARI-recommended rotation window, LE should exclude us
from rate limits. If we're not within that window, the order still
succeeds, so there's no risk in including the old cert.
Fixes#15542
Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
The test suite had grown to about 20s on my machine, but it doesn't
do much taxing work so was a good candidate to parallelise. Now runs
in under 2s on my machine.
Updates #cleanup
Change-Id: I2fcc6be9ca226c74c0cb6c906778846e959492e4
Signed-off-by: Tom Proctor <tomhjp@users.noreply.github.com>
The earlier #15534 prevent some dup string flags. This does it for all
flag types.
Updates #6813
Change-Id: Iec2871448394ea9a5b604310bdbf7b499434bf01
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
tsconsensus enables tsnet.Server instances to form a consensus.
tsconsensus wraps hashicorp/raft with
* the ability to do discovery via tailscale tags
* inter node communication over tailscale
* routing of commands to the leader
Updates #14667
Signed-off-by: Fran Bull <fran@tailscale.com>
So we can link open source contributors to it.
Updates #cleanup
Change-Id: I02f612b38db9594f19b3be5d982f58c136120e9a
Co-authored-by: James Sanderson <jsanderson@tailscale.com>
Co-authored-by: Will Norris <will@tailscale.com>
Co-authored-by: James Tucker <james@tailscale.com>
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Some CLI flags support multiple values separated by commas. These flags
are intended to be declared only once and will silently ignore subsequent
instances. This will now throw an error if multiple instances of advertise-tags
and advertise-routes are detected.
Fixes#6813
Signed-off-by: Jason O'Donnell <2160810+jasonodonnell@users.noreply.github.com>
Ensure that the upstream is always queried, so that if upstream is going
to NXDOMAIN natc will also return NXDOMAIN rather than returning address
allocations.
At this time both IPv4 and IPv6 are still returned if upstream has a
result, regardless of upstream support - this is ~ok as we're proxying.
Rewrite the tests to be once again slightly closer to integration tests,
but they're still very rough and in need of a refactor.
Further refactors are probably needed implementation side too, as this
removed rather than added units.
Updates #15367
Signed-off-by: James Tucker <james@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 (03b47a55c7956) 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>
I added yet another one in 6d117d64a256234 but that new one is at the
best place int he dependency graph and has the best name, so let's use
that one for everything possible.
types/lazy can't use it for circular dependency reasons, so unexport
that copy at least.
Updates #cleanup
Change-Id: I25db6b6a0d81dbb8e89a0a9080c7f15cbf7aa770
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
We want to be able to use the netx.Network (and RealNetwork
implemementation) outside of tests, without linking "testing".
So split out the non-test stuff of nettest into its own package.
We tend to use "foox" as the convention for things we wish were in the
standard library's foo package, so "netx" seems consistent.
Updates tailscale/corp#27636
Change-Id: I1911d361f4fbdf189837bf629a20f2ebfa863c44
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
This fixes a bug in the local client where the DELETE request was
not being sent correctly. The route was missing a slash before the url
and this now matches the switch profile function.
Signed-off-by: Esteban-Bermudez <esteban@bermudezaguirre.com>
To avoid ephemeral port / TIME_WAIT exhaustion with high --count
values, and to eventually detect leaked connections in tests. (Later
the memory network will register a Cleanup on the TB to verify that
everything's been shut down)
Updates tailscale/corp#27636
Change-Id: Id06f1ae750d8719c5a75d871654574a8226d2733
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
For future in-memory network changes (#15558) to be able to be
stricter and do automatic leak detection when it's safe to do so, in
non-parallel tests.
Updates tailscale/corp#27636
Change-Id: I50f03b16a3f92ce61a7ed88264b49d8c6628f638
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Make the perPeerState objects able to function independently without a
shared reference to the connector.
We don't currently change the values from connector that perPeerState
uses at runtime. Explicitly copying them at perPeerState creation allows
us to, for example, put the perPeerState into a consensus algorithm in
the future.
Updates #14667
Signed-off-by: Fran Bull <fran@tailscale.com>
This shouldn't be necessary, but while we're continuing to figure out
the root cause, this is better than having to restart the app or switch
profiles on the command line.
Updates #15528
Change-Id: Ia101a4a3005adb9118051b3416f5a64a4a45987d
Signed-off-by: Will Norris <will@tailscale.com>
Android >=14 forbids the use of netlink sockets, and in some configurations
can kill apps that try.
Fixes#9836
Signed-off-by: David Anderson <dave@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>
