./rpcping tcp 127.0.0.1 failed in [email protected] on centos8_aarch64

[Tom-python0121]

Hello, I meet a problem :./rpcping tcp 127.0.0.1 failed in [email protected] on centos8_aarch64 when I want to test it.

Steps to reproduce the issue

[root@localhost tests]# ls
CMakeFiles  Makefile  cmake_install.cmake  rpcping
[root@localhost tests]# ./rpcping
Usage: rpcping <raw|rdma|tcp|udp> <host> [--rpcbind] [--count=<n>] [--threads=<n>] [--workers=<n>] [--port=<n>] [--program=<n>] [--version=<n>] [--procedure=<n>]
[root@localhost tests]# ./rpcping raw 127.0.0.1
Segmentation fault (core dumped)
[root@localhost tests]# ./rpcping rdma 127.0.0.1
Segmentation fault (core dumped)
[root@localhost tests]# ./rpcping tcp 127.0.0.1
Segmentation fault (core dumped)
[root@localhost tests]# ./rpcping udp 127.0.0.1
Segmentation fault (core dumped)

Why did this error occur, and how did you fix it to make it work?

[Tom-python0121]

The same error is reported on the x86_64 system. However, in [email protected], both aarch64 and x86_64 tests are normal, as shown in the following figure: [root@localhost tests]# ./rpcping tcp 127.0.0.1
rpcping tcp 127.0.0.1 count=500 threads=1 workers=5 (port=2049 program=100003 version=3 procedure=0): failures 0 timeouts 500 mean 16.3974, total 16.3974

[Tom-python0121]

Does the problem that memory leak occurs when source codes of other functions are added to [email protected] or later? Segmentation fault (core dump)

[ffilz]

Is there a reason to use rpcping instead of rpcinfo which has "ping" capability also?

[ffilz]

This is a duplicate of #220. Closing.