forked from I2P_Developers/i2p.i2p
dont log a RI DSM down a client tunnel as an error, just silently drop, as probably the result of a FVSJ
This commit is contained in:
@@ -66,6 +66,16 @@ public class InboundMessageDistributor implements GarlicMessageReceiver.CloveRec
|
|||||||
newMsg.setFromHash(orig.getFromHash());
|
newMsg.setFromHash(orig.getFromHash());
|
||||||
newMsg.setSearchKey(orig.getSearchKey());
|
newMsg.setSearchKey(orig.getSearchKey());
|
||||||
msg = newMsg;
|
msg = newMsg;
|
||||||
|
} else if ( (_client != null) &&
|
||||||
|
(msg.getType() == DatabaseStoreMessage.MESSAGE_TYPE) &&
|
||||||
|
(((DatabaseStoreMessage)msg).getValueType() == DatabaseStoreMessage.KEY_TYPE_ROUTERINFO)) {
|
||||||
|
// FVSJ may result in an unsolicited RI store if the peer went non-ff.
|
||||||
|
// Maybe we can figure out a way to handle this safely, so we don't ask him again.
|
||||||
|
// For now, just hope we eventually find out through other means.
|
||||||
|
// Todo: if peer was ff and RI is not ff, queue for exploration in netdb (but that isn't part of the facade now)
|
||||||
|
if (_log.shouldLog(Log.WARN))
|
||||||
|
_log.warn("Dropping DSM down a tunnel for " + _client.toBase64() + ": " + msg);
|
||||||
|
return;
|
||||||
} else if ( (_client != null) &&
|
} else if ( (_client != null) &&
|
||||||
(msg.getType() != DeliveryStatusMessage.MESSAGE_TYPE) &&
|
(msg.getType() != DeliveryStatusMessage.MESSAGE_TYPE) &&
|
||||||
(msg.getType() != GarlicMessage.MESSAGE_TYPE) &&
|
(msg.getType() != GarlicMessage.MESSAGE_TYPE) &&
|
||||||
|
Reference in New Issue
Block a user