5883 Commits (7aba0b0d78bd698a7a4a78d3c782f921f37d4cda)

 

All Branches   

Search

Author	SHA1	Message	Date
Brad Fitzpatrick	162488a775	net/interfaces: cache "home" router lookup on big Linux routers ... This is a continuation of the earlier 2a67beaacf but more aggressive; this now remembers that we failed to find the "home" router IP so we don't try again later on the next call. Updates #7621 Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	2 years ago
James Tucker	c5150eae67	net/netcheck: reenable TestNodeAddrResolve on Windows ... Updates #7876 Co-authored-by: Andrew Dunham <andrew@du.nham.ca> Signed-off-by: Andrew Dunham <andrew@du.nham.ca> Signed-off-by: James Tucker <james@tailscale.com> Change-Id: Idb2e6cc2edf6ca123b751d6c8f8729b0cba86023	2 years ago
Andrew Dunham	80b138f0df	wgengine/magicsock: keep advertising endpoints after we stop discovering them ... Previously, when updating endpoints we would immediately stop advertising any endpoint that wasn't discovered during determineEndpoints. This could result in, for example, a case where we performed an incremental netcheck, didn't get any of our three STUN packets back, and then dropped our STUN endpoint from the set of advertised endpoints... which would result in clients falling back to a DERP connection until the next call to determineEndpoints. Instead, let's cache endpoints that we've discovered and continue reporting them to clients until a timeout expires. In the above case where we temporarily don't have a discovered STUN endpoint, we would continue reporting the old value, then re-discover the STUN endpoint again and continue reporting it as normal, so clients never see a withdrawal. Updates tailscale/coral#108 Signed-off-by: Andrew Dunham <andrew@du.nham.ca> Change-Id: I42de72e7418ab328a6c732bdefc74549708cf8b9	2 years ago
Brad Fitzpatrick	4b49ca4a12	wgengine/magicsock: update comments on what implements conn.Bind ... The comment still said *magicsock.Conn implemented wireguard-go conn.Bind. That wasn't accurate anymore. A doc #cleanup. Change-Id: I7fd003b939497889cc81147bfb937b93e4f6865c Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	2 years ago
Brad Fitzpatrick	10f1c90f4d	wgengine/magicsock, types/nettype, etc: finish ReadFromUDPAddrPort netip migration ... So we're staying within the netip.Addr/AddrPort consistently and avoiding allocs/conversions to the legacy net addr types. Updates #5162 Change-Id: I59feba60d3de39f773e68292d759766bac98c917 Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	2 years ago
Brad Fitzpatrick	29f7df9d8f	wgengine/magicsock, etc: remove mostly unused WriteTo methods ... Updates #2331 Updates #5162 Change-Id: I8291884425481eeaedde38a54adfd8ed7292a497 Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	2 years ago
Brad Fitzpatrick	83c41f3697	net/netcheck: remove unused method from interface ... Updates #2331 Updates #5162 Change-Id: I77ff956c2d59bde611d47523659a07afb4a6da2d Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	2 years ago
James Tucker	20f17d6e7b	wgengine/magicsock: reenable magicsock tests on Windows ... These tests are passing locally and on CI. They had failed earlier in the day when first fixing up CI, and it is not immediately clear why. I have cycled IPv6 support locally, but this should not have a substantial effect. Updates #7876 Signed-off-by: James Tucker <jftucker@gmail.com>	2 years ago
James Tucker	bd0c32ca21	tsnet: disable TestLoopbackSOCKS5 on Windows ... This test is not regularly passing on CI, but seems to pass reliably locally. Needs deeper debugging. Updates #7876 Signed-off-by: James Tucker <jftucker@gmail.com>	2 years ago
James Tucker	b7f51a1468	.github/workflows: add artifact caching and remove double build on race ... Go artifact caching will help provided that the cache remains small enough - we can reuse the strategy from the Windows build where we only cache and pull the zips, but let go(1) do the many-file unpacking as it does so faster. The race matrix was building once without race, then running all the tests with race, so change the matrix to incldue a `buildflags` parameter and use that both in the build and test steps. Updates #cleanup Signed-off-by: James Tucker <james@tailscale.com>	2 years ago
Andrew Dunham	f352f8a0e6	util/set: move Slice type from corp to oss ... This is an exact copy of the files misc/set/set{,_test}.go from tailscale/corp@a5415daa9c, plus the license headers. For use in #7877 Signed-off-by: Andrew Dunham <andrew@du.nham.ca> Change-Id: I712d09c6d1a180c6633abe3acf8feb59b27e2866	2 years ago
James Tucker	8dec1a8724	.github/workflows: reenable Windows CI, disable broken tests ... We accidentally switched to ./tool/go in 4022796484 which resulted in no longer running Windows builds, as this is attempting to run a bash script. I was unable to quickly fix the various tests that have regressed, so instead I've added skips referencing #7876, which we need to back and fix. Updates #7262 Updates #7876 Signed-off-by: James Tucker <james@tailscale.com>	2 years ago
Will Norris	4ecc7fdf5f	api.md: update example auth key value ... example was missing the "-auth" type in the key prefix, which all new keys now contain. Also update key ID to match the full key, and fix indenting of closing braces. Signed-off-by: Will Norris <will@tailscale.com>	2 years ago
Brad Fitzpatrick	6866aaeab3	wgengine/magicsock: factor out receiveIPv4 & receiveIPv6 common code ... Updates #2331 Change-Id: I801df38b217f5d17203e8dc3b8654f44747e0f4b Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	2 years ago
Brad Fitzpatrick	c889254b42	net/art: skip tests on CI for now ... To get the tree green again for other people. Updates #7866 Change-Id: Ibdad2e1408e5f0c97e49a148bfd77aad17c2c5e5 Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	2 years ago
Andrew Dunham	228d0c6aea	net/netcheck: use dnscache.Resolver when resolving DERP IPs ... This also adds a bunch of tests for this function to ensure that we're returning the proper IP(s) in all cases. Signed-off-by: Andrew Dunham <andrew@du.nham.ca> Change-Id: I0d9d57170dbab5f2bf07abdf78ecd17e0e635399	2 years ago
Maisem Ali	64bbf1738e	tailcfg: make SelfNodeV4MasqAddrForThisPeer a pointer ... This makes `omitempty` actually work, and saves bytes in each map response. Updates tailscale/corp#8020 Signed-off-by: Maisem Ali <maisem@tailscale.com>	2 years ago
David Anderson	a5fd51ebdc	net/art: disable the IPv6 100k routes benchmark. ... At the current unoptimized memory utilization of the various data structures, 100k IPv6 routes consumes in the ballpark of 3-4GiB, which risks OOMing our 386 test machine. Until we have the optimizations to (drastically) reduce that consumption, skip the test that bloats too much for 32-bit machines. Signed-off-by: David Anderson <danderson@tailscale.com>	2 years ago
David Anderson	a7c910e361	net/art: implement the Table type, a multi-level art route table. ... Updates #7781 │ sec/op │ TableInsertion/ipv4/10 1.562µ ± 2% TableInsertion/ipv4/100 2.398µ ± 5% TableInsertion/ipv4/1000 2.097µ ± 3% TableInsertion/ipv4/10000 2.756µ ± 4% TableInsertion/ipv4/100000 2.473µ ± 13% TableInsertion/ipv6/10 7.649µ ± 2% TableInsertion/ipv6/100 12.09µ ± 3% TableInsertion/ipv6/1000 14.84µ ± 5% TableInsertion/ipv6/10000 14.72µ ± 8% TableInsertion/ipv6/100000 13.23µ ± 41% TableDelete/ipv4/10 378.4n ± 5% TableDelete/ipv4/100 366.9n ± 3% TableDelete/ipv4/1000 418.6n ± 3% TableDelete/ipv4/10000 609.2n ± 11% TableDelete/ipv4/100000 679.2n ± 28% TableDelete/ipv6/10 504.2n ± 4% TableDelete/ipv6/100 959.5n ± 12% TableDelete/ipv6/1000 1.436µ ± 6% TableDelete/ipv6/10000 1.772µ ± 15% TableDelete/ipv6/100000 1.172µ ± 113% TableGet/ipv4/10 32.14n ± 11% TableGet/ipv4/100 38.58n ± 2% TableGet/ipv4/1000 45.03n ± 2% TableGet/ipv4/10000 52.90n ± 7% TableGet/ipv4/100000 135.2n ± 11% TableGet/ipv6/10 41.55n ± 1% TableGet/ipv6/100 44.78n ± 2% TableGet/ipv6/1000 49.03n ± 2% TableGet/ipv6/10000 65.38n ± 5% TableGet/ipv6/100000 525.0n ± 39% │ avg-B/op │ TableInsertion/ipv4/10 25.18Ki ± 0% TableInsertion/ipv4/100 17.63Ki ± 0% TableInsertion/ipv4/1000 14.14Ki ± 0% TableInsertion/ipv4/10000 12.92Ki ± 0% TableInsertion/ipv4/100000 11.13Ki ± 0% TableInsertion/ipv6/10 76.87Ki ± 0% TableInsertion/ipv6/100 98.33Ki ± 0% TableInsertion/ipv6/1000 91.44Ki ± 0% TableInsertion/ipv6/10000 90.39Ki ± 0% TableInsertion/ipv6/100000 87.19Ki ± 0% TableDelete/ipv4/10 3.230 ± 0% TableDelete/ipv4/100 4.020 ± 0% TableDelete/ipv4/1000 3.990 ± 0% TableDelete/ipv4/10000 4.000 ± 0% TableDelete/ipv4/100000 4.000 ± 0% TableDelete/ipv6/10 16.00 ± 0% TableDelete/ipv6/100 16.00 ± 0% TableDelete/ipv6/1000 16.00 ± 0% TableDelete/ipv6/10000 16.00 ± 0% TableDelete/ipv6/100000 16.00 ± 0% │ avg-allocs/op │ TableInsertion/ipv4/10 2.900 ± 0% TableInsertion/ipv4/100 2.330 ± 0% TableInsertion/ipv4/1000 2.070 ± 0% TableInsertion/ipv4/10000 1.980 ± 0% TableInsertion/ipv4/100000 1.840 ± 0% TableInsertion/ipv6/10 6.800 ± 0% TableInsertion/ipv6/100 8.420 ± 0% TableInsertion/ipv6/1000 7.900 ± 0% TableInsertion/ipv6/10000 7.820 ± 0% TableInsertion/ipv6/100000 7.580 ± 0% TableDelete/ipv4/10 1.000 ± 0% TableDelete/ipv4/100 1.000 ± 0% TableDelete/ipv4/1000 1.000 ± 0% TableDelete/ipv4/10000 1.000 ± 0% TableDelete/ipv4/100000 1.000 ± 0% TableDelete/ipv6/10 1.000 ± 0% TableDelete/ipv6/100 1.000 ± 0% TableDelete/ipv6/1000 1.000 ± 0% TableDelete/ipv6/10000 1.000 ± 0% TableDelete/ipv6/100000 1.000 ± 0% │ routes/s │ TableInsertion/ipv4/10 640.3k ± 2% TableInsertion/ipv4/100 417.1k ± 5% TableInsertion/ipv4/1000 477.0k ± 3% TableInsertion/ipv4/10000 362.8k ± 5% TableInsertion/ipv4/100000 404.5k ± 15% TableInsertion/ipv6/10 130.7k ± 1% TableInsertion/ipv6/100 82.69k ± 3% TableInsertion/ipv6/1000 67.37k ± 5% TableInsertion/ipv6/10000 67.93k ± 9% TableInsertion/ipv6/100000 75.63k ± 29% TableDelete/ipv4/10 2.642M ± 6% TableDelete/ipv4/100 2.726M ± 3% TableDelete/ipv4/1000 2.389M ± 3% TableDelete/ipv4/10000 1.641M ± 12% TableDelete/ipv4/100000 1.472M ± 27% TableDelete/ipv6/10 1.984M ± 4% TableDelete/ipv6/100 1.042M ± 11% TableDelete/ipv6/1000 696.5k ± 6% TableDelete/ipv6/10000 564.4k ± 13% TableDelete/ipv6/100000 853.6k ± 53% │ addrs/s │ TableGet/ipv4/10 31.11M ± 10% TableGet/ipv4/100 25.92M ± 2% TableGet/ipv4/1000 22.21M ± 2% TableGet/ipv4/10000 18.91M ± 8% TableGet/ipv4/100000 7.397M ± 12% TableGet/ipv6/10 24.07M ± 1% TableGet/ipv6/100 22.33M ± 2% TableGet/ipv6/1000 20.40M ± 2% TableGet/ipv6/10000 15.30M ± 5% TableGet/ipv6/100000 1.905M ± 28% │ B/op │ TableGet/ipv4/10 4.000 ± 0% TableGet/ipv4/100 4.000 ± 0% TableGet/ipv4/1000 4.000 ± 0% TableGet/ipv4/10000 4.000 ± 0% TableGet/ipv4/100000 4.000 ± 0% TableGet/ipv6/10 16.00 ± 0% TableGet/ipv6/100 16.00 ± 0% TableGet/ipv6/1000 16.00 ± 0% TableGet/ipv6/10000 16.00 ± 0% TableGet/ipv6/100000 16.00 ± 0% │ allocs/op │ TableGet/ipv4/10 1.000 ± 0% TableGet/ipv4/100 1.000 ± 0% TableGet/ipv4/1000 1.000 ± 0% TableGet/ipv4/10000 1.000 ± 0% TableGet/ipv4/100000 1.000 ± 0% TableGet/ipv6/10 1.000 ± 0% TableGet/ipv6/100 1.000 ± 0% TableGet/ipv6/1000 1.000 ± 0% TableGet/ipv6/10000 1.000 ± 0% TableGet/ipv6/100000 1.000 ± 0% Signed-off-by: David Anderson <danderson@tailscale.com>	2 years ago
Mihai Parparita	edb02b63f8	net/sockstats: pass in logger to sockstats.WithSockStats ... Using log.Printf may end up being printed out to the console, which is not desirable. I noticed this when I was investigating some client logs with `sockstats: trace "NetcheckClient" was overwritten by another`. That turns to be harmless/expected (the netcheck client will fall back to the DERP client in some cases, which does its own sockstats trace). However, the log output could be visible to users if running the `tailscale netcheck` CLI command, which would be needlessly confusing. Updates tailscale/corp#9230 Signed-off-by: Mihai Parparita <mihai@tailscale.com>	2 years ago
Mihai Parparita	782ccb5655	.github/workflows: run one set of tests with the tailscale_go build tag ... We use it to gate code that depends on custom Go toolchain, but it's currently only passed in the corp runners. Add a set on OSS so that we can catch regressions earlier. To specifically test sockstats this required adding a build tag to explicitly enable them -- they're normally on for iOS, macOS and Android only, and we don't run tests on those platforms normally. Signed-off-by: Mihai Parparita <mihai@tailscale.com>	2 years ago
Will Norris	bb34589748	log/sockstatlog: limit sockstat logs to 5 MB ... Updates tailscale/corp#9230 Signed-off-by: Will Norris <will@tailscale.com>	2 years ago
Brad Fitzpatrick	9e50da321b	client/tailscale: dial LocalAPI at 127.0.0.1 on macOS ... Updates #7851 Change-Id: Ib53cf53cdfee277ef42f7833352bc51ecb5db959 Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>	2 years ago
Mihai Parparita	bdc7a61c24	tool/gocross: add ts_macext build tag for Xcode builds ... It's used to control various opt-in functionality for the macOS and iOS apps, and was lost in the migration to gocross. Updates tailscale/tailscale#7769 Signed-off-by: Mihai Parparita <mihai@tailscale.com>	2 years ago
License Updater	33b006cacf	licenses: update tailscale{,d} licenses ... Signed-off-by: License Updater <noreply@tailscale.com>	2 years ago
License Updater	e5d272f445	licenses: update win/apple licenses ... Signed-off-by: License Updater <noreply+license-updater@tailscale.com>	2 years ago
Flakes Updater	7c95734907	go.mod.sri: update SRI hash for go.mod changes ... Signed-off-by: Flakes Updater <noreply@tailscale.com>	2 years ago
Anton Tolchanov	8546ff98fb	tsweb: move varz handler(s) into separate modules ... This splits Prometheus metric handlers exposed by tsweb into two modules: - `varz.Handler` exposes Prometheus metrics generated by our expvar converter; - `promvarz.Handler` combines our expvar-converted metrics and native Prometheus metrics. By default, tsweb will use the promvarz handler, however users can keep using only the expvar converter. Specifically, `tailscaled` now uses `varz.Handler` explicitly, which avoids a dependency on the (heavyweight) Prometheus client. Updates https://github.com/tailscale/corp/issues/10205 Signed-off-by: Anton Tolchanov <anton@tailscale.com>	2 years ago
Anton Tolchanov	c153e6ae2f	prober: migrate to Prometheus metric library ... This provides an example of using native Prometheus metrics with tsweb. Prober library seems to be the only user of PrometheusVar, so I am removing support for it in tsweb. Updates https://github.com/tailscale/corp/issues/10205 Signed-off-by: Anton Tolchanov <anton@tailscale.com>	2 years ago
Anton Tolchanov	11e6247d2a	tsweb: expose native Prometheus metrics in /debug/varz ... The handler will expose built-in process and Go metrics by default, which currently duplicate some of the expvar-proxied metrics (`goroutines` vs `go_goroutines`, `memstats` vs `go_memstats`), but as long as their names are different, Prometheus server will just scrape both. This will change /debug/varz behaviour for most tsweb binaries, but notably not for control, which configures a `tsweb.VarzHandler` [explicitly](a5b5d5167f/cmd/tailcontrol/tailcontrol.go (L779)) Updates https://github.com/tailscale/corp/issues/10205 Signed-off-by: Anton Tolchanov <anton@tailscale.com>	2 years ago
License Updater	690446c784	licenses: update android licenses ... Signed-off-by: License Updater <noreply@tailscale.com>	2 years ago
Mihai Parparita	cef0a474f8	ipn/ipnlocal: check that sockstatLogger is available in c2n endpoint ... Otherwise there may be a panic if it's nil (and the control side of the c2n call will just time out). Signed-off-by: Mihai Parparita <mihai@tailscale.com>	2 years ago
Mihai Parparita	03b2c44a21	ipn/ipnlocal: more explicitly say if sockstats are not available ... Makes it more apparent in the PeerAPI endpoint that the client was not built with the appropriate toolchain or build tags. Updates tailscale/corp#9230 Signed-off-by: Mihai Parparita <mihai@tailscale.com>	2 years ago
Flakes Updater	1bec2cbbd5	go.mod.sri: update SRI hash for go.mod changes ... Signed-off-by: Flakes Updater <noreply@tailscale.com>	2 years ago
Jordan Whited	f571536598	go.mod: bump wireguard-go (#7836) ... This pulls in a synchronization optimization, see tailscale/wireguard-go@af17262. Updates tailscale/corp#8734 Signed-off-by: Jordan Whited <jordan@tailscale.com>	2 years ago
James Tucker	e09c434e5d	wgengine/magicsock: remove locking sync requirements on conn disco keys ... The lazy initialization of the disco key is not necessary, and contributes to unnecessary locking and state checking. Updates #cleanup Signed-off-by: James Tucker <james@tailscale.com>	2 years ago
James Tucker	e1b71c83ac	wgengine/magicsock: remove unused fields on discoInfo ... Updates #cleanup Signed-off-by: James Tucker <james@tailscale.com>	2 years ago
James Tucker	a257b2f88b	wgengine/magicsock: add immutability documentation to endpointDisco ... Updates #7825 Signed-off-by: James Tucker <james@tailscale.com>	2 years ago
James Tucker	fb18af5564	wgengine/netstack: fix data-race on startup ... Running tailscaled with the race detector enabled immediately fires on this field, as it is updated after first read. Updates #cleanup Signed-off-by: James Tucker <james@tailscale.com>	2 years ago
Charlotte Brandhorst-Satzkorn	c573bef0aa	tailcfg,wgengine: add initial support for WireGuard only peers ... A peer can have IsWireGuardOnly, which means it will not support DERP or Disco, and it must have Endpoints filled in order to be usable. In the present implementation only the first Endpoint will be used as the bestAddr. Updates tailscale/corp#10351 Co-authored-by: Charlotte Brandhorst-Satzkorn <charlotte@tailscale.com> Co-authored-by: James Tucker <james@tailscale.com> Signed-off-by: James Tucker <james@tailscale.com>	2 years ago
James Tucker	6cfcb3cae4	wgengine/magicsock: fix synchronization of endpoint disco fields ... Identified in review in #7821 endpoint.discoKey and endpoint.discoShort are often accessed without first taking endpoint.mu. The arrangement with endpoint.mu is inconvenient for a good number of those call-sites, so it is instead replaced with an atomic pointer to carry both pieces of disco info. This will also help with #7821 that wants to add explicit checks/guards to disable disco behaviors when disco keys are missing which is necessarily implicitly mostly covered by this change. Updates #7821 Signed-off-by: James Tucker <james@tailscale.com>	2 years ago
Mihai Parparita	e978299bf0	net/sockstats: disable deltas for the cell radio power state metric ... Updates tailscale/corp#9230 Signed-off-by: Mihai Parparita <mihai@tailscale.com>	2 years ago
Will Norris	22680a11ae	net/sockstats: return early if no radio period length ... Signed-off-by: Will Norris <will@tailscale.com>	2 years ago
Will Norris	75784e10e2	sockstats: add client metrics for radio power state ... power state is very roughly approximated based on observed network activity and AT&T's state transition timings for a typical 3G radio. Updates tailscale/corp#9230 Updates #3363 Signed-off-by: Will Norris <will@tailscale.com>	2 years ago
Tom DNetto	6a627e5a33	net, wgengine/capture: encode NAT addresses in pcap stream ... Signed-off-by: Tom DNetto <tom@tailscale.com>	2 years ago
License Updater	92459a9248	licenses: update win/apple licenses ... Signed-off-by: License Updater <noreply+license-updater@tailscale.com>	2 years ago
License Updater	7012bf7981	licenses: update tailscale{,d} licenses ... Signed-off-by: License Updater <noreply@tailscale.com>	2 years ago
License Updater	07b29f13dc	licenses: update android licenses ... Signed-off-by: License Updater <noreply@tailscale.com>	2 years ago
Mihai Parparita	f49b9f75b8	util/clientmetric: allow client metric values to be provided by a function ... Adds NewGaugeFunc and NewCounterFunc (inspired by expvar.Func) which change the current value to be reported by a function. This allows some client metric values to be computed on-demand during uploading (at most every 15 seconds), instead of being continuously updated. clientmetric uploading had a bunch of micro-optimizations for memory access (#3331) which are not possible with this approach. However, any performance hit from function-based metrics is contained to those metrics only, and we expect to have very few. Also adds a DisableDeltas() option for client metrics, so that absolute values are always reported. This makes server-side processing of some metrics easier to reason about. Updates tailscale/corp#9230 Signed-off-by: Mihai Parparita <mihai@tailscale.com>	2 years ago
phirework	c0e0a5458f	cmd/tailscale: show reauth etc. links even if no login name (#7803) ... Signed-off-by: Jenny Zhang <jz@tailscale.com>	2 years ago