[selftests] `mptcp_connect.sh`: `copyfd_io_poll`: poll timed out

[matttbe]

I don't think it happened on our CI (I still need to double-check), but NIPA can "easily" hit such issue:

# 49 ns1 MPTCP -> ns1 (dead:beef:1::1:10046) TCP       (duration    36ms) [ OK ]
# 50 ns1 TCP   -> ns1 (dead:beef:1::1:10047) MPTCP     copyfd_io_poll: poll timed out (events: POLLIN 1, POLLOUT 0)
# (duration 30279ms) [FAIL] file received by server does not match (in, out):
# -rw------- 1 root root 5171914 Jul 16 05:24 /tmp/tmp.W2c96hxSIz
# Trailing bytes are: 
# w,ѐ)-rw------- 1 root root 5166208 Jul 16 05:24 /tmp/tmp.s33PNcrN6M
# Trailing bytes are: 
# (<v /&^<ֱrnFsaC7INFO: with peek mode: saveAfterPeek

logs
history

BTW, as Jakub suggested, we should probably print a hexdump instead of random data :)

[matttbe]

If someone is able to have a look, don't hesitate. A first step is certainly to try reproducing this issue to better understand the root cause, e.g. using stress-ng in parallel or in the VM, changing the sched priorities for the VM, etc.

[matttbe]

Note: the regressions might be more frequent due to this series.

See the discussion here