ping output and DNS failure semantics differ from Linux iputils

[matt-dz]

What was run

ping -c 2 -W 1 127.0.0.1
ping -c 1 -W 1 kafka

Expected output

For supported ping invocations on Linux, rshell should match GNU iputils semantics closely enough for diagnostic use.

Oracle environment:

Image: pshim-kind-node:v1.32.0, built FROM kindest/node:v1.32.0
Linux flavor: Debian GNU/Linux 12 (bookworm), aarch64
Oracle shell: /usr/bin/bash --noprofile --norc <temp-script>
bash package: 5.2.15-2+b10
iputils-ping package: 3:20221126-1+deb12u1
ping -V: ping from iputils 20221126

Representative oracle output for a successful ping:

PING 127.0.0.1 (127.0.0.1) 56(84) bytes of data.
64 bytes from 127.0.0.1: icmp_seq=1 ttl=64 time=0.046 ms

--- 127.0.0.1 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1062ms
rtt min/avg/max/mdev = 0.044/0.045/0.046/0.001 ms

Representative oracle output for DNS failure:

ping: kafka: Temporary failure in name resolution

with exit code 2.

Actual output

rshell emits BSD-style success output and different DNS failure semantics:

PING 127.0.0.1 (127.0.0.1): 56 data bytes
64 bytes from 127.0.0.1: icmp_seq=1 ttl=64 time=0.181 ms

--- 127.0.0.1 ping statistics ---
2 packets transmitted, 2 received, 0.0% packet loss
round-trip min/avg/max/stddev = 0.151/0.166/0.181/0.014 ms

For DNS failure, rshell returned exit code 1 and stderr like:

ping: lookup kafka on 172.18.0.1:53: server misbehaving

Why this matters

Agents and scripts often inspect ping output, packet-loss summaries, and exit codes while diagnosing networking incidents. The current differences can make Linux-focused diagnostics behave differently under rshell even when the same supported command line is used.

These differences were found during rshell QA by comparing supported ping invocations against a Linux oracle.