Commit Graph

862 Commits (058d427fa60143347d13eeb187a7978e67bfe7ff)

Author SHA1 Message Date
David Anderson 52212f4323 all: update exp/slices and fix call sites
slices.SortFunc suffered a late-in-cycle API breakage.

Updates #cleanup

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 6afffece8a net/art: use more intelligible, go-ish state tracking in table.Get
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 4f14ed2ad6 net/art: use encoding/binary for address bit twiddling
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson f1cd67488d net/art: move slice closer to its use
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 44ad7b3746 net/art: factor out picking the right strideTable for addr family
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 125b982ba5 net/art: make Table.Get alloc-free
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson b76d8a88ae net/art: document return value of strideTable.delete
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson b242e2c2cb net/art: reword confusing function docstring
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 8478358d77 net/art: use "index", not "idx" in function names
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson de5c6ed4be net/art: document valid values of strideTable.prefix
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 736a44264f net/art: fix comment typo
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 1e6f0bb608 net/art: fix slowPrefixTable bugs found by fuzzing
One is a straight "I forgot how to Go" bug, the others are semantic
mismatches with the main implementation around masking the prefixes
passed to insert/delete.

Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson aaca911904 net/art: add another consistency test for insert/delete
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson b145a22f55 net/art: add more exhaustive table testing
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 9cc3f7a3d6 net/art: fix format of debug output
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson ac657caaf1 net/art: add debug hooks to strideTable
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson fcf4d044fa net/art: implement path compression optimization
Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 486195edf0 net/art: make each strideTable track the IP prefix it represents
This is a prerequisite for path compression, so that insert/delete
can determine when compression occurred.

Updates #7781

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 45b5d0983c net/art: fix running tests outside of CI
Updates #7866

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
Michael Stapelberg 2a6c237d4c net/dns: overwrite /tmp/resolv.conf on gokrazy
Appliances built using https://gokrazy.org/ have a read-only root file system,
including /etc/resolv.conf, which is a symlink to /tmp/resolv.conf.

The system’s dhcp client overwrites /tmp/resolv.conf instead,
so we need to use this path in Tailscale, too.

related to https://github.com/gokrazy/gokrazy/issues/209

fixes https://github.com/tailscale/tailscale/issues/8689

Signed-off-by: Michael Stapelberg <michael@stapelberg.de>
1 year ago
Anton Tolchanov 388b124513 net/dns: detect when libnss_resolve is used
Having `127.0.0.53` is not the only way to use `systemd-resolved`. An
alternative way is to enable `libnss_resolve` module, which seems to now
be used by default on Debian 12 bookworm.

Fixes #8549

Signed-off-by: Anton Tolchanov <anton@tailscale.com>
1 year ago
Andrew Dunham 9b5e29761c net/netcheck: ignore PreferredDERP changes that are small
If the absolute value of the difference between the current
PreferredDERP's latency and the best latency is <= 10ms, don't change
it and instead prefer the previous value.

This is in addition to the existing hysteresis that tries to remain
on the previous DERP region if the relative improvement is small, but
handles nodes that have low latency to >1 DERP region better.

Updates #8603

Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I1e34c94178f8c9a68a69921c5bc0227337514c70
1 year ago
Andrew Dunham 7aba0b0d78 net/netcheck, tailcfg: add DERPHomeParams and use it
This allows providing additional information to the client about how to
select a home DERP region, such as preferring a given DERP region over
all others.

Updates #8603

Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I7c4a270f31d8585112fab5408799ffba5b75266f
1 year ago
Andrew Dunham a7648a6723 net/dnsfallback: run recursive resolver and compare results
When performing a fallback DNS query, run the recursive resolver in a
separate goroutine and compare the results returned by the recursive
resolver with the results we get from "regular" bootstrap DNS. This will
allow us to gather data about whether the recursive DNS resolver works
better, worse, or about the same as "regular" bootstrap DNS.

Updates #5853

Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: Ifa0b0cc9eeb0dccd6f7a3d91675fe44b3b34bd48
1 year ago
Adrian Dewhurst cd4c71c122 tstest: prepare for Clock API changes
This change introduces tstime.NewClock and tstime.ClockOpts as a new way
to construct tstime.Clock. This is a subset of #8464 as a stepping stone
so that we can update our internal code to use the new API before making
the second round of changes.

Updates #8463

Change-Id: Ib26edb60e5355802aeca83ed60e4fdf806c90e27
Signed-off-by: Adrian Dewhurst <adrian@tailscale.com>
1 year ago
Andrew Dunham d9eca20ee2 net/tcpinfo: add package to allow fetching TCP information
This package contains platform-independent abstractions for fetching
information about an open TCP connection.

Updates #8413

Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I236657b1060d7e6a45efc7a2f6aacf474547a2fe
1 year ago
KevinLiang10 243ce6ccc1 util/linuxfw: decoupling IPTables logic from linux router
This change is introducing new netfilterRunner interface and moving iptables manipulation to a lower leveled iptables runner.

For #391

Signed-off-by: KevinLiang10 <kevinliang@tailscale.com>
1 year ago
Ross Zurowski 832f1028c7
net/netutil: parse IP forwarding val as int, not bool (#8455)
This commit updates our IP forwarding parsing logic to allow the less
common but still valid value of `2` to be parsed as `true`, which fixes
an error some users encountered.

Fixes #8375

Signed-off-by: Ross Zurowski <ross@rosszurowski.com>
1 year ago
Brad Fitzpatrick a874f1afd8 all: adjust case of "IPv4" and "IPv6"
Updates #docs

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Andrew Dunham b6d20e6f8f go.mod, net/dns/recursive: update github.com/miekg/dns
Updates #cleanup

Change-Id: If4de6a84448a17dd81cc2a8af788bd18c3d0bbe3
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
1 year ago
Brad Fitzpatrick a076213f58 net/memnet: add optional Listener.NewConn config knob
Updates tailscale/corp#11620

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 1543e233e6 net/tstun, tsnet: make happier on WASI
Also fix a js/wasm issue with tsnet in the process. (same issue as WASI)

Updates #8320
Fixes #8315

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Andrew Dunham f077b672e4 net/dns/recursive: add initial implementation of recursive DNS resolver
We've talked in the past about reworking how bootstrap DNS works to
instead do recursive DNS resolution from the root; this would better
support on-prem customers and Headscale users where the DERP servers
don't currently resolve their DNS server. This package is an initial
implementation of recursive resolution for A and AAAA records.

Updates #5853

Change-Id: Ibe974d78709b4b03674b47c4ef61f9a00addf8b4
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
1 year ago
Brad Fitzpatrick eefee6f149 all: use cmpx.Or where it made sense
I left a few out where writing it explicitly was better
for various reasons.

Updates #8296

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Denton Gentry 64f16f7f38 net/dnscache: use PreferGo on Windows.
Updates https://github.com/tailscale/tailscale/issues/5161

Signed-off-by: Denton Gentry <dgentry@tailscale.com>
1 year ago
James Tucker da90fab899 net/netcheck: reenable TestBasic on Windows
This test was either fixed by intermediate changes or was mis-flagged as
failing during #7876 triage.

Updates #7876

Signed-off-by: James Tucker <jftucker@gmail.com>
2 years ago
Will Norris ea84fc9ad2 net/sockstats: wait before reporting battery usage
Wait 2 minutes before we start reporting battery usage. There is always
radio activity on initial startup, which gets reported as 100% high
power usage.  Let that settle before we report usage data.

Updates tailscale/corp#9230

Signed-off-by: Will Norris <will@tailscale.com>
2 years ago
James Tucker 9ed3a061c3 net/netns: fix segv when no tailscale interface is found
`interfaces.Tailscale()` returns all zero values when it finds no
Tailscale interface and encounters no errors. The netns package was
treating no error as a signal that it would receive a non-zero pointer
value leading to nil pointer dereference.

Observed in:

```
--- FAIL: TestGetInterfaceIndex (0.00s)
    --- FAIL: TestGetInterfaceIndex/IP_and_port (0.00s)
panic: runtime error: invalid memory address or nil pointer dereference [recovered]
        panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x2 addr=0x0 pc=0x1029eb7d8]

goroutine 7 [running]:
testing.tRunner.func1.2({0x102a691e0, 0x102bc05c0})
        /Users/raggi/.cache/tailscale-go/src/testing/testing.go:1526 +0x1c8
testing.tRunner.func1()
        /Users/raggi/.cache/tailscale-go/src/testing/testing.go:1529 +0x384
panic({0x102a691e0, 0x102bc05c0})
        /Users/raggi/.cache/tailscale-go/src/runtime/panic.go:884 +0x204
tailscale.com/net/netns.getInterfaceIndex(0x14000073f28, 0x1028d0284?, {0x1029ef3b7, 0xa})
        /Users/raggi/src/github.com/tailscale/tailscale/net/netns/netns_darwin.go:114 +0x228
tailscale.com/net/netns.TestGetInterfaceIndex.func2(0x14000138000)
        /Users/raggi/src/github.com/tailscale/tailscale/net/netns/netns_darwin_test.go:37 +0x54
testing.tRunner(0x14000138000, 0x140000551b0)
        /Users/raggi/.cache/tailscale-go/src/testing/testing.go:1576 +0x10c
created by testing.(*T).Run
        /Users/raggi/.cache/tailscale-go/src/testing/testing.go:1629 +0x368
FAIL    tailscale.com/net/netns 0.824s
```

Fixes #8064

Signed-off-by: James Tucker <jftucker@gmail.com>
2 years ago
Brad Fitzpatrick 6e967446e4 tsd: add package with System type to unify subsystem init, discovery
This is part of an effort to clean up tailscaled initialization between
tailscaled, tailscaled Windows service, tsnet, and the mac GUI.

Updates #8036

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Maisem Ali 3ae7140690 net/tstun: handle exit nodes in NAT configs
In the case where the exit node requires SNAT, we would SNAT all traffic not just the
traffic meant to go through the exit node. This was a result of the default route being
added to the routing table which would match basically everything.

In this case, we need to account for all peers in the routing table not just the ones
that require NAT.

Fix and add a test.

Updates tailscale/corp#8020

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Charlotte Brandhorst-Satzkorn 161d1d281a
net/ping,netcheck: add v6 pinging capabilities to pinger (#7971)
This change adds a v6conn to the pinger to enable sending pings to v6
addrs.

Updates #7826

Signed-off-by: Charlotte Brandhorst-Satzkorn <charlotte@tailscale.com>
2 years ago
Maisem Ali f58751eb2b net/packet: add NAT support for DCCP and GRE
Updates tailscale/corp#8020

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Brad Fitzpatrick 90ba26cea1 net/netcheck: fix crash when IPv6 kinda but not really works
Looks like on some systems there's an IPv6 address, but then opening
a IPv6 UDP socket fails later. Probably some firewall. Tolerate it
better and don't crash.

To repro: check the "udp6" to something like "udp7" (something that'll
fail) and run "go run ./cmd/tailscale netcheck" on a machine with
active IPv6. It used to crash and now it doesn't.

Fixes #7949

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Andrew Dunham 04a3118d45 net/tstun: add tests for captureHook
Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I630f852d9f16c951c721b34f2bc4128e68fe9475
2 years ago
Mihai Parparita 7330aa593e all: avoid repeated default interface lookups
On some platforms (notably macOS and iOS) we look up the default
interface to bind outgoing connections to. This is both duplicated
work and results in logspam when the default interface is not available
(i.e. when a phone has no connectivity, we log an error and thus cause
more things that we will try to upload and fail).

Fixed by passing around a netmon.Monitor to more places, so that we can
use its cached interface state.

Fixes #7850
Updates #7621

Signed-off-by: Mihai Parparita <mihai@tailscale.com>
2 years ago
Mihai Parparita 4722f7e322 all: move network monitoring from wgengine/monitor to net/netmon
We're using it in more and more places, and it's not really specific to
our use of Wireguard (and does more just link/interface monitoring).

Also removes the separate interface we had for it in sockstats -- it's
a small enough package (we already pull in all of its dependencies
via other paths) that it's not worth the extra complexity.

Updates #7621
Updates #7850

Signed-off-by: Mihai Parparita <mihai@tailscale.com>
2 years ago
Andrew Dunham f85dc6f97c
ci: add more lints (#7909)
This is a follow-up to #7905 that adds two more linters and fixes the corresponding findings. As per the previous PR, this only flags things that are "obviously" wrong, and fixes the issues found.

Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I8739bdb7bc4f75666a7385a7a26d56ec13741b7c
2 years ago
Maisem Ali 85de580455 net/tsdial: do not use proxies when dialing out to PeerAPI
Found this when adding a test that does a ping over PeerAPI.

Our integration tests set up a trafficTrap to ensure that tailscaled
does not call out to the internet, and it does so via a HTTP_PROXY.

When adding a test for pings over PeerAPI, it triggered the trap and investigation
lead to the realization that we were not removing the Proxy when trying to
dial out to the PeerAPI.

Updates tailscale/corp#8020

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Mihai Parparita d0906cda97 net/sockstats: expose debug info
Exposes some internal state of the sockstats package via the C2N and
PeerAPI endpoints, so that it can be used for debugging. For now this
includes the estimated radio on percentage and a second-by-second view
of the times the radio was active.

Also fixes another off-by-one error in the radio on percentage that
was leading to >100% values (if n seconds have passed since we started
to monitor, there may be n + 1 possible seconds where the radio could
have been on).

Updates tailscale/corp#9230

Signed-off-by: Mihai Parparita <mihai@tailscale.com>
2 years ago
Will Norris 7c386ca6d2 net/sockstats: fix calculation of radio power usage
When splitting the radio monitor usage array, we were splitting at now %
3600 to get values into chronological order.  This caused the value for
the final second to be included at the beginning of the ordered slice
rather than the end.  If there was activity during that final second, an
extra five seconds of high power usage would get recorded in some cases.
This could result in a final calculation of greater than 100% usage.

This corrects that by splitting values at (now+1 % 3600).

This also simplifies the percentage calculation by always rounding
values down, which is sufficient for our usage.

Signed-off-by: Will Norris <will@tailscale.com>
2 years ago