Skip to content

libtailscale: do not close old TUNs during reconfig #790

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
nickkhyl merged 2 commits into from
May 15, 2026
+17 −27
Merged

libtailscale: do not close old TUNs during reconfig #790

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
870630f
libtailscale: do not close old TUNs during reconfig, except on ChromeOS
nickkhyl May 12, 2026
2393aa2
libtailscale: do not close old TUNs during reconfig on ChromeOS
nickkhyl May 12, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
44 changes: 17 additions & 27 deletions libtailscale/net.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -80,36 +80,27 @@ func (b *backend) updateTUN(rcfg *router.Config, dcfg *dns.OSConfig) (err error)
b.logger.Logf("updateTUN: changed")
defer b.logger.Logf("updateTUN: finished")

// Close previous tunnel(s).
// This is necessary for ChromeOS, native Android devices
// seem to handle seamless handover between tunnels correctly.
//
// TODO(eliasnaur): If seamless handover becomes a desirable feature, skip
// the closing on ChromeOS.
b.logger.Logf("updateTUN: closing old TUNs")
b.CloseTUNs()
b.logger.Logf("updateTUN: closed old TUNs")

// Since the previous tunnel(s) are closed, the [multiTUN] device is
// not operational until a new underlying tunnel is created and added,
// which may never happen in case of an error or an empty [router.Config].
//
// Therefore, to prevent deadlocks where a [multiTUN.Write] would
// block waiting for a new tunnel to be added, we bring the multiTUN
// device down on exit unless a new [tun.Device] is created and added
// successfully. See tailscale/tailscale#18679.
//
// TODO(nickkhyl): revisit and simplify the [multiTUN] implementation?
hasTunnel := false
defer func() {
if !hasTunnel && b.devices.Down() {
if disableTUN := len(rcfg.LocalAddrs) == 0; disableTUN {
b.logger.Logf("updateTUN: closing old TUNs")
b.CloseTUNs()
b.logger.Logf("updateTUN: closed old TUNs")

// Since the previous tunnel(s) are closed, the [multiTUN] device is
// not operational until a new underlying tunnel is created and added,
// which may never happen in case of an error or an empty [router.Config].
//
// Therefore, to prevent deadlocks where a [multiTUN.Write] would
// block waiting for a new tunnel to be added, we bring the multiTUN
// device down on exit unless a new [tun.Device] is created and added
// successfully. See tailscale/tailscale#18679.
//
// TODO(nickkhyl): revisit and simplify the [multiTUN] implementation?
if b.devices.Down() {
b.logger.Logf("updateTUN: tunnel brought down: %v", err)
}
}()

if len(rcfg.LocalAddrs) == 0 {
return nil
}

builder := vpnService.service.NewBuilder()
b.logger.Logf("updateTUN: got new builder")

Expand Down Expand Up @@ -244,7 +235,6 @@ func (b *backend) updateTUN(rcfg *router.Config, dcfg *dns.OSConfig) (err error)
b.logger.Logf("updateTUN: created TUN device")

b.devices.add(tunDev)
hasTunnel = true
b.logger.Logf("updateTUN: added TUN device")

if b.devices.Up() {
Expand Down
Loading