Commit Graph

2866 Commits (1f006025c219eb726f78890bcd9239e6646c9982)
 

Author SHA1 Message Date
Maisem Ali 1f006025c2 net/tstun: fix build on arm
Signed-off-by: Maisem Ali <maisem@tailscale.com>
3 years ago
Josh Bleecher Snyder fcca374fa7 tstest/integration/testcontrol: sort peers in map response
This is part of the control protocol.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Brad Fitzpatrick cd426eaf4c net/portmapper: fix t.Log-after-test-done race in tests
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Maisem Ali 9f62cc665e tailscaled: try migrating old state on synology devices
Signed-off-by: Maisem Ali <maisem@tailscale.com>
3 years ago
Maisem Ali 5c383bdf5d wgengine/router: pass in AmbientCaps when calling `ip rule`
Signed-off-by: Maisem Ali <maisem@tailscale.com>
3 years ago
Brad Fitzpatrick 56db3e2548 ipn/localapi: refresh ACME certs in background two weeks in advance
Updates #1235

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick 6f8c8c771b control/controlclient: tweak a couple error messages
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick b7ae529ecc client/tailscale: make GetCertificate guess cert if SNI lacks dots
Updates #1235

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Simeng He e199e407d2 tailcfg: add IP and Types field to PingRequest
Signed-off-by: Simeng He <simeng@tailscale.com>
3 years ago
Brad Fitzpatrick d5e1abd0c4 cmd/tailscale/cli: only write cert file if it changed
Updates #1235

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick 57b794c338 ipn/localapi: move cert fetching code to localapi, cache, add cert subcommand
Updates #1235

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Josh Bleecher Snyder 4c8b5fdec4 control/controlclient: do not periodically print full netmap
The netmaps can get really large.
Printing, processing, and uploading them is expensive.
Only print the header on an ongoing basis.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Josh Bleecher Snyder a666b546fb ipn/ipnlocal: log number of packet filters rather than entire filter
The number of packet filters can grow very large,
so this log entry can be very large.
We can get the packet filter server-side,
so reduce verbosity here to just the number of filters present.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Josh Bleecher Snyder 0c038b477f logtail: add a re-usable buffer for uploads
This avoids a per-upload alloc (which in practice
often means per-log-line), up to 4k.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Josh Bleecher Snyder 278e7de9c9 logtail: always send a json array
The code goes to some effort to send a single JSON object
when there's only a single line and a JSON array when there
are multiple lines.

It makes the code more complex and more expensive;
when we add a second line, we have to use a second buffer
to duplicate the first one after adding a leading square brackets.

The savings come to two bytes. Instead, always send an array.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Josh Bleecher Snyder 93284209bc logtail/filch: preallocate a scanner buffer
Scanning log lines is a frequent source of allocations.
Pre-allocate a re-usable buffer.

This still doesn't help when there are giant log lines.
Those will still be problematic from an iOS memory perspective.
For more on that, see https://github.com/tailscale/corp/issues/2423.

(For those who cannot follow that link, it is a discussion
of particular problematic types of log lines for
particular categories of customers. The "categories of customers"
part is the reason that it is a private issue.)

There is also a latent bug here. If we ever encounter
a log line longer than bufio.MaxScanTokenSize,
then bufio.Scan will return an error,
and we'll truncate the file and discard the rest of the log.
That's not good, but bufio.MaxScanTokenSize is really big,
so it probably doesn't matter much in practice now.
Unfortunately, it does prevent us from easily capping the potential
memory usage here, on pain of losing log entries.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Matt Layher 8ab44b339e net/tstun: use unix.Ifreq type for Linux TAP interface configuration
Signed-off-by: Matt Layher <mdlayher@gmail.com>
3 years ago
Josh Bleecher Snyder 6da6d47a83 all: simplify build tags involving iOS
Prior to Go 1.16, iOS used GOOS=darwin,
so we had to distinguish macOS from iOS during GOARCH.

We now require Go 1.16 in our go.mod, so we can simplify.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Josh Bleecher Snyder a24cee0d67 all: simplify ts_macext build tags
Now that we have the easier-to-parse go:build build tags,
it is straightforward to simplify them. Yay.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Josh Bleecher Snyder d2aa144dcc syncs: bump known good version to include Go 1.17
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Brad Fitzpatrick 25e060a841 cmd/tailscale/cli: fix cert fetch WaitOrder retry loop, misc cleanups
Updates #1235

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick 833200da6f net/tstun: don't exec uname -r on Linux in TUN failure diagnostics
Fixes https://twitter.com/zekjur/status/1425557520513486848

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick e804ab29fd net/tstun: move TUN failure diagnostics to OS-specific files
Mostly so the Linux one can use Linux-specific stuff in package
syscall and not use os/exec for uname for portability.

But also it helps deps a tiny bit on iOS.

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick b2eea1ee00 cmd/tailscale/cli: make cert fetch registration automatic, show valid domains
Updates #1235

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick 39610aeb09 wgengine/magicsock: move debug knobs to their own file, compile out on iOS
No need for these knobs on iOS where you can set the environment
variables anyway.

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Denton Gentry 98d557dd24 Dockerfile: use alpine:3.14
golang:1.16-alpine has updated to Alpine 3.14,
update the system image to match.

Reported by @kubeworm
https://twitter.com/kubeworm/status/1426751941519020033

Signed-off-by: Denton Gentry <dgentry@tailscale.com>
3 years ago
Denton Gentry 3e7ff5ff98 cmd/tailscaled: enable hybrid netstack mode for FreeBSD.
Allows FreeBSD to function as an exit node in the same way
that Windows and Tailscaled-on-MacOS do.

RELNOTE=FreeBSD can now function as an exit node.

Fixes https://github.com/tailscale/tailscale/issues/2498

Signed-off-by: Denton Gentry <dgentry@tailscale.com>
3 years ago
Brad Fitzpatrick 954867fef5 words: fix parser to handle missing newline at end
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick c992504375 words: group some scales, support comments
Not sure how we missed mixolydian.

Updates #1235

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick 1bca722824 words: add five types of tales, best dog
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick 00b4c2331b words: add accessors and tests for a few of our favorite words
Updates #1235

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Charlotte Brandhorst-Satzkorn 9547669787 words: these are a few more of my favorite words
Updates #1235

Signed-off-by: Charlotte Brandhorst-Satzkorn <charlotte@catzkorn.dev>
3 years ago
Charlotte Brandhorst-Satzkorn b5a41ff381 words: these are a few of my favorite words
Updates #1235

Signed-off-by: Charlotte Brandhorst-Satzkorn <charlotte@catzkorn.dev>
3 years ago
Brad Fitzpatrick ec9f3f4cc0 cmd/tailscale: update depaware
Missing from prior commit.

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick c68a12afe9 cmd/tailscale: add temporary debug command for getting DNS-01 LetsEncrypt cert
Not even close to usable or well integrated yet, but submitting this before
it bitrots or I lose it.

Updates #1235

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
David Anderson d2d55bd63c cmd/microproxy: delete.
Thank goodness.

Fixes #2635

Signed-off-by: David Anderson <danderson@tailscale.com>
3 years ago
Brad Fitzpatrick c6740da624 tsweb: make VarzHandler support untyped expvar.Maps for compatibility
Updates #2635

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick 7c7eb8094b tsweb: make VarzHandler support expvar.Funcs returning ints/floats
Updates #2635
3 years ago
Brad Fitzpatrick 5aba620fb9 tsweb: make VarzHandler capable of walking structs with reflect
To be used by control, per linked bug's plan.

Updates #2635

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
julianknodt b9bd7dbc5d net/portmapper: log upnp information
This logs some basic statistics for UPnP, so that tailscale can better understand what routers
are being used and how to connect to them.

Signed-off-by: julianknodt <julianknodt@gmail.com>
3 years ago
julianknodt 26b6fe7f02 net/portmapper: add PCP integration test
This adds a PCP test to the IGD test server, by hardcoding in a few observed packets from
Denton's box.

Signed-off-by: julianknodt <julianknodt@gmail.com>
3 years ago
Brad Fitzpatrick 3700cf9ea4 tsweb: also support LabelMaps from expvar.Map, without metrics
We want to use tsweb to format Prometheus-style metrics from
our temporary golang.org/x/net/http2 fork, but we don't want http2
to depend on the tailscale.com module to use the concrete type
tailscale.com/metrics.LabelMap. Instead, let a expvar.Map be used
instead of it's annotated sufficiently in its name.

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Brad Fitzpatrick 5f45d8f8e6 tsweb: add VarzHandler tests
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Josh Bleecher Snyder a4e19f2233 version: remove rsc.io/goversion dependency
rsc.io/goversion is really expensive.
Running version.ReadExe on tailscaled on darwin
allocates 47k objects, almost 11mb.

All we want is the module info. For that, all we need to do
is scan through the binary looking for the magic start/end strings
and then grab the bytes in between them.

We can do that easily and quickly with nothing but a 64k buffer.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
3 years ago
Brad Fitzpatrick bdb93c5942 net/portmapper: actually test something in TestProbeIntegration
And use dynamic port numbers in tests, as Linux on GitHub Actions and
Windows in general have things running on these ports.

Co-Author: Julian Knodt <julianknodt@gmail.com>
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
3 years ago
Denton Gentry 26c1183941 hostinfo: add fly.io detection
Signed-off-by: Denton Gentry <dgentry@tailscale.com>
3 years ago
Denton Gentry 0796c53404 tsnet: add AuthKey support.
Set a TS_AUTHKEY environment variable to "tskey-01234..."

Signed-off-by: Denton Gentry <dgentry@tailscale.com>
3 years ago
Adrian Dewhurst 8bdf878832 net/dns/resolver: use forwarded dns txid directly
Previously, we hashed the question and combined it with the original
txid which was useful when concurrent queries were multiplexed on a
single local source port. We encountered some situations where the DNS
server canonicalizes the question in the response (uppercase converted
to lowercase in this case), which resulted in responses that we couldn't
match to the original request due to hash mismatches. This includes a
new test to cover that situation.

Fixes #2597

Signed-off-by: Adrian Dewhurst <adrian@tailscale.com>
3 years ago
David Crawshaw 360223fccb types/dnstype: introduce new package for Resolver
So the type can be used in net/dns without introducing a tailcfg
dependency.

For #2596

Signed-off-by: David Crawshaw <crawshaw@tailscale.com>
3 years ago
Christine Dodrill 4d19db7c9f
scripts/installer: work on Oracle Linux (#2604)
Before we didn't detect it properly. Since Oracle Linux is diet centos,
we can just make the centos logic detect Oracle linux and everything
should be fine.

Signed-off-by: Christine Dodrill <xe@tailscale.com>
3 years ago