On Sat, Nov 30, 2019 at 01:55:40AM +0100, Klemens Nanni wrote: > OpenBSD 6.6-current (GENERIC.MP) #494: Thu Nov 28 13:24:18 MST 2019 > [email protected]:/usr/src/sys/arch/amd64/compile/GENERIC.MP > > Sitting in the hut with above snapshot, no unwind.conf(5). > unwind crashed yesterday, I just found the core dump in /var/crash/; > with DEBUG=-g build of today, I get the following trace:
Unfortunately that trace doesn't make a lot of sense. For starters there is no path from new_frowarders to try_next_resolver. I don't think we can mix and match core dumps with binaries build later. I do recall that I got lucky once or twice but that's a long time ago. > > (gdb) bt > #0 0x00001f31d2bea877 in try_next_resolver (rq=0x1f34b254f600) at > /s/sbin/unwind/resolver.c:752 > #1 0x00001f31d2be8d14 in new_forwarders (oppdot=0) at > /s/sbin/unwind/resolver.c:1029 > #2 resolver_dispatch_frontend (fd=<optimized out>, event=<optimized out>, > bula=0x1f348e451000) at /s/sbin/unwind/resolver.c:500 > #3 0x00001f31d2c887ff in gettime (base=<optimized out>, tp=<optimized out>) > at /usr/src/lib/libevent/event.c:93 > #4 event_base_loop (base=0x1f344be5f800, flags=1273363456) at > /usr/src/lib/libevent/event.c:478 > #5 0x00001f31d2be825f in resolver (debug=<optimized out>, verbose=<optimized > out>) at /s/sbin/unwind/resolver.c:373 > #6 0x00001f31d2bf0dd2 in config_clear (conf=0x1f31d2e5d728 <yyval+8>) at > /s/sbin/unwind/unwind.c:692 > #7 main_shutdown () at /s/sbin/unwind/unwind.c:312 > #8 0x00001f31d2be7486 in ___start () > #9 0x0000000000000000 in ?? () -- I'm not entirely sure you are real.
