Commit Graph
6 Commits
Author SHA1 Message Date
Alex ChanandAlex Chan cd153aa644 control, ipn, tailcfg: enable seamless key renewal by default
Previously, seamless key renewal was an opt-in feature.  Customers had
to set a `seamless-key-renewal` node attribute in their policy file.

This patch enables seamless key renewal by default for all clients.

It includes a `disable-seamless-key-renewal` node attribute we can set
in Control, so we can manage the rollout and disable the feature for
clients with known bugs.  This new attribute makes the feature opt-out.

Updates tailscale/corp#31479

Signed-off-by: Alex Chan <alexc@tailscale.com>
2025-09-18 09:59:46 +01:00
Alex ChanandAlex Chan db0b9a361c net/dns: don't timeout if inotify sends multiple events
This fixes a flaky test which has been occasionally timing out in CI.

In particular, this test times out if `watchFile` receives multiple
notifications from inotify before we cancel the test context. We block
processing the second notification, because we've stopped listening to
the `callbackDone` channel.

This patch changes the test so we only send on the first notification.

Testing this locally with `stress` confirms that the test is no longer
flaky.

Fixes #17172
Updates #14699

Signed-off-by: Alex Chan <alexc@tailscale.com>
2025-09-17 14:51:17 +01:00
Alex ChanandAlex Chan 312582bdbf ssh/tailssh: mark TestSSHRecordingCancelsSessionsOnUploadFailure as flaky
Updates https://github.com/tailscale/tailscale/issues/7707

Signed-off-by: Alex Chan <alexc@tailscale.com>
2025-09-17 11:27:00 +01:00
Alex ChanandAlex Chan 2b0f59cd38 logpolicy: remove the deprecated and now-unused NewWithConfigPath
Updates #cleanup

Signed-off-by: Alex Chan <alexc@tailscale.com>
2025-09-16 19:24:30 +01:00
Alex ChanandAlex Chan 84659b1dc6 ipn: fix the string representation of an empty ipn.Notify
Before: `ipn.Notify}`
After:  `ipn.Notify{}`

Updates #cleanup

Signed-off-by: Alex Chan <alexc@tailscale.com>
2025-09-16 18:13:49 +01:00
Alex ChanandAlex Chan 5c24f0ed80 wgengine/magicsock: send a valid payload in TestNetworkDownSendErrors
This test ostensibly checks whether we record an error metric if a packet
is dropped because the network is down, but the network connectivity is
irrelevant -- the send error is actually because the arguments to Send()
are invalid:

    RebindingUDPConn.WriteWireGuardBatchTo:
    [unexpected] offset (0) != Geneve header length (8)

This patch changes the test so we try to send a valid packet, and we
verify this by sending it once before taking the network down.  The new
error is:

    magicsock: network down

which is what we're trying to test.

We then test sending an invalid payload as a separate test case.

Updates tailscale/corp#22075

Signed-off-by: Alex Chan <alexc@tailscale.com>
2025-09-15 23:26:32 +01:00