BPF links require that the owning FD remains open, this FD is embedded
into the RawLink returned by the attach function and must live for the
duration of the server.
Updates ENG-4274
Signed-off-by: James Tucker <james@tailscale.com>
Detection of duplicate Network Lock signature chains added in
01847e0123 failed to account for chains
originating with a SigCredential signature, which is used for wrapped
auth keys. This results in erroneous removal of signatures that
originate from the same re-usable auth key.
This change ensures that multiple nodes created by the same re-usable
auth key are not getting filtered out by the network lock.
Updates tailscale/corp#19764
Signed-off-by: Anton Tolchanov <anton@tailscale.com>
This change moves handling of wrapped auth keys to the `tka` package and
adds a test covering auth key originating signatures (SigCredential) in
netmap.
Updates tailscale/corp#19764
Signed-off-by: Anton Tolchanov <anton@tailscale.com>
When auto-udpates are enabled, we don't need to nag users to update
after a new release, before we release auto-updates.
Updates https://github.com/tailscale/corp/issues/20081
Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
This is not valid in many situations, specifically when running a local astro site that listens on localhost, but ignores 127.0.0.1
Fixes: https://github.com/tailscale/tailscale/issues/12201
Signed-off-by: Josh McKinney <joshka@users.noreply.github.com>
In hopes it'll be found more.
Updates tailscale/corp#20844
Change-Id: Ic92ee9908f45b88f8770de285f838333f9467465
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
A few other minor language updates.
Updates tailscale/corp#20844
Change-Id: Idba85941baa0e2714688cc8a4ec3e242e7d1a362
Signed-off-by: James Tucker <james@tailscale.com>
And some misc doc tweaks for idiomatic Go style.
Updates #cleanup
Change-Id: I3ca45f78aaca037f433538b847fd6a9571a2d918
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
We cannot directly pass a flat domain name into NetUserGetInfo; we must
resolve the address of a domain controller first.
This PR implements the appropriate resolution mechanisms to do that, and
also exposes a couple of new utility APIs for future needs.
Fixes#12627
Signed-off-by: Aaron Klotz <aaron@tailscale.com>
This turns the checklocks workflow into a real check, and adds
annotations to a few basic packages as a starting point.
Updates #12625
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I2b0185bae05a843b5257980fc6bde732b1bdd93f
The exit node suggestion CLI command was written with the assumption
that it's possible to provide a stableid on the command line, but this
is incorrect. Instead, it will now emit the name of the exit node.
Fixes#12618
Change-Id: Id7277f395b5fca090a99b0d13bfee7b215bc9802
Signed-off-by: Adrian Dewhurst <adrian@tailscale.com>
The context can get canceled during backoff, and binding after that
makes the listener impossible to close afterwards.
Fixes#12620.
Signed-off-by: Naman Sood <mail@nsood.in>
To complement the existing `onCompletion` callback, which is called
after request handler.
Updates tailscale/corp#17075
Signed-off-by: Anton Tolchanov <anton@tailscale.com>
We can observe a data race in tests when logging after a test is
finished. `b.onHealthChange` is called in a goroutine after being
registered with `health.Tracker.RegisterWatcher`, which calls callbacks
in `setUnhealthyLocked` in a new goroutine.
See: https://github.com/tailscale/tailscale/actions/runs/9672919302/job/26686038740
Updates #12054
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: Ibf22cc994965d88a9e7236544878d5373f91229e
This PR ties together pseudoconsoles, user profiles, s4u logons, and
process creation into what is (hopefully) a simple API for various
Tailscale services to obtain Windows access tokens without requiring
knowledge of any Windows passwords. It works both for domain-joined
machines (Kerberos) and non-domain-joined machines. The former case
is fairly straightforward as it is fully documented. OTOH, the latter
case is not documented, though it is fully defined in the C headers in
the Windows SDK. The documentation blanks were filled in by reading
the source code of Microsoft's Win32 port of OpenSSH.
We need to do a bit of acrobatics to make conpty work correctly while
creating a child process with an s4u token; see the doc comments above
startProcessInternal for details.
Updates #12383
Signed-off-by: Aaron Klotz <aaron@tailscale.com>
Previously, if we had a umask set (e.g. 0027) that prevented creating a
world-readable file, /etc/resolv.conf would be created without the o+r
bit and thus other users may be unable to resolve DNS.
Since a umask only applies to file creation, chmod the file after
creation and before renaming it to ensure that it has the appropriate
permissions.
Updates #12609
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I2a05d64f4f3a8ee8683a70be17a7da0e70933137
The logic we added in #11378 would prevent selecting a home DERP if we
have no control connection.
Updates tailscale/corp#18095
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I44bb6ac4393989444e4961b8cfa27dc149a33c6e
Fixestailscale/corp#20677
Replaces the original attempt to rectify this (by injecting a netMon
event) which was both heavy handed, and missed cases where the
netMon event was "minor".
On apple platforms, the fetching the interface's nameservers can
and does return an empty list in certain situations. Apple's API
in particular is very limiting here. The header hints at notifications
for dns changes which would let us react ahead of time, but it's all
private APIs.
To avoid remaining in the state where we end up with no
nameservers but we absolutely need them, we'll react
to a lack of upstream nameservers by attempting to re-query
the OS.
We'll rate limit this to space out the attempts. It seems relatively
harmless to attempt a reconfig every 5 seconds (triggered
by an incoming query) if the network is in this broken state.
Missing nameservers might possibly be a persistent condition
(vs a transient error), but that would also imply that something
out of our control is badly misconfigured.
Tested by randomly returning [] for the nameservers. When switching
between Wifi networks, or cell->wifi, this will randomly trigger
the bug, and we appear to reliably heal the DNS state.
Signed-off-by: Jonathan Nobels <jonathan@tailscale.com>
So non-local users (e.g. Kerberos on FreeIPA) on Linux can be looked
up. Our default binaries are built with pure Go os/user which only
supports the classic /etc/passwd and not any libc-hooked lookups.
Updates #12601
Change-Id: I9592db89e6ca58bf972f2dcee7a35fbf44608a4f
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
stunstamp now sends data to Prometheus via remote write, and Prometheus
can serve the same data. Retaining and cleaning up old data in sqlite
leads to long probing pauses, and it's not worth investing more effort
to optimize the schema and/or concurrency model.
Updates tailscale/corp#20344
Signed-off-by: Jordan Whited <jordan@tailscale.com>
PeerPresentFlags was added in 5ffb2668ef but wasn't plumbed through to
the RunConnectionLoop. Rather than add yet another parameter (as
IP:port was added earlier), pass in the raw PeerPresentMessage and
PeerGoneMessage struct values, which are the same things, plus two
fields: PeerGoneReasonType for gone and the PeerPresentFlags from
5ffb2668ef.
Updates tailscale/corp#17816
Change-Id: Ib19d9f95353651ada90656071fc3656cf58b7987
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
When the store-appc-routes flag is on for a tailnet we are writing the
routes more often than seems necessary. Investigation reveals that we
are doing so ~every time we observe a dns response, even if this causes
us not to advertise any new routes. So when we have no new routes,
instead do not advertise routes.
Fixes#12593
Signed-off-by: Fran Bull <fran@tailscale.com>
I couldn't convince myself the old way was safe and couldn't lose
writes.
And it seemed too complicated.
Updates tailscale/corp#21104
Change-Id: I17ba7c7d6fd83458a311ac671146a1f6a458a5c1
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
sendMeshUpdates tries to write as much as possible without blocking,
being careful to check the bufio.Writer.Available size before writes.
Except that regressed in 6c791f7d60 which made those messages larger, which
meants we were doing network I/O with the Server mutex held.
Updates tailscale/corp#13945
Change-Id: Ic327071d2e37de262931b9b390cae32084811919
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
This adds the ability to "peek" at the value of a SyncValue, so that
it's possible to observe a value without computing this.
Updates tailscale/corp#17122
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Co-authored-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Change-Id: I06f88c22a1f7ffcbc7ff82946335356bb0ef4622
This is implemented via GetBestInterfaceEx. Should we encounter errors
or fail to resolve a valid, non-Tailscale interface, we fall back to
returning the index for the default interface instead.
Fixes#12551
Signed-off-by: Aaron Klotz <aaron@tailscale.com>
Timeouts could already be identified as NaN values on
stunstamp_derp_stun_rtt_ns, but we can't use NaN effectively with
promql to visualize them. So, this commit adds a timeouts metric that
we can use with rate/delta/etc promql functions.
Updates tailscale/corp#20689
Signed-off-by: Jordan Whited <jordan@tailscale.com>
Changes "Accept" TCP logs to display in verbose logs only,
and removes lines from default logging behavior.
Updates #12158
Signed-off-by: Keli Velazquez <keli@tailscale.com>
