ipn/ipnlocal: don't compact TKA state on startup

Compacting on startup means nodes may compact at a different cadence
based on whether they're long-running or restarting frequently.

We already compact after every sync, which only occurs when the TKA
state has changed. Waiting for TKA changes to trigger compaction on
nodes means compaction will occur more consistently across a tailnet.

Updates tailscale/corp#33537

Change-Id: Ia0aa6d9e5e362e9ab08450fde69772841790d5b5
Signed-off-by: Alex Chan <alexc@tailscale.com>
This commit is contained in:
Alex Chan authored and Alex Chan committed 2026-05-01 13:27:12 +01:00
1 parent a6c5d23742
commit cac94f51cc
1 file changed
-4
-4
View File
@@ -96,10 +96,6 @@ func (b *LocalBackend) initTKALocked() error {
return fmt.Errorf("initializing tka: %v", err)
}
if err := authority.Compact(storage, tka.CompactionDefaults); err != nil {
b.logf("tka compaction failed: %v", err)
}
b.tka = &tkaState{
profile: cp.ID(),
authority: authority,