[Swan] pluto lockup in syslog() ?

Paul Wouters pwouters at redhat.com
Fri Mar 1 17:49:40 EET 2013


I just experienced a lock up calling syslog()

pluto had completely frozen. gdb showed:

(gdb) bt
#0  0x00007fbd349a69bb in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#1  0x00007fbd34994e54 in ?? () from /lib/x86_64-linux-gnu/libc.so.6
#2  0x00007fbd34994844 in __vsyslog_chk () from /lib/x86_64-linux-gnu/libc.so.6
#3  0x00007fbd34994cdf in syslog () from /lib/x86_64-linux-gnu/libc.so.6
#4  0x000000000041c3d9 in libreswan_DBG_log (message=0x4dd1f0 "state: %ld requesting event %s to be deleted by %s:%d")
     at /root/libreswan-3.1dr2/programs/pluto/log.c:758
#5  0x000000000042a096 in _delete_dpd_event (st=0x2946870, file=0x4d9e70 "/root/libreswan-3.1dr2/programs/pluto/state.c",
     lineno=427) at /root/libreswan-3.1dr2/programs/pluto/timer.c:745
#6  0x000000000041d48a in delete_state (st=0x2946870) at /root/libreswan-3.1dr2/programs/pluto/state.c:427
#7  0x000000000043b271 in accept_delete (st=0x2946870, md=0x26ee130, p=0x26ee338)
     at /root/libreswan-3.1dr2/programs/pluto/ikev1_main.c:2931
#8  0x0000000000432152 in process_packet_tail (mdp=0x733e48) at /root/libreswan-3.1dr2/programs/pluto/ikev1.c:2032
#9  0x0000000000431217 in process_v1_packet (mdp=0x733e48) at /root/libreswan-3.1dr2/programs/pluto/ikev1.c:1622
#10 0x0000000000466426 in process_packet (mdp=0x733e48) at /root/libreswan-3.1dr2/programs/pluto/demux.c:171
#11 0x00000000004664b3 in comm_handle (ifp=0x26e0340) at /root/libreswan-3.1dr2/programs/pluto/demux.c:225
#12 0x00000000004272c1 in call_server () at /root/libreswan-3.1dr2/programs/pluto/server.c:787
#13 0x0000000000423c33 in main (argc=4, argv=0x7fffe5dd53e8) at /root/libreswan-3.1dr2/programs/pluto/plutomain.c:1262
(gdb)

Apart from merging in the sd_notify patches and having systemd restart
pluto when its heart beat failed, is there anything we could have
done? Should have done?

Paul


More information about the Swan mailing list