jech / shncpd Goto Github PK

While it does remove it if it is killed, should it not set a lifetime instead?

root@dancer:/git/shncpd# ip -6 addr show eno1
2: eno1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000
inet6 2601:646:4102:c5::123/128 scope global
valid_lft forever preferred_lft forever
inet6 2601:646:4180:d9f:e4cc:7e3:3277:57a8/128 scope global
valid_lft forever preferred_lft forever
inet6 fe80::eea8:6bff:fefe:9a2/64 scope link
valid_lft forever preferred_lft forever
root@dancer:/git/shncpd# fg
shncpd -M wlp2s0 -M eno1 -D
^Croot@dancer:~/git/shncpd# ip -6 addr show eno1
2: eno1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000
inet6 2601:646:4102:c5::123/128 scope global
valid_lft forever preferred_lft forever
inet6 fe80::eea8:6bff:fefe:9a2/64 scope link
valid_lft forever preferred_lft forever

I took shcncpd for a spin yesterday, finding the basic functionality working, only to find this morning
that it was eating 100% of cpu. It is not making a lot of syscalls, one of these goes by every few seconds, but that is certainly not enough to explain the cpu usage. I will try to fire it up again in a debugger to see if I can make it happen again.

sendto(3, "\0\3\0\10\231\212a\370\0\0\0\3\0\4\0\10R\242i\377]{\202\31", 24, 0, {sa_family=AF_INET6, sin6_port=htons(8808), inet_pton(AF_INET6, "ff02::8808", &sin6_addr), sin6_flowinfo=0, sin6_scope_id=if_nametoindex("wlp2s0")}, 28) = 24
socket(PF_LOCAL, SOCK_DGRAM|SOCK_CLOEXEC, 0) = 4
ioctl(4, SIOCGIFINDEX, {ifr_name="wlp2s0", ifr_index=3}) = 0
close(4) = 0
open("/proc/sys/net/ipv6/conf/all/forwarding", O_RDONLY) = 4
read(4, "1\n", 99) = 2
close(4) = 0
rt_sigaction(SIGINT, {SIG_IGN, [], SA_RESTORER, 0x7f2364a6d2f0}, {0x402b16, [], SA_RESTORER, 0x7f2364a6d2f0}, 8) = 0
rt_sigaction(SIGQUIT, {SIG_IGN, [], SA_RESTORER, 0x7f2364a6d2f0}, {SIG_DFL, [], SA_RESTORER, 0x7f2364a6d2f0}, 8) = 0
rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0
clone(child_stack=0, flags=CLONE_PARENT_SETTID|SIGCHLD, parent_tidptr=0x7ffcf738a3ac) = 9824
wait4(9824, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0, NULL) = 9824
rt_sigaction(SIGINT, {0x402b16, [], SA_RESTORER, 0x7f2364a6d2f0}, NULL, 8) = 0
rt_sigaction(SIGQUIT, {SIG_DFL, [], SA_RESTORER, 0x7f2364a6d2f0}, NULL, 8) = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
--- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=9824, si_status=0, si_utime=0, si_stime=0} ---

It is configured on this x86 box as:

d 3632 99.2 0.0 6540 1668 pts/30 R Apr11 806:29 shncpd -D -M wlp2s0
root 3655 0.0 0.0 6540 1940 pts/30 S Apr11 0:02 shncpd -D -M wlp2s0

Elsewhere I have it configured to serve addresses - and it is behaving normally. This is on a pi2:

root 3424 0.0 0.1 1984 1340 ? S 04:01 0:08 shncpd -E fd10::/48 -E 172.26.192.0/20 eth0 wlan0
root 3696 0.0 0.1 1984 1360 ? S 04:05 0:09 shncpd -D -E fd10::/48 -E 172.26.192.0/20 -M wlan0 eth0

I've set it up to run like this:

/usr/bin/shncpd -d 3 -p 8808 -m ff02::8808 -s /usr/lib/shncpd/shncpd-script.sh wlan0 eth0

In the logs I see:

HNCP_DEBUG_LEVEL=3
+ case "$1" in
+ '[' -n '' ']'

Using a DHCP client, DNS and NTP are configured.

Addresses are assigned and routing works.

The main router has OpenWRT CC.