Commit Graph

4623 Commits (4eed2883db7875e3f2efc66da6c6924da103458a)
 

Author SHA1 Message Date
Charlotte Brandhorst-Satzkorn 4eed2883db
words: space, the final frontier (#5952)
Captains log. Stardate 100386.37.

Work is proceeding on the Words list as Tailscalars are forced to scavenge for more taily and scaley things.

Signed-off-by: Charlotte Brandhorst-Satzkorn <charlotte@tailscale.com>
2 years ago
Andrew Dunham c32f9f5865
cmd/tailscale, ipn: enable debug logs when --report flag is passed to bugreport (#5830)
Change-Id: Id22e9f4a2dcf35cecb9cd19dd844389e38c922ec
Signed-off-by: Andrew Dunham <andrew@tailscale.com>
2 years ago
Andrew Dunham 64ea60aaa3
derp: add TCP RTT metric on Linux (#5949)
Periodically poll the TCP RTT metric from all open TCP connections and
update a (bucketed) histogram metric.

Signed-off-by: Andrew Dunham <andrew@du.nham.ca>
Change-Id: I6214902196b05bf7829c9d0ea501ce0e13d984cf
2 years ago
Brad Fitzpatrick a04f1ff9e6 logtail: default to 2s log flush delay on all platforms
Per chat. This is close enough to realtime but massively reduces
number of HTTP requests. (which you can verify with
TS_DEBUG_LOGTAIL_WAKES and watching tailscaled run at start)

By contrast, this is set to 2 minutes on mobile.

Change-Id: Id737c7924d452de5c446df3961f5e94a43a33f1f
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Mihai Parparita 63ad49890f cmd/tsconnect: pre-compress main.wasm when building the NPM package
This way we can do that once (out of band, in the GitHub action),
instead of increasing the time of each deploy that uses the package.

.wasm is removed from the list of automatically pre-compressed
extensions, an OSS bump and small change on the corp side is needed to
make use of this change.

Signed-off-by: Mihai Parparita <mihai@tailscale.com>
2 years ago
License Updater 899b4cae10 licenses: update win/apple licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Tom DNetto a515fc517b ipn/ipnlocal: make tkaSyncIfNeeded exclusive with a mutex
Running corp/ipn#TestNetworkLockE2E has a 1/300 chance of failing, and
deskchecking suggests thats whats happening are two netmaps are racing each
other to be processed through tkaSyncIfNeededLocked. This happens in the
first place because we release b.mu during network RPCs.

To fix this, we make the tka sync logic an exclusive section, so two
netmaps will need to wait for tka sync to complete serially (which is what
we would want anyway, as the second run through probably wont need to
sync).

Signed-off-by: Tom DNetto <tom@tailscale.com>
2 years ago
Tom DNetto 227777154a control/controlclient,ipn/ipnlocal,tailcfg: rotate node-key signature on register
CAPVER 47

Signed-off-by: Tom DNetto <tom@tailscale.com>
2 years ago
Anton Tolchanov 26af329fde prober: expand certificate verification logic in the TLS prober
TLS prober now checks validity period for all server certificates
and verifies OCSP revocation status for the leaf cert.

Signed-off-by: Anton Tolchanov <anton@tailscale.com>
2 years ago
License Updater 39d03b6b63 licenses: update win/apple licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Maisem Ali 3555a49518 net/dns: always attempt to read the OS config on macOS/iOS
Also reconfigure DNS on iOS/macOS on link changes.

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
James Tucker 539c073cf0 wgengine/magicsock: set UDP socket buffer sizes to 7MB
- At high data rates more buffer space is required in order to avoid
  packet loss during any cause of delay.
- On slower machines more buffer space is required in order to avoid
  packet loss while decryption & tun writing is underway.
- On higher latency network paths more buffer space is required in order
  to overcome BDP.
- On Linux set with SO_*BUFFORCE to bypass net.core.{r,w}mem_max.
- 7MB is the current default maximum on macOS 12.6
- Windows test is omitted, as Windows does not support getsockopt for
  these options.

Signed-off-by: James Tucker <james@tailscale.com>
2 years ago
Brad Fitzpatrick a315336287 logtail: change batched upload mechanism to not use CPU when idle
The mobile implementation had a 2 minute ticker going all the time
to do a channel send. Instead, schedule it as needed based on activity.

Then we can be actually idle for long periods of time.

Updates #3363

Change-Id: I0dba4150ea7b94f74382fbd10db54a82f7ef6c29
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Will Norris d05dd41bc1 api.md: document using '-' value as default tailnet
Signed-off-by: Will Norris <will@tailscale.com>
2 years ago
Brad Fitzpatrick 9a264dac01 net/netcheck: fix crash in checkCaptivePortal
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>
2 years ago
Mihai Parparita b2855cfd86 derp/derphttp: fix nil pointer dereference when closing a netcheck client
NewNetcheckClient only initializes a subset of fields of derphttp.Client,
and the Close() call added by #5707 was result in a nil pointer dereference.
Make Close() safe to call when using NewNetcheckClient() too.

Fixes #5919

Signed-off-by: Mihai Parparita <mihai@tailscale.com>
2 years ago
James Tucker 4ec6d41682 wgengine/router: fix MTU configuration on Windows
Always set the MTU to the Tailscale default MTU. In practice we are
missing applying an MTU for IPv6 on Windows prior to this patch.

This is the simplest patch to fix the problem, the code in here needs
some more refactoring.

Fixes #5914

Signed-off-by: James Tucker <james@tailscale.com>
2 years ago
Joe Tsai a1a43ed266
wgengine/netlog: add support for magicsock statistics (#5913)
This sets up Logger to handle statistics at the magicsock layer,
where we can correlate traffic between a particular tailscale IP address
and any number of physical endpoints used to contact the node
that hosts that tailscale address.

We also export Message and TupleCounts to better document the JSON format
that is being sent to the logging infrastructure.

This commit does NOT yet enable the actual logging of magicsock statistics.
That will be a future commit.

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
2 years ago
License Updater db863bf00f licenses: update android licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Joe Tsai f9120eee57
wgengine: start network logger in Userspace.Reconfig (#5908)
If the wgcfg.Config is specified with network logging arguments,
then Userspace.Reconfig starts up an asynchronous network logger,
which is shutdown either upon Userspace.Close or when Userspace.Reconfig
is called again without network logging or route arguments.

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
2 years ago
Joe Tsai 49bae7fd5c
wgengine: fix typo in Engine.PeerForIP (#5912)
Signed-off-by: Joe Tsai <joetsai@digital-static.net>
2 years ago
Sonia Appasamy 5363a90272 types/view: add ContainsNonExitSubnetRoutes func
Signed-off-by: Sonia Appasamy <sonia@tailscale.com>
2 years ago
Mihai Parparita b49eb7d55c cmd/tsconnect: move NPM package to being under the @tailscale org
Signed-off-by: Mihai Parparita <mihai@tailscale.com>
2 years ago
Joe Tsai 1b4e4cc1e8
wgengine/netlog: new package for traffic flow logging (#5864)
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>
2 years ago
Brad Fitzpatrick 79755d3ce5 tstest/natlab: add Firewall.Reset method to drop firewall state
For future use in magicsock tests.

Updates #540

Change-Id: I2f07d1a2924f20b36e357c4533ff0a1a974d5061
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Denton Gentry 1b9ed9f365 VERSION.txt: this is 1.33.
We did not get this VERSION.txt file checked in at the correct time,
the prior 10 commits in `main` between the v1.32.0 tag point and
this commit were not part of release 1.32. We did no unstable builds
during this time, so the error should have no impact.

Signed-off-by: Denton Gentry <dgentry@tailscale.com>
2 years ago
License Updater e7519adc18 licenses: update tailscale{,d} licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Brad Fitzpatrick e24de8a617 ssh/tailssh: add password-forcing workaround for buggy SSH clients
If the username includes a suffix of +password, then we accept
password auth and just let them in like it were no auth.

This exists purely for SSH clients that get confused by seeing success
to their initial auth type "none".

Co-authored-by: Maisem Ali <maisem@tailscale.com>
Change-Id: I616d4c64d042449fb164f615012f3bae246e91ec
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Anton Tolchanov c070d39287 cmd/tailscaled: handle tailscaled symlink on macOS
When Tailscale is installed via Homebrew, `/usr/local/bin/tailscaled`
is a symlink to the actual binary.

Now when `tailscaled install-system-daemon` runs, it will not attempt
to overwrite that symlink if it already points to the tailscaled binary.
However, if executed binary and the link target differ, the path will
he overwritten - this can happen when a user decides to replace
Homebrew-installed tailscaled with a one compiled from source code.

Fixes #5353

Signed-off-by: Anton Tolchanov <anton@tailscale.com>
2 years ago
Denton Gentry 51d488673a scripts/installer.sh: add OSMC
Fixes https://github.com/tailscale/tailscale/issues/4960

Signed-off-by: Denton Gentry <dgentry@tailscale.com>
2 years ago
Emmanuel T Odeke 680f8d9793 all: fix more resource leaks found by staticmajor
Updates #5706

Signed-off-by: Emmanuel T Odeke <emmanuel@orijtech.com>
2 years ago
Brad Fitzpatrick 614a24763b tsweb: sort top-level expvars after removing type prefixes
Fixes #5778

Change-Id: I56c367338fa5686da288cc6545209ef4d6b88549
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick 0475ed4a7e cmd/ssh-auth-none-demo: put the hostname in the package doc
188.166.70.128 port 2222 for now. Some hostname later maybe.

Change-Id: I9c329410035221ed6cdff7a482727d30b77eea8b
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Maisem Ali 7df85c6031 cmd/ssh-auth-none-demo: add banner as part of the demo
Send two banners with a second in between, this demonstrates the case
where all banners are shown after auth completes and not during.

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Brad Fitzpatrick 718914b697 tsweb: remove allocs introduced by earlier change
This removes the ~9 allocs added by #5869, while still keeping struct
fields sorted (the previous commit's tests still pass). And add a test
to lock it in that this shouldn't allocate.

Updates #5778

Change-Id: I4c12b9e2a1334adc1ea5aba1777681cb9fc18fbf
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
License Updater 529e893f70 licenses: update win/apple licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Brad Fitzpatrick 8a187159b2 cmd/ssh-auth-none-demo: add demo SSH server that acts like Tailscale SSH
For SSH client authors to fix their clients without setting up
Tailscale stuff.

Change-Id: I8c7049398512de6cb91c13716d4dcebed4d47b9c
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Brad Fitzpatrick b2994568fe ipn/localapi: put all the LocalAPI methods into a map
Rather than a bunch of switch cases.

Change-Id: Id1db813ec255bfab59cbc982bee351eb36373245
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Maisem Ali f172fc42f7 ssh/tailssh: close sshContext on context cancellation
This was preventing tailscaled from shutting down properly if there were
active sessions in certain states (e.g. waiting in check mode).

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Hasnain Lakhani 8fe04b035c tsweb: sort varz by name after stripping prefix (#5778)
This makes it easier to view prometheus metrics.

Added a test case which demonstrates the new behavior - the test
initially failed as the output was ordered in the same order
as the fields were declared in the struct (i.e. foo_a, bar_a, foo_b,
bar_b). For that reason, I also had to change an existing test case
to sort the fields in the new expected order.

Signed-off-by: Hasnain Lakhani <m.hasnain.lakhani@gmail.com>
2 years ago
License Updater d29ec4d7a4 licenses: update tailscale{,d} licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Maisem Ali 4de1601ef4 ssh/tailssh: add support for sending multiple banners
Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
License Updater 91b5c50b43 licenses: update win/apple licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Maisem Ali ecf6cdd830 ssh/tailssh: add TestSSHAuthFlow
Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Maisem Ali f16b77de5d ssh/tailssh: do the full auth flow during ssh auth
Fixes #5091

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
License Updater c8a3d02989 licenses: update android licenses
Signed-off-by: License Updater <noreply@tailscale.com>
2 years ago
Brad Fitzpatrick 6d76764f37 ipn/ipnlocal: fix taildrop target list UI bug
The macOS and iOS apps that used the /localapi/v0/file-targets handler
were getting too many candidate targets. They wouldn't actually accept
the file. This is effectively just a UI glitch in the wrong hosts
being listed as valid targets from the source side.

Change-Id: I6907a5a1c3c66920e5ec71601c044e722e7cb888
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
2 years ago
Maisem Ali b84ec521bf ssh/tailssh: do not send EOT on session disconnection
This was assumed to be the fix for mosh not working, however turns out
all we really needed was the duplicate fd also introduced in the same
commit (af412e8874).

Fixes #5103

Signed-off-by: Maisem Ali <maisem@tailscale.com>
2 years ago
Joe Tsai 82f5f438e0
wgengine/wgcfg: plumb down audit log IDs (#5855)
The node and domain audit log IDs are provided in the map response,
but are ultimately going to be used in wgengine since
that's the layer that manages the tstun.Wrapper.

Do the plumbing work to get this field passed down the stack.

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
2 years ago
Mihai Parparita 92ad56ddcb cmd/tsconnect: close the SSH session an unload event instead of beforeunload
The window may not end up getting unloaded (if other beforeunload
handlers prevent the event), thus we should only close the SSH session
if it's truly getting unloaded.

Updates tailscale/corp#7304

Signed-off-by: Mihai Parparita <mihai@tailscale.com>
2 years ago