"On 2018-12-16, Michael Pagano <thepagano...@gmail.com> wrote: > Hello, > > I seem to be having a difficulty with DDB. I am intending to send in a > bug report for a kernel panic I'm having, but unfortunately, I am unable > to acquire the proper details necessary to send in a meaningful report > because of DDB hanging when this kernel panic occurs. Is there a > way to obtain this information (i. e., traceback and ps output) on the > next boot so I could make said report? > > Thank you very much in advance. >
Does "sysctl machdep.forceukbd=1" help? If you have "ddb.console=1" (which must be set at boot e.g. via sysctl.conf - you can't do that later at runtime) you can test entry without a panic by pressing ctrl-alt-esc. If not, set ddb.panic=0, it will print some things, attempt to save a kernel core to disk, and reboot - if you're watching when it happens hopefully you can take a photo whilr it's writing to disk." Unfortunately, all "sysctl machdep forceukbd=1" does is cause the terminal to be completely unusable. It acts as though I'm holding down the enter key, and there is nothing I can do about it except to manually reboot. "ddb.console=1" was something I found to not work whatsoever. Control-alt-escape did nothing, even though I did assign it using sysctl.conf and I attempted to use it on the next boot. I did find "ddb.panic=0" to be useful to get a trace. However, it would not dump the kernel to disk and I'm still missing the ps output. Thank you very much for your input. If you have any further advice, please notify me.
