Retry mechanism when getting a broadcasted block fail on all peers (h…

…yperledger#4271)

Signed-off-by: Fabio Di Fabio <[email protected]>

CHANGELOG.md

@@ -5,6 +5,7 @@
 ### Additions and Improvements
 - Upgrade besu-native to 0.6.0 and use Blake2bf native implementation if available by default [#4264](https://github.com/hyperledger/besu/pull/4264)
 - Better management of jemalloc presence/absence in startup script [#4237](https://github.com/hyperledger/besu/pull/4237)
+- Retry mechanism when getting a broadcasted block fail on all peers [#4271](https://github.com/hyperledger/besu/pull/4271)
 - Filter out disconnected peers when fetching available peers [#4269](https://github.com/hyperledger/besu/pull/4269)
 - Updated the default value of fast-sync-min-peers post merge [#4298](https://github.com/hyperledger/besu/pull/4298)
 - Log imported block info post merge [#4310](https://github.com/hyperledger/besu/pull/4310)

ethereum/eth/src/main/java/org/hyperledger/besu/ethereum/eth/manager/task/AbstractRetryingPeerTask.java

@@ -114,9 +114,9 @@ protected void handleTaskError(final Throwable error) {
 
     if (cause instanceof NoAvailablePeersException) {
       LOG.debug(
-          "No useful peer found, checking remaining current peers for usefulness: {}",
+          "No useful peer found, wait max 5 seconds for new peer to connect: current peers {}",
           ethContext.getEthPeers().peerCount());
-      // Wait for new peer to connect
+
       final WaitForPeerTask waitTask = WaitForPeerTask.create(ethContext, metricsSystem);
       executeSubTask(
           () ->