Ken Smith wrote: >On Thu, Nov 04, 2004 at 03:34:26PM +0100, Michael Nottebrock wrote: > > >>On Thursday, 4. November 2004 07:31, Alex Dupre wrote: >> >> >>>Michael Nottebrock wrote: >>> >>> >>>>Could all people who are seeing this please post their kernel >>>>configurations, sysctl.conf and perhaps some system details (platform, >>>>UP/MP), too. >>>> >>>> >>Looking at all of these, I fail to see anything "special" that all these >>machines have in common - looks like there's a full range from tuned >>configurations with eyebrowraisers like PREEMPTION to almost-unmodified >>GENERIC and pretty different hardware, too. :( >> >>So if anybody spots something interesting here after all, or thinks Marc >>Ramirez backtraces / debug sessions reveal anything about the bug, please >>take over... >> >> if you can get to teh kernel debugger on a serial link, (there is a sysctl to drop yuo into it) then the appropriate lines of 'ps' would be interesting. also x/xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx {address} where address is the address of the ksegrp in question (in threaded processes) myght be interesting. (you can get that from the ps in ddb) >> >> > >In particular if you manage to wedge processes in an unkillable state >the output of "ps -o f -l" might be useful. I'd like to see all of the >flags set for a process that gets wedged. > > please also add the -H option to show the threads in question.Received on Thu Nov 04 2004 - 17:25:02 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:21 UTC