Re: kernel panic | |
| [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] | |
Hi Mrunal, ----8<---- > > Is this problem a common one? Do I need to compile my kernel to provide a > stack trace on console or is it configured by default? I am using a Open > SUSE 10.1 kernel 2.6.18. ----8<----- Some points from my experience; might give you some hints: 1. I might be wrong, but openSUSE 10.1 comes with kdump support. it can help you generate the kdump and save a core file at a prescribed location. (man kdump should help). I am not sure whether this option is enabled as default, but probably you try finding it in "zcat /proc/config.gz" whether this is available, before trying other methods required a kernel recompile. 2. Also, even I have faced such crashed when nothing was getting printed in /var/log/messages. That is because syslog is a userspace utility which accesses /proc/kmsg and prints the log buffer. It may not be able to do so in case of server crashes (like interrupts or deadlock). In such cases, /var/log is unreliable. 3. This has generally worked for me - try putting sleep(s) in the control path for module initialization (something like sleep_on_timeout for 2 seconds) and loads of printks telling you what is happening in your kernel path. This way atleast you can narrow down where/why the problem exists. Sleeps would give syslog time to print something. Sincerely hope this helps. If problem already solved, apologies for the noice. -- Shreyansh -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
[Newbies FAQ] [Kernel List] [Site Home] [IETF Annouce] [Git] [Networking] [Security] [Bugtraq] [Rubini] [Photo] [Yosemite] [MIPS Linux] [ARM Linux] [Linux Security] [Linux Networking] [Linux RAID] [Linux SCSI] [Linux ACPI] [DDR & Rambus] [UNIX Filesystems] [Linux Resources]
![]() |
![]() |