diff options
| author | Simon Law <sfllaw@tailscale.com> | 2025-09-29 19:16:54 -0700 |
|---|---|---|
| committer | Simon Law <sfllaw@tailscale.com> | 2025-10-06 15:42:26 -0700 |
| commit | c386bc9e9cca9563e7e63f159c506fc7dc098d96 (patch) | |
| tree | 7bc57e3877896ef6e30a9a4620a345b28afdb7f0 /ssh/tailssh | |
| parent | 5555bbcd222f6f0e757d458f4251d87450d67b41 (diff) | |
| download | tailscale-sfllaw/tailscale-ping-for-client-side-reachability.tar.xz tailscale-sfllaw/tailscale-ping-for-client-side-reachability.zip | |
ipn/ipnlocal: determine client-side-reachability by pingingsfllaw/tailscale-ping-for-client-side-reachability
PR #17367 introduced the `client-side-reachability` node attribute
that switches the node to completely ignore the online signal from
control. This was to work around the times when the control plane
determines that a node is not online, while the node is still able to
connect to its peers. However, this breaks high-availability failover,
since nodes that are actually offline are never treated that way.
This PR makes nodeBackend.PeerIsReachable actually test reachability
by disco-pinging the peer node.
In the future, the client itself should collect reachability data from
active Wireguard flows.
Updates #17366
Updates tailscale/corp#30379
Updates tailscale/corp#32686
Updates tailscale/corp#33033
Signed-off-by: Simon Law <sfllaw@tailscale.com>
Diffstat (limited to 'ssh/tailssh')
0 files changed, 0 insertions, 0 deletions