Commit Graph

6308 Commits (8428a64b562542aebce3812ad196757661e390c8)
 

Author SHA1 Message Date
Andrew Lytvynov eef15b4ffc
cmd/dist,release/dist: sign release tarballs with an ECDSA key (#8759)
Pass an optional PEM-encoded ECDSA key to `cmd/dist` to sign all built
tarballs. The signature is stored next to the tarball with a `.sig`
extension.

Tested this with an `openssl`-generated key pair and verified the
resulting signature.

Updates #8760

Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
1 year ago
David Anderson ed46442cb1 client/tailscale/apitype: document never-nil property of WhoIsResponse
Every time I use WhoIsResponse I end up writing mildly irritating nil-checking
for both Node and UserProfile, but it turns out our code guarantees that both
are non-nil in successful whois responses.

Updates #cleanup

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
Brad Fitzpatrick 5ebb271322 derp/derphttp: add optional Client.BaseContext hook
Like net/http.Server.BaseContext, this lets callers specify a base
context for dials.

Updates tailscale/corp#12702

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Maisem Ali 058d427fa6 tailcfg: add helper to unmarshal PeerCap values
Updates #4217

Signed-off-by: Maisem Ali <maisem@tailscale.com>
1 year ago
salman aljammaz 68f8e5678e
wgengine/magicsock: remove dead code (#8745)
The nonce value is not read by anything, and di.sharedKey.Seal()
a few lines below generates its own. #cleanup

Signed-off-by: salman <salman@tailscale.com>
1 year ago
License Updater 0554deb48c licenses: update tailscale{,d} licenses
Signed-off-by: License Updater <noreply+license-updater@tailscale.com>
1 year ago
David Anderson 6114247d0a types/logid: add a Compare method
Updates #cleanup

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
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
Claire Wang 90a7d3066c
derp: use tstime (#8634)
Updates #8587

Signed-off-by: Claire Wang <claire@tailscale.com>
1 year ago
Claire Wang 2315bf246a
ipn: use tstime (#8597)
Updates #8587
Signed-off-by: Claire Wang <claire@tailscale.com>
1 year ago
Andrew Lytvynov c1ecae13ab
ipn/{ipnlocal,localapi}: actually renew certs before expiry (#8731)
While our `shouldStartDomainRenewal` check is correct, `getCertPEM`
would always bail if the existing cert is not expired. Add the same
`shouldStartDomainRenewal` check to `getCertPEM` to make it proceed with
renewal when existing certs are still valid but should be renewed.

The extra check is expensive (ARI request towards LetsEncrypt), so cache
the last check result for 1hr to not degrade `tailscale serve`
performance.

Also, asynchronous renewal is great for `tailscale serve` but confusing
for `tailscale cert`. Add an explicit flag to `GetCertPEM` to force a
synchronous renewal for `tailscale cert`.

Fixes #8725

Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
1 year ago
tinku-tailscale aa37be70cf
api.md: update device authorize API docs to allow for deauth (#8728)
Signed-off-by: tinku-tailscale <139132124+tinku-tailscale@users.noreply.github.com>
1 year ago
Charlotte Brandhorst-Satzkorn 35bdbeda9f cli: introduce exit-node subcommand to list and filter exit nodes
This change introduces a new subcommand, `exit-node`, along with a
subsubcommand of `list` and a `--filter` flag.

Exit nodes without location data will continue to be displayed when
`status` is used. Exit nodes with location data will only be displayed
behind `exit-node list`, and in status if they are the active exit node.

The `filter` flag can be used to filter exit nodes with location data by
country.

Exit nodes with Location.Priority data will have only the highest
priority option for each country and city listed. For countries with
multiple cities, a <Country> <Any> option will be displayed, indicating
the highest priority node within that country.

Updates tailscale/corp#13025

Signed-off-by: Charlotte Brandhorst-Satzkorn <charlotte@tailscale.com>
1 year ago
David Anderson 9d89e85db7 wgengine/magicsock: document mysterious-looking assignment
Updates #cleanup

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 84777354a0 wgengine/magicsock: factor out more separable parts
Updates #8720

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson 9a76deb4b0 disco: move disco pcap helper to disco package
Updates tailscale/corp#13464

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson cde37f5307 wgengine/magicsock: factor out peerMap into separate file
Updates tailscale/corp#13464

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson f7016d8c00 wgengine/magicsock: factor out endpoint into its own file
Updates tailscale/corp#13464

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
David Anderson c2831f6614 wgengine/magicsock: delete unused stuff
Updates tailscale/corp#13464

Signed-off-by: David Anderson <danderson@tailscale.com>
1 year ago
Andrew Lytvynov 9edb848505
cmd/tailscale/cli: implement update on FreeBSD (#8710)
Implement `tailscale update` on FreeBSD. This is much simpler than other
platforms because `pkg rquery` lets us get the version in their repos
without any extra parsing.

Updates #6995

Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
1 year ago
Maisem Ali 1ecc16da5f tailcfg,ipn/ipnlocal,wgengine: add values to PeerCapabilities
Define PeerCapabilty and PeerCapMap as the new way of sending down
inter-peer capability information.

Previously, this was unstructured and you could only send down strings
which got too limiting for certain usecases. Instead add the ability
to send down raw JSON messages that are opaque to Tailscale but provide
the applications to define them however they wish.

Also update accessors to use the new values.

Updates #4217

Signed-off-by: Maisem Ali <maisem@tailscale.com>
1 year ago
Andrew Lytvynov 306deea03a
cmd/tailscale/cli,version/distro: update support for Alpine (#8701)
Similar to Arch support, use the latest version info from the official
`apk` repo and don't offer explicit track or version switching.
Add detection for Alpine Linux in version/distro along the way.

Updates #6995

Signed-off-by: Andrew Lytvynov <awly@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
License Updater 4c05d43008 licenses: update tailscale{,d} licenses
Signed-off-by: License Updater <noreply+license-updater@tailscale.com>
1 year ago
Andrew Lytvynov 894b237a70
cmd/tailscale/cli: implement update for dnf/yum-based distros (#8678)
This is the Fedora family of distros, including CentOS, RHEL and others.
Tested in `fedora:latest` and `centos:7` containers.

Updates #6995

Signed-off-by: Andrew Lytvynov <awly@tailscale.com>
1 year ago
Brad Fitzpatrick f1cc8ab3f9 tailcfg: add UserProfile.Groups
Updates tailscale/corp#13375

Signed-off-by: Brad Fitzpatrick <bradfitz@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
Brad Fitzpatrick 453620dca1 go.toolchain.rev: bump Go version
For a96a9eddc0

Updates tailscale/corp#12702

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Brad Fitzpatrick 41db1d7bba tailcfg: deprecate Debug, flesh out Node.DERP docs
Updates #docs

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago
Jenny Zhang 907c56c200 api.md: add documentation to API endpoint about SCIM group warnings
Updates #8645
Signed-off-by: Jenny Zhang <jz@tailscale.com>
1 year ago
Claire Wang e1bcecc393
logtail: use tstime (#8607)
Updates #8587
Signed-off-by: Claire Wang <claire@tailscale.com>
1 year ago
Joe Tsai bb4b35e923 ssh: ignore io.EOF from sftp.Server.Serve
If the connection provided to sftp.NewServer is closed,
Serve returns the io.EOF error verbatim from io.Reader.Read.
This is an odd error since this is an expected situation,
so we manually ignore io.EOF.
This is somewhat buggy since the sftp package itself
incorrectly reports io.EOF in cases where it should actually
be reporting io.ErrUnexpectedEOF.
See https://github.com/pkg/sftp/pull/554 which patches Serve to
return nil on clean closes and fixes buggy uses of io.ReadFull.

Fixes #8592

Signed-off-by: Joe Tsai <joetsai@digital-static.net>
1 year ago
Brad Fitzpatrick 88cc0ad9f7 util/linuxfw: remove yet-unused code to fix linux/arm64 crash
The util/linuxfw/iptables.go had a bunch of code that wasn't yet used
(in prep for future work) but because of its imports, ended up
initializing code deep within gvisor that panicked on init on arm64
systems not using 4KB pages.

This deletes the unused code to delete the imports and remove the
panic. We can then cherry-pick this back to the branch and restore it
later in a different way.

A new test makes sure we don't regress in the future by depending on
the panicking package in question.

Fixes #8658

Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
1 year ago