You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
when send one transaction to one rpc neb service, the neb rpc service will receive the same transaction message again .
Error Log
msg="Failed to push a tx into tx pool." err="duplicated transaction" file=asm_amd64.s func=runtime.goexit goroutine=16 line=2338 messageType=newtx transaction="{"chainID":1001, "hash":"cc29e71ce53682be97663f2fbcc57490b9b2ac355828b46604576c06c30a63fe", "from":"e81f56a0bccf6500c8458abfa79b4287715d74edc59c2ad9", "to":"bbb213b48cf7554885177ad889c92be409e4ab08595273ab", "nonce":171, "value":"1", "timestamp":1517024148, "gasprice": "1000000", "gaslimit":"20000", "type":"binary"}"
Cause analysis
in p2p stream ,when neb node received other node msg, it will add the message (node.pid+message.hash) to bloom filter. it avoid to send parent node again.
But when the first node send one message to second node, the second node send the same message to the third node , the third node may send the same message to the first node。
the msg send routing path
A--> B --> C --> A
Solution
the firs solution : When send message to other node , add the messages to bloom filter , it avoid to filter message not to push the pool, but not avoid to receive again
the second solution : avoid send A again A--> B --> C -XX-> A , every node shoud record all all the received the same message nodes.
We will fix the bug later
The text was updated successfully, but these errors were encountered:
Description
when send one transaction to one rpc neb service, the neb rpc service will receive the same transaction message again .
Error Log
Cause analysis
in p2p stream ,when neb node received other node msg, it will add the message (node.pid+message.hash) to bloom filter. it avoid to send parent node again.
But when the first node send one message to second node, the second node send the same message to the third node , the third node may send the same message to the first node。
the msg send routing path
Solution
the firs solution : When send message to other node , add the messages to bloom filter , it avoid to filter message not to push the pool, but not avoid to receive again
the second solution : avoid send A again A--> B --> C -XX-> A , every node shoud record all all the received the same message nodes.
We will fix the bug later
The text was updated successfully, but these errors were encountered: