beacon/light: keep retrying checkpoint init if failed (ethereum#33966)

This PR changes the blsync checkpoint init logic so that even if the
initialization fails with a certain server and an error log message is
printed, the server goes back to its initial state and is allowed to
retry initialization after the failure delay period. The previous logic
had an `ssDone` server state that did put the server in a permanently
unusable state once the checkpoint init failed for an apparently
permanent reason. This was not the correct behavior because different
servers behave differently in case of overload and sometimes the
response to a permanently missing item is not clearly distinguishable
from an overload response. A safer logic is to never assume anything to
be permanent and always give a chance to retry.
The failure delay formula is also fixed; now it is properly capped at
`maxFailureDelay`. The previous formula did allow the delay to grow
unlimited if a retry was attempted immediately after each delay period.

Author: zsfelfoldi

beacon/light/request/server.go

@@ -438,14 +438,11 @@ func (s *serverWithLimits) fail(desc string) {
 // failLocked calculates the dynamic failure delay and applies it.
 func (s *serverWithLimits) failLocked(desc string) {
 	log.Debug("Server error", "description", desc)
-	s.failureDelay *= 2
 	now := s.clock.Now()
 	if now > s.failureDelayEnd {
 		s.failureDelay *= math.Pow(2, -float64(now-s.failureDelayEnd)/float64(maxFailureDelay))
 	}
-	if s.failureDelay < float64(minFailureDelay) {
-		s.failureDelay = float64(minFailureDelay)
-	}
+	s.failureDelay = max(min(s.failureDelay*2, float64(maxFailureDelay)), float64(minFailureDelay))
 	s.failureDelayEnd = now + mclock.AbsTime(s.failureDelay)
 	s.delay(time.Duration(s.failureDelay))
 }

beacon/light/sync/update_sync.go

@@ -62,7 +62,6 @@ const (
 	ssNeedParent             // cp header slot %32 != 0, need parent to check epoch boundary
 	ssParentRequested        // cp parent header requested
 	ssPrintStatus            // has all necessary info, print log message if init still not successful
-	ssDone                   // log message printed, no more action required
 )
 
 type serverState struct {
@@ -180,7 +179,8 @@ func (s *CheckpointInit) Process(requester request.Requester, events []request.E
 		default:
 			log.Error("blsync: checkpoint not available, but reported as finalized; specified checkpoint hash might be too old", "server", server.Name())
 		}
-		s.serverState[server] = serverState{state: ssDone}
+		s.serverState[server] = serverState{state: ssDefault}
+		requester.Fail(server, "checkpoint init failed")
 	}
 }