ubuntu_kernel_selftests:net:fib_nexthops.sh: IP4 torture test commands get killed

[Stefan Bader]

On kernel06 (s390x.zvm) with hirsute:linux (5.11.0-40.44 ) for sru-20211018 I found the following:

11:29:40 DEBUG| [stdout] # IPv4 runtime torture
11:29:40 DEBUG| [stdout] # --------------------
11:34:40 DEBUG| [stdout] # TEST: IPv4 torture test [ OK ]
11:34:40 DEBUG| [stdout] # ./fib_nexthops.sh: line 186: 48485 Killed ipv4_del_add_loop1
11:34:40 DEBUG| [stdout] # ./fib_nexthops.sh: line 186: 48486 Killed ipv4_grp_replace_loop
11:34:40 DEBUG| [stdout] # ./fib_nexthops.sh: line 186: 48487 Killed ip netns exec me ping -f 172.16.101.1 > /dev/null 2>&1
11:34:40 DEBUG| [stdout] # ./fib_nexthops.sh: line 186: 48488 Killed ip netns exec me ping -f 172.16.101.2 > /dev/null 2>&1
11:34:40 DEBUG| [stdout] # ./fib_nexthops.sh: line 186: 48490 Killed ip netns exec me mausezahn veth1 -B 172.16.101.2 -A 172.16.1.1 -c 0 -t tcp "dp=1-1023, flags=syn" > /dev/null 2>&1

I don't think that was in the log of the first crank of that kernel (which only changed the zfs module), so I don't think it is a regression.

[Stefan Bader (smb)]

I got fooled by this this time. The IP4 and IP6 torture test seems to pass with the killed commands. What fails is the ping with nexthops test. (bug #1945305).