eyjhb/headscale
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

replace ephemeral deletion logic (#2008)

Browse source 
* replace ephemeral deletion logic

this commit replaces the way we remove ephemeral nodes,
currently they are deleted in a loop and we look at last seen
time. This time is now only set when a node disconnects and
there was a bug (#2006) where nodes that had never disconnected
was deleted since they did not have a last seen.

The new logic will start an expiry timer when the node disconnects
and delete the node from the database when the timer is up.

If the node reconnects within the expiry, the timer is cancelled.

Fixes #2006

Signed-off-by: Kristoffer Dalby <kristoffer@tailscale.com>

* use uint64 as authekyid and ptr helper in tests

Signed-off-by: Kristoffer Dalby <kristoffer@tailscale.com>

* add test db helper

Signed-off-by: Kristoffer Dalby <kristoffer@tailscale.com>

* add list ephemeral node func

Signed-off-by: Kristoffer Dalby <kristoffer@tailscale.com>

* schedule ephemeral nodes for removal on startup

Signed-off-by: Kristoffer Dalby <kristoffer@tailscale.com>

* fix gorm query for postgres

Signed-off-by: Kristoffer Dalby <kristoffer@tailscale.com>

* add godoc

Signed-off-by: Kristoffer Dalby <kristoffer@tailscale.com>

---------

Signed-off-by: Kristoffer Dalby <kristoffer@tailscale.com>
This commit is contained in:
Kristoffer Dalby 2024-07-18 10:01:59 +02:00 • committed by GitHub
parent 58bd38a609
commit 7e62031444
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
13 changed files with 417 additions and 206 deletions
Download patch file Download diff file Expand all files Collapse all files

65
hscontrol/app.go
View file

@ -91,6 +91,7 @@ type Headscale struct {
db *db.HSDatabase
ipAlloc *db.IPAllocator
noisePrivateKey *key.MachinePrivate
ephemeralGC *db.EphemeralGarbageCollector
DERPMap *tailcfg.DERPMap
DERPServer *derpServer.DERPServer
@ -153,6 +154,12 @@ func NewHeadscale(cfg *types.Config) (*Headscale, error) {
return nil, err
}
app.ephemeralGC = db.NewEphemeralGarbageCollector(func(ni types.NodeID) {
if err := app.db.DeleteEphemeralNode(ni); err != nil {
log.Err(err).Uint64("node.id", ni.Uint64()).Msgf("failed to delete ephemeral node")
}
})
if cfg.OIDC.Issuer != "" {
err = app.initOIDC()
if err != nil {
@ -217,47 +224,6 @@ func (h *Headscale) redirect(w http.ResponseWriter, req *http.Request) {
http.Redirect(w, req, target, http.StatusFound)
}
// deleteExpireEphemeralNodes deletes ephemeral node records that have not been
// seen for longer than h.cfg.EphemeralNodeInactivityTimeout.
func (h *Headscale) deleteExpireEphemeralNodes(ctx context.Context, every time.Duration) {
ticker := time.NewTicker(every)
for {
select {
case <-ctx.Done():
ticker.Stop()
return
case <-ticker.C:
var removed []types.NodeID
var changed []types.NodeID
if err := h.db.Write(func(tx *gorm.DB) error {
removed, changed = db.DeleteExpiredEphemeralNodes(tx, h.cfg.EphemeralNodeInactivityTimeout)
return nil
}); err != nil {
log.Error().Err(err).Msg("database error while expiring ephemeral nodes")
continue
}
if removed != nil {
ctx := types.NotifyCtx(context.Background(), "expire-ephemeral", "na")
h.nodeNotifier.NotifyAll(ctx, types.StateUpdate{
Type: types.StatePeerRemoved,
Removed: removed,
})
}
if changed != nil {
ctx := types.NotifyCtx(context.Background(), "expire-ephemeral", "na")
h.nodeNotifier.NotifyAll(ctx, types.StateUpdate{
Type: types.StatePeerChanged,
ChangeNodes: changed,
})
}
}
}
}
// expireExpiredNodes expires nodes that have an explicit expiry set
// after that expiry time has passed.
func (h *Headscale) expireExpiredNodes(ctx context.Context, every time.Duration) {
@ -557,9 +523,18 @@ func (h *Headscale) Serve() error {
return errEmptyInitialDERPMap
}
expireEphemeralCtx, expireEphemeralCancel := context.WithCancel(context.Background())
defer expireEphemeralCancel()
go h.deleteExpireEphemeralNodes(expireEphemeralCtx, updateInterval)
// Start ephemeral node garbage collector and schedule all nodes
// that are already in the database and ephemeral. If they are still
// around between restarts, they will reconnect and the GC will
// be cancelled.
go h.ephemeralGC.Start()
ephmNodes, err := h.db.ListEphemeralNodes()
if err != nil {
return fmt.Errorf("failed to list ephemeral nodes: %w", err)
}
for _, node := range ephmNodes {
h.ephemeralGC.Schedule(node.ID, h.cfg.EphemeralNodeInactivityTimeout)
}
expireNodeCtx, expireNodeCancel := context.WithCancel(context.Background())
defer expireNodeCancel()
@ -809,7 +784,7 @@ func (h *Headscale) Serve() error {
Msg("Received signal to stop, shutting down gracefully")
expireNodeCancel()
expireEphemeralCancel()
h.ephemeralGC.Close()
trace("waiting for netmap stream to close")
h.pollNetMapStreamWG.Wait()