We saw a few cases where we hit this limit; bumping to 4k seems
relatively uncontroversial.
Change-Id: I218fee3bc0d2fa5fde16eddc36497a73ebd7cbda
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
We change our invocations of GetExtendedTcpTable to request additional
information about the "module" responsible for the port. In addition to pid,
this output also includes sufficient metadata to enable Windows to resolve
process names and disambiguate svchost processes.
We store the OS-specific output in an OSMetadata field in netstat.Entry, which
portlist may then use as necessary to actually resolve the process/module name.
Signed-off-by: Aaron Klotz <aaron@tailscale.com>
The Tailscale logging service has a hard limit on the maximum
log message size that can be accepted.
We want to ensure that netlog messages never exceed
this limit otherwise a client cannot transmit logs.
Move the goroutine for periodically dumping netlog messages
from wgengine/netlog to net/connstats.
This allows net/connstats to manage when it dumps messages,
either based on time or by size.
Updates tailscale/corp#8427
Signed-off-by: Joe Tsai <joetsai@digital-static.net>
Previously, if a DNS-over-TCP message was received while there were
existing queries in-flight, and it was over the size limit, we'd close
the 'responses' channel. This would cause those in-flight queries to
send on the closed channel and panic.
Instead, don't close the channel at all and rely on s.ctx being
canceled, which will ensure that in-flight queries don't hang.
Fixes#6725
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I8267728ac37ed7ae38ddd09ce2633a5824320097
This is temporary while we work to upstream performance work in
https://github.com/WireGuard/wireguard-go/pull/64. A replace directive
is less ideal as it breaks dependent code without duplication of the
directive.
Signed-off-by: Jordan Whited <jordan@tailscale.com>
We would call parsedPacketPool.Get() for all packets received in Read/Write.
This was wasteful and not necessary, fetch a single *packet.Parsed for
all packets.
Signed-off-by: Maisem Ali <maisem@tailscale.com>
This commit updates the wireguard-go dependency and implements the
necessary changes to the tun.Device and conn.Bind implementations to
support passing vectors of packets in tailscaled. This significantly
improves throughput performance on Linux.
Updates #414
Signed-off-by: Jordan Whited <jordan@tailscale.com>
Signed-off-by: James Tucker <james@tailscale.com>
Co-authored-by: James Tucker <james@tailscale.com>
x/exp/slices now has ContainsFunc (golang/go#53983) so we can delete
our versions.
Change-Id: I5157a403bfc1b30e243bf31c8b611da25e995078
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
We were previously only doing this for tailscaled-on-Darwin, but it also
appears to help on iOS. Otherwise, when we rebind magicsock UDP
connections after a cellular -> WiFi interface change they still keep
using cellular one.
To do this correctly when using exit nodes, we need to exclude the
Tailscale interface when getting the default route, otherwise packets
cannot leave the tunnel. There are native macOS/iOS APIs that we can
use to do this, so we allow those clients to override the implementation
of DefaultRouteInterfaceIndex.
Updates #6565, may also help with #5156
Signed-off-by: Mihai Parparita <mihai@tailscale.com>
I couldn't find any logs that indicated which mode it was running in so adding that.
Also added a gauge metric for dnsMode.
Signed-off-by: Maisem Ali <maisem@tailscale.com>
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>
The fix in 4fc8538e2 was sufficient for IPv6. Browsers (can?) send the
IPv6 literal, even without a port number, in brackets.
Updates tailscale/corp#7948
Change-Id: I0e429d3de4df8429152c12f251ab140b0c8f6b77
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
There was a mechanism in tshttpproxy to note that a Windows proxy
lookup failed and to stop hitting it so often. But that turns out to
fire a lot (no PAC file configured at all results in a proxy lookup),
so after the first proxy lookup, we were enabling the "omg something's
wrong, stop looking up proxies" bit for awhile, which was then also
preventing the normal Go environment-based proxy lookups from working.
This at least fixes environment-based proxies.
Plenty of other Windows-specific proxy work remains (using
WinHttpGetIEProxyConfigForCurrentUser instead of just PAC files,
ignoring certain types of errors, etc), but this should fix
the regression reported in #4811.
Updates #4811
Change-Id: I665e1891897d58e290163bda5ca51a22a017c5f9
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Run an inotify goroutine and watch if another program takes over
/etc/inotify.conf. Log if so.
For now this only logs. In the future I want to wire it up into the
health system to warn (visible in "tailscale status", etc) about the
situation, with a short URL to more info about how you should really
be using systemd-resolved if you want programs to not fight over your
DNS files on Linux.
Updates #4254 etc etc
Change-Id: I86ad9125717d266d0e3822d4d847d88da6a0daaa
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Noticed this while debugging something else, we would reset all routes if
either `--advertise-exit-node` or `--advertise-routes` were set. This handles
correctly updating them.
Also added tests.
Signed-off-by: Maisem Ali <maisem@tailscale.com>
The derpers don't allow whitespace in the challenge.
Change-Id: I93a8b073b846b87854fba127b5c1d80db205f658
Signed-off-by: Andrew Dunham <andrew@tailscale.com>
The //go:build syntax was introduced in Go 1.17:
https://go.dev/doc/go1.17#build-lines
gofmt has kept the +build and go:build lines in sync since
then, but enough time has passed. Time to remove them.
Done with:
perl -i -npe 's,^// \+build.*\n,,' $(git grep -l -F '+build')
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
To collect some data on how widespread this is and whether there's
any correlation between different versions of Windows, etc.
Updates #4811
Change-Id: I003041d0d7e61d2482acd8155c1a4ed413a2c5c4
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Instead of returning a custom error, use ErrGetBaseConfigNotSupported
that seems to be intended for this use case. This fixes DNS resolution
on macOS clients compiled from source.
Signed-off-by: Anton Tolchanov <anton@tailscale.com>
The netlog.Message type is useful to depend on from other packages,
but doing so would transitively cause gvisor and other large packages
to be linked in.
Avoid this problem by moving all network logging types to a single package.
We also update staticcheck to take in:
003d277bcf
Signed-off-by: Joe Tsai <joetsai@digital-static.net>
It looks like this was left by mistake in 4a3e2842.
Change-Id: Ie4e3d5842548cd2e8533b3552298fb1ce9ba761a
Signed-off-by: Andrew Dunham <andrew@tailscale.com>
On Android, the system resolver can return IPv4 addresses as IPv6-mapped
addresses (i.e. `::ffff:a.b.c.d`). After the switch to `net/netip`
(19008a3), this case is no longer handled and a response like this will
be seen as failure to resolve any IPv4 addresses.
Handle this case by simply calling `Unmap()` on the returned IPs. Fixes#5698.
Signed-off-by: Peter Cai <peter@typeblog.net>
We had previously added this to the netcheck report in #5087 but never
copied it into the NetInfo struct. Additionally, add it to log lines so
it's visible to support.
Change-Id: Ib6266f7c6aeb2eb2a28922aeafd950fe1bf5627e
Signed-off-by: Andrew Dunham <andrew@tailscale.com>
The ResolvConfMode property is documented to return how systemd-resolved
is currently managing /etc/resolv.conf. Include that information in the
debug line, when available, to assist in debugging DNS issues.
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I1ae3a257df1d318d0193a8c7f135c458ec45093e
The Lufthansa in-flight wifi generates a synthetic 204 response to the
DERP server's /generate_204 endpoint. This PR adds a basic
challenge/response to the endpoint; something sufficiently complicated
that it's unlikely to be implemented by a captive portal. We can then
check for the expected response to verify whether we're being MITM'd.
Follow-up to #5601
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I94a68c9a16a7be7290200eea6a549b64f02ff48f
Instead of treating any interface with a non-ifscope route as a
potential default gateway, now verify that a given route is
actually a default route (0.0.0.0/0 or ::/0).
Fixes#5879
Signed-off-by: Anton Tolchanov <anton@tailscale.com>
We removed it in #4806 in favor of the built-in functionality from the
nhooyr.io/websocket package. However, it has an issue with deadlines
that has not been fixed yet (see nhooyr/websocket#350). Temporarily
go back to using a custom wrapper (using the fix from our fork) so that
derpers will stop closing connections too aggressively.
Updates #5921
Signed-off-by: Mihai Parparita <mihai@tailscale.com>
If netcheck happens before there's a derpmap.
This seems to only affect Headscale because it doesn't send a derpmap
as early?
Change-Id: I51e0dfca8e40623e04702bc9cc471770ca20d2c2
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
The Logger type managers a logtail.Logger for extracting
statistics from a tstun.Wrapper.
So long as Shutdown is called, it ensures that logtail
and statistic gathering resources are properly cleared up.
Signed-off-by: Joe Tsai <joetsai@digital-static.net>
Rename StatisticsEnable as SetStatisticsEnabled to be consistent
with other similarly named methods.
Rename StatisticsExtract as ExtractStatistics to follow
the convention where methods start with a verb.
It was originally named with Statistics as a prefix so that
statistics related methods would sort well in godoc,
but that property no longer holds.
Signed-off-by: Joe Tsai <joetsai@digital-static.net>
If Wrapper.StatisticsEnable is enabled,
then per-connection counters are maintained.
If enabled, Wrapper.StatisticsExtract must be periodically called
otherwise there is unbounded memory growth.
Signed-off-by: Joe Tsai <joetsai@digital-static.net>
High-level API:
type Statistics struct { ... }
type Counts struct { TxPackets, TxBytes, RxPackets, RxBytes uint64 }
func (*Statistics) UpdateTx([]byte)
func (*Statistics) UpdateRx([]byte)
func (*Statistics) Extract() map[flowtrack.Tuple]Counts
The API accepts a []byte instead of a packet.Parsed so that a future
implementation can directly hash the address and port bytes,
which are contiguous in most IP packets.
This will be useful for a custom concurrent-safe hashmap implementation.
Signed-off-by: Joe Tsai <joetsai@digital-static.net>
Most visible when using tsnet.Server, but could have resulted in dropped
messages in a few other places too.
Fixes#5743
Signed-off-by: Mihai Parparita <mihai@tailscale.com>
I brain-o'ed the math earlier. The NextDNS prefix is /32 (actually
/33, but will guarantee last bit is 0), so we have 128-32 = 96 bits
(12 bytes) of config/profile ID that we can extract. NextDNS doesn't
currently use all those, but might.
Updates #2452
Change-Id: I249bd28500c781e45425fd00fd3f46893ae226a2
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
- removed some in-flow time calls
- increase buffer size to 2MB to overcome syscall cost
- move relative time computation from record to report time
Signed-off-by: James Tucker <james@tailscale.com>
The fragment offset is an 8 byte offset rather than a byte offset, so
the short packet limit is now in fragment block size in order to compare
with the offset value.
The packet flags are in the first 3 bits of the flags/frags byte, and
so after conversion to a uint16 little endian value they are at the
start, not the end of the value - the mask for extracting "more
fragments" is adjusted to match this byte.
Extremely short fragments less than 80 bytes are dropped, but fragments
over 80 bytes are now accepted.
Fixes#5727
Signed-off-by: James Tucker <james@tailscale.com>
This doesn't change any behaviour for now, other than maybe running a
full netcheck more often. The intent is to start gathering data on
captive portals, and additionally, seeing this in the 'tailscale
netcheck' command should provide a bit of additional information to
users.
Updates #1634
Change-Id: I6ba08f9c584dc0200619fa97f9fde1a319f25c76
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
The io/ioutil package has been deprecated as of Go 1.16 [1]. This commit
replaces the existing io/ioutil functions with their new definitions in
io and os packages.
Reference: https://golang.org/doc/go1.16#ioutil
Signed-off-by: Eng Zer Jun <engzerjun@gmail.com>
The plan has changed. Doing query parameters rather than path +
heades. NextDNS added support for query parameters.
Updates #2452
Change-Id: I4783c0a06d6af90756d9c80a7512644ba702388c
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
For debugging a macOS-specific magicsock issue. macOS runs in
bind-to-interface mode always. This lets me force Linux into the same
mode as macOS, even if the Linux kernel supports SO_MARK, as it
usually does.
Updates #2331 etc
Change-Id: Iac9e4a7429c1781337e716ffc914443b7aa2869d
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Clarify & verify that some DoH URLs can be sent over tailcfg
in some limited cases.
Updates #2452
Change-Id: Ibb25db77788629c315dc26285a1059a763989e24
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
NextDNS is unique in that users create accounts and then get
user-specific DNS IPs & DoH URLs.
For DoH, the customer ID is in the URL path.
For IPv6, the IP address includes the customer ID in the lower bits.
For IPv4, there's a fragile "IP linking" mechanism to associate your
public IPv4 with an assigned NextDNS IPv4 and that tuple maps to your
customer ID.
We don't use the IP linking mechanism.
Instead, NextDNS is DoH-only. Which means using NextDNS necessarily
shunts all DNS traffic through 100.100.100.100 (programming the OS to
use 100.100.100.100 as the global resolver) because operating systems
can't usually do DoH themselves.
Once it's in Tailscale's DoH client, we then connect out to the known
NextDNS IPv4/IPv6 anycast addresses.
If the control plane sends the client a NextDNS IPv6 address, we then
map it to the corresponding NextDNS DoH with the same client ID, and
we dial that DoH server using the combination of v4/v6 anycast IPs.
Updates #2452
Change-Id: I3439d798d21d5fc9df5a2701839910f5bef85463
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
This is especially helpful as we launch newer DERPs over time, and older
clients have progressively out-of-date static DERP maps baked in. After
this, as long as the client has successfully connected once, it'll cache
the most recent DERP map it knows about.
Resolves an in-code comment from @bradfitz
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
If ExtraRecords (Hosts) are specified without a corresponding split
DNS route and global DNS is specified, then program the host OS DNS to
use 100.100.100.100 so it can blend in those ExtraRecords.
Updates #1543
Change-Id: If49014a5ecc8e38978ff26e54d1f74fe8dbbb9bc
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Somehow I accidentally set the wrong registry value here.
It should be DisableDynamicUpdate=1 and not EnableDNSUpdate=0.
This is a regression from 545639e.
Signed-off-by: Maisem Ali <maisem@tailscale.com>
See https://mullvad.net/en/help/dns-over-https-and-dns-over-tls/
The Mullvad DoH servers appear to only speak HTTP/2 and
the use of a non-nil DialContext in the http.Transport
means that ForceAttemptHTTP2 must be set to true to be
able to use them.
Signed-off-by: Nahum Shalman <nahamu@gmail.com>
This works around the 2.3s delay in short name lookups when SNR is
enabled.
C:\Windows\System32\drivers\etc\hosts file. We only add known hosts that
match the search domains, and we populate the list in order of
Search Domains so that our matching algorithm mimics what Windows would
otherwise do itself if SNR was off.
Updates #1659
Signed-off-by: Maisem Ali <maisem@tailscale.com>
Convert ParseResponse and Response to use netip.AddrPort instead of
net.IP and separate port.
Fixes#5281
Signed-off-by: Kris Brandow <kris.brandow@gmail.com>
Like LLMNR, NetBIOS also adds resolution delays and we don't support it
anyway so just disable it on the interface.
Updates #1659
Signed-off-by: Maisem Ali <maisem@tailscale.com>
Currently we forward unmatched queries to the default resolver on
Windows. This results in duplicate queries being issued to the same
resolver which is just wasted.
Updates #1659
Signed-off-by: Maisem Ali <maisem@tailscale.com>
Otherwise we just keep looping over the same thing again and again.
```
dns udp query: upstream nameservers not set
dns udp query: upstream nameservers not set
dns udp query: upstream nameservers not set
```
Signed-off-by: Maisem Ali <maisem@tailscale.com>
