On 4 August 2017 at 13:05, Jesper Wallin <[email protected]> wrote: > > On Fri, Aug 04, 2017 at 10:47:31AM +0100, Stuart Henderson wrote: > > To clarify: you ran syspatch, and after fetching the patches (but still > > on the old kernel) you attempted to reboot, and it stopped responding > > at that point - is that correct? > > Correct. I applied all patches and attempted to reboot into the new > kernel. > > > Which kernel was it running previously? > > OpenBSD 6.1 (GENERIC.MP) #8: Tue Jun 27 08:50:26 CEST 2017 > > > If it happens again, post the panic message - if you don't have > > anything logging the serial at the time you can repeat it with "show > > panic". Also the process listing ("ps" in DDB) - this occurred when > > a process was exiting and it would be useful to know which one. > > Yeah, the plan was of course to run the ddb commands suggested on > https://www.openbsd.org/ddb.html, but I was unable to after it froze. > > My bad for not starting with "show panic" though. Taking notes. :-) > > > Information about config of the machine might help too. Which PF > > features are you using? Any source tracking? What daemons are you > > using? (ftp-proxy? relayd? etc.) > > The machine acts as our gateway running the default daemons plus > unbound, dhcpd, relayd and sshd. PF is using nat, rdr-to, normalization, > queueing (both queue and prio) and keep/modulate state. I have added > pf.conf.txt, relayd.conf.txt and ifconfig.txt to the directory > previously linked. >
Another thing to try would be this change that didn't make it to stable. http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys/net/pf.c.diff?r1=1.1034&r2=1.1035
