The intention was always that files only get written to *.partial
files and renamed at the end once fully received, but somewhere in the
process that got lost in buffered mode and *.partial files were only
being used in direct receive mode. This fix prevents WaitingFiles
from returning files that are still being transferred.
Updates tailscale/corp#1626
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
If DeleteFile fails on Windows due to another process (anti-virus,
probably) having our file open, instead leave a marker file that the
file is logically deleted, and remove it from API calls and clean it
up lazily later.
Updates tailscale/corp#1626
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
The old decay-based one took a while to converge. This new one (based
very loosely on TCP BBR) seems to converge quickly on what seems to be
the best speed.
Signed-off-by: Avery Pennarun <apenwarr@tailscale.com>
This tries to generate traffic at a rate that will saturate the
receiver, without overdoing it, even in the event of packet loss. It's
unrealistically more aggressive than TCP (which will back off quickly
in case of packet loss) but less silly than a blind test that just
generates packets as fast as it can (which can cause all the CPU to be
absorbed by the transmitter, giving an incorrect impression of how much
capacity the total system has).
Initial indications are that a syscall about every 10 packets (TCP bulk
delivery) is roughly the same speed as sending every packet through a
channel. A syscall per packet is about 5x-10x slower than that.
The whole tailscale wireguard-go + magicsock + packet filter
combination is about 4x slower again, which is better than I thought
we'd do, but probably has room for improvement.
Note that in "full" tailscale, there is also a tundev read/write for
every packet, effectively doubling the syscall overhead per packet.
Given these numbers, it seems like read/write syscalls are only 25-40%
of the total CPU time used in tailscale proper, so we do have
significant non-syscall optimization work to do too.
Sample output:
$ GOMAXPROCS=2 go test -bench . -benchtime 5s ./cmd/tailbench
goos: linux
goarch: amd64
pkg: tailscale.com/cmd/tailbench
cpu: Intel(R) Core(TM) i7-4785T CPU @ 2.20GHz
BenchmarkTrivialNoAlloc/32-2 56340248 93.85 ns/op 340.98 MB/s 0 %lost 0 B/op 0 allocs/op
BenchmarkTrivialNoAlloc/124-2 57527490 99.27 ns/op 1249.10 MB/s 0 %lost 0 B/op 0 allocs/op
BenchmarkTrivialNoAlloc/1024-2 52537773 111.3 ns/op 9200.39 MB/s 0 %lost 0 B/op 0 allocs/op
BenchmarkTrivial/32-2 41878063 135.6 ns/op 236.04 MB/s 0 %lost 0 B/op 0 allocs/op
BenchmarkTrivial/124-2 41270439 138.4 ns/op 896.02 MB/s 0 %lost 0 B/op 0 allocs/op
BenchmarkTrivial/1024-2 36337252 154.3 ns/op 6635.30 MB/s 0 %lost 0 B/op 0 allocs/op
BenchmarkBlockingChannel/32-2 12171654 494.3 ns/op 64.74 MB/s 0 %lost 1791 B/op 0 allocs/op
BenchmarkBlockingChannel/124-2 12149956 507.8 ns/op 244.17 MB/s 0 %lost 1792 B/op 1 allocs/op
BenchmarkBlockingChannel/1024-2 11034754 528.8 ns/op 1936.42 MB/s 0 %lost 1792 B/op 1 allocs/op
BenchmarkNonlockingChannel/32-2 8960622 2195 ns/op 14.58 MB/s 8.825 %lost 1792 B/op 1 allocs/op
BenchmarkNonlockingChannel/124-2 3014614 2224 ns/op 55.75 MB/s 11.18 %lost 1792 B/op 1 allocs/op
BenchmarkNonlockingChannel/1024-2 3234915 1688 ns/op 606.53 MB/s 3.765 %lost 1792 B/op 1 allocs/op
BenchmarkDoubleChannel/32-2 8457559 764.1 ns/op 41.88 MB/s 5.945 %lost 1792 B/op 1 allocs/op
BenchmarkDoubleChannel/124-2 5497726 1030 ns/op 120.38 MB/s 12.14 %lost 1792 B/op 1 allocs/op
BenchmarkDoubleChannel/1024-2 7985656 1360 ns/op 752.86 MB/s 13.57 %lost 1792 B/op 1 allocs/op
BenchmarkUDP/32-2 1652134 3695 ns/op 8.66 MB/s 0 %lost 176 B/op 3 allocs/op
BenchmarkUDP/124-2 1621024 3765 ns/op 32.94 MB/s 0 %lost 176 B/op 3 allocs/op
BenchmarkUDP/1024-2 1553750 3825 ns/op 267.72 MB/s 0 %lost 176 B/op 3 allocs/op
BenchmarkTCP/32-2 11056336 503.2 ns/op 63.60 MB/s 0 %lost 0 B/op 0 allocs/op
BenchmarkTCP/124-2 11074869 533.7 ns/op 232.32 MB/s 0 %lost 0 B/op 0 allocs/op
BenchmarkTCP/1024-2 8934968 671.4 ns/op 1525.20 MB/s 0 %lost 0 B/op 0 allocs/op
BenchmarkWireGuardTest/32-2 1403702 4547 ns/op 7.04 MB/s 14.37 %lost 467 B/op 3 allocs/op
BenchmarkWireGuardTest/124-2 780645 7927 ns/op 15.64 MB/s 1.537 %lost 420 B/op 3 allocs/op
BenchmarkWireGuardTest/1024-2 512671 11791 ns/op 86.85 MB/s 0.5206 %lost 411 B/op 3 allocs/op
PASS
ok tailscale.com/wgengine/bench 195.724s
Updates #414.
Signed-off-by: Avery Pennarun <apenwarr@tailscale.com>
NetworkManager fixed the bug that forced us to use NetworkManager
if it's programming systemd-resolved, and in the same release also
made NetworkManager ignore DNS settings provided for unmanaged
interfaces... Which breaks what we used to do. So, with versions
1.26.6 and above, we MUST NOT use NetworkManager to indirectly
program systemd-resolved, but thankfully we can talk to resolved
directly and get the right outcome.
Fixes#1788
Signed-off-by: David Anderson <danderson@tailscale.com>
The existing implementation was completely, embarrassingly conceptually broken.
We aren't able to see whether wireguard-go's receive function goroutines
are running or not. All we can do is model that based on what we have done.
This commit fixes that model.
Fixes#1781
Signed-off-by: Josh Bleecher Snyder <josharian@gmail.com>
Avery reported a sub-ms health transition from "receiveIPv4 not running" to "ok".
To avoid these transient false-positives, be more precise about
the expected lifetime of receive funcs. The problematic case is one in which
they were started but exited prior to a call to connBind.Close.
Explicitly represent started vs running state, taking care with the order of updates.
Signed-off-by: Josh Bleecher Snyder <josharian@gmail.com>
The connection failure diagnostic code was never updated enough for
exit nodes, so disable its misleading output when the node it picks
(incorrectly) to diagnose is only an exit node.
Fixes#1754
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
The new "tailscale up" checks previously didn't protect against
--advertise-exit-node being omitted in the case that
--advertise-routes was also provided. It wasn't done before because
there is no corresponding pref for "--advertise-exit-node"; it's a
helper flag that augments --advertise-routes. But that's an
implementation detail and we can still help users. We just have to
special case that pref and look whether the current routes include
both the v4 and v6 /0 routes.
Fixes#1767
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
This doesn't make --operator implicit (which we might do in the
future), but it at least doesn't require repeating it in the future
when it already matches $USER.
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
It was getting cleared on notify.
Document that authURL is cleared on notify and add a new field that
isn't, using the new field for the JSON status.
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
I've spent two days searching for a theoretical wireguard-go bug
around receive functions exiting early.
I've found many bugs, but none of the flavor we're looking for.
Restore wireguard-go's logging around starting and stopping receive functions,
so that we can definitively rule in or out this particular theory.
Signed-off-by: Josh Bleecher Snyder <josharian@gmail.com>
I see a bunch of these in some logs I'm looking at,
separated only by a few seconds.
Log the error so we can tell what's going on here.
Signed-off-by: Josh Bleecher Snyder <josharian@gmail.com>
These were getting rate-limited for nodes with many peers.
Consolate the output into single lines, which are nicer anyway.
Signed-off-by: Josh Bleecher Snyder <josharian@gmail.com>
With this change, the ipnserver's safesocket.Listen (the localhost
tcp.Listen) happens right away, before any synchronous
TUN/DNS/Engine/etc setup work, which might be slow, especially on
early boot on Windows.
Because the safesocket.Listen starts up early, that means localhost
TCP dials (the safesocket.Connect from the GUI) complete successfully
and thus the GUI avoids the MessageBox error. (I verified that
pacifies it, even without a Listener.Accept; I'd feared that Windows
localhost was maybe special and avoided the normal listener backlog).
Once the GUI can then connect immediately without errors, the various
timeouts then matter less, because the backend is no longer trying to
race against the GUI's timeout. So keep retrying on errors for a
minute, or 10 minutes if the system just booted in the past 10
minutes.
This should fix the problem with Windows 10 desktops auto-logging in
and starting the Tailscale frontend which was then showing a
MessageBox error about failing to connect to tailscaled, which was
slow coming up because the Windows networking stack wasn't up
yet. Fingers crossed.
Fixes#1313 (previously #1187, etc)
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
This change implements Windows version of install-system-daemon and
uninstall-system-daemon subcommands. When running the commands the
user will install or remove Tailscale Windows service.
Updates #1232
Signed-off-by: Alex Brainman <alex.brainman@gmail.com>
This used to not be necessary, because MagicDNS always did full proxying.
But with split DNS, we need to know which names to route to our resolver,
otherwise reverse lookups break.
This captures the entire CGNAT range, as well as our Tailscale ULA.
Signed-off-by: David Anderson <danderson@tailscale.com>
Otherwise, the existence of authoritative domains forces full
DNS proxying even when no other DNS config is present.
Signed-off-by: David Anderson <danderson@tailscale.com>
