Tuvie commented on issue #2265:
URL: https://github.com/apache/brpc/issues/2265#issuecomment-1580845198
> 当某个节点出现异常日志是，另一端（10.156.8.16）上没有任何异常日志
> ```
> W0607 18:08:44.002235 661
external/brpc/src/brpc/input_messenger.cpp:240] Close Socket{id=8589934640
fd=1395 addr=10.156.8.16:25696:10643} (0x5616d4012c00) due to unknown message:
\00\00\00\00\00\00\80?\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\80?\00\00\00\00\00\00\00\00\00\00\00\00\00\00\80?...<skipping
8096 bytes>
> W0607 18:08:44.002683 661
external/brpc/src/brpc/policy/baidu_rpc_protocol.cpp:265] Fail to write into
Socket{id=8589934640 fd=1395 addr=10.156.8.16:25696:10643} (0x5616d4012c00):
Invalid argument
> W0607 18:08:44.106501 626
external/brpc/src/brpc/rdma/rdma_endpoint.cpp:564] Fail to read Hello Message
from client:Socket{id=2149 fd=1395 addr=10.156.8.16:26662:10643}
(0x0x561c9d2e8480) 10.156.8.16:26662: Got EOF
> ```
这个问题没有本地复现的话不是很容易定位。上面显示的unknown的数据有什么特征吗？错误发生的时间前后有什么迹象吗？错误的发生和规模、执行时长、数据长度有什么关联吗？上层业务调用rpc的模式是否可能用rdma的example来表达？
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
