The branch main has been updated by glebius: URL: https://cgit.FreeBSD.org/src/commit/?id=38edf96b1787ce3d8c00e4466348dab891c7a9ea
commit 38edf96b1787ce3d8c00e4466348dab891c7a9ea Author: Gleb Smirnoff <[email protected]> AuthorDate: 2026-02-19 02:39:00 +0000 Commit: Gleb Smirnoff <[email protected]> CommitDate: 2026-02-19 02:53:16 +0000 tests/ipfw: fix log:bpf test flakyness There were several problems: o Using 'netstat -B' is not a reliable way to make sure that all tcpdumps have attached to bpf(4). The problem is that tcpdump (via libpcap) does several ioctl(2)s after the attach including two BIOCSETF. Each of them flushes the input buffer. So we can see tcpdump attached in 'netstat -B' and start sending packets and the packet will be captured by bpf(4) before BIOCSETF and freed and tcpdump won't read anything. Instead of using netstat(1), use ps(1) and make sure each tcpdump is blocked on the "bpf" wait channel, which guarantees it is done with ioctl(2)s and is now blocked in read(2). o Using 'nc -w 0' sets timeout not only on the connect(2) (as documented) but also on poll(2), which is not documented. There is a race in shell that will make stdin not yet filled by 'echo foo' when nc(1) does poll(2). With zero timeout, this poll(2) will immediately return and nc will exit. o The waiting loop had two errors: using wrong variable name as well as invoking a subshell, that actually can't wait on the pid. o The reading tcpdump was lacking '-q' option, that prevents any protocol interpretations. Sometimes, when random port chosen by nc(1) would match some well-known (to tcpdump) port, the output would differ from the expected. PR: 293241 --- tests/sys/netpfil/ipfw/log.sh | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-) diff --git a/tests/sys/netpfil/ipfw/log.sh b/tests/sys/netpfil/ipfw/log.sh index 7df5c69e4219..cde909e25d1e 100755 --- a/tests/sys/netpfil/ipfw/log.sh +++ b/tests/sys/netpfil/ipfw/log.sh @@ -57,31 +57,34 @@ bpf_body() pids="${pids} $!" done - # wait for tcpdumps to attach, include netstat(1) header in ${count} - count=$(( $(echo ${rules} ${auto} | wc -w) + 1)) - while [ $(jexec alcatraz netstat -B | wc -l) -ne ${count} ]; do - sleep 0.01; + # wait for tcpdumps to fully attach and block in bpfread() + for p in ${pids}; do + while [ $(ps -o wchan ${p} | tr "\n" " " | cut -w -f 2) != \ + "bpf" ]; do + sleep 0.01; + done done for p in ${rules} 666; do - echo foo | nc -u 192.0.2.1 10${p} -w 0 + echo foo | nc -u 192.0.2.1 10${p} done for p in ${pids}; do - atf_check -s exit:0 sh -c "wait $pid; exit $?" + wait ${p} + atf_check_equal 0 $? done # statically numbered taps for p in ${rules}; do atf_check -o match:"192.0.2.0.[0-9]+ > 192.0.2.1.10${p}: UDP" \ -e match:"reading from file [a-zA-Z0-9/.]+${p}.pcap" \ - tcpdump -nr ${PWD}/${p}.pcap + tcpdump -qnr ${PWD}/${p}.pcap done # autonumbered tap with 10666 port atf_check -o match:"192.0.2.0.[0-9]+ > 192.0.2.1.10666: UDP" \ -e match:"reading from file [a-zA-Z0-9/.]+${auto}.pcap" \ - tcpdump -nr ${PWD}/${auto}.pcap + tcpdump -qnr ${PWD}/${auto}.pcap } bpf_cleanup()
