Cannot get kernel core dump
Hello! I'm trying to report a problem which involves kernel panic, but I cannot get a core dump. The system is 7.0-STABLE from June 18th. I've set the following in /etc/rc.conf (and of course rebooted, even several times): dumpdev=/dev/ad0s1b dumpdir=/home/crash The machine has 512 MB RAM and swap partition is also 512 MB, but according to dumpon(8) this shouldn't be a problem, because hw.physmem is still somewhat smaller than the swap partition: bsd# sysctl hw.physmem hw.physmem: 527863808 bsd# swapinfo Device 1K-blocks UsedAvail Capacity /dev/ad0s1b5242880 524288 0% /home partition, where I've created the crash directory, has 55 GB free space. When the machine boots up, it dutifully reports: kernel dumps on /dev/ad0s1b Then I do the thing that causes the panic. System prints the panic message and hangs. Only thing left to do is press the reset button. On next boot, the log says savecore: no dumps found When the system panics and prints the panic message, it doesn't say anything about saving the coredump. I haven't had a kernel panic for a long time, but I seem to remember that in the past there was some kind of message to that effect. Are there known conditions under which the core dump really isn't recorded, or am I simply missing something obvious? -- Toomas ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Cannot get kernel core dump
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Toomas Aas wrote: | Hello! | | I'm trying to report a problem which involves kernel panic, but I cannot get a | core dump. The system is 7.0-STABLE from June 18th. | | I've set the following in /etc/rc.conf (and of course rebooted, even several | times): | dumpdev=/dev/ad0s1b | dumpdir=/home/crash | | The machine has 512 MB RAM and swap partition is also 512 MB, but according to | dumpon(8) this shouldn't be a problem, because hw.physmem is still somewhat | smaller than the swap partition: | | bsd# sysctl hw.physmem | hw.physmem: 527863808 | | bsd# swapinfo | Device 1K-blocks UsedAvail Capacity | /dev/ad0s1b5242880 524288 0% | | /home partition, where I've created the crash directory, has 55 GB free space. | | When the machine boots up, it dutifully reports: | kernel dumps on /dev/ad0s1b | | Then I do the thing that causes the panic. System prints the panic message | and hangs. Only thing left to do is press the reset button. On next boot, the | log says | savecore: no dumps found | | When the system panics and prints the panic message, it doesn't say anything | about saving the coredump. I haven't had a kernel panic for a long time, but | I seem to remember that in the past there was some kind of message to that | effect. | | Are there known conditions under which the core dump really isn't recorded, or | am I simply missing something obvious? Yes, if the panic occurs before the filesystem supposed to contain the crash dumps is mounted. Is it the case? | | -- | Toomas | ___ | freebsd-questions@freebsd.org mailing list | http://lists.freebsd.org/mailman/listinfo/freebsd-questions | To unsubscribe, send any mail to [EMAIL PROTECTED] - -- Pietro Cerutti [EMAIL PROTECTED] PGP Public Key: http://gahr.ch/pgp -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.9 (FreeBSD) iEYEAREKAAYFAkhiVTgACgkQwMJqmJVx9454AwCgowIDacy0X33iBwbC0QRVCYjw zGcAnig8zjnlrq1njlwqO7pJOGIyOV3O =4wCh -END PGP SIGNATURE- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Panic when disconnecting Coolpix L5 (was: Cannot get kernel core dump)
Wednesday 25 June 2008 17:24:58 kirjutas Pietro Cerutti: Toomas Aas wrote: | Are there known conditions under which the core dump really isn't | recorded, or am I simply missing something obvious? Yes, if the panic occurs before the filesystem supposed to contain the crash dumps is mounted. Is it the case? No, the panic occurs when the system is in full multi-user mode and all filesystems are mounted, including /home where I have configured the crash dumps. Swap partition /dev/ad0s1b is also active. However, it just occurred to me that the system actually goes somewhat awry even before the panic occurs, so perhaps this is why the core dump is not recorded. The sequence of events is like this: 1. I attach a digital camera (Nikon Coolpix L5) via USB to my computer, which has Asrock K7S41GX motherboard, with onboard SiS 5571 USB controller. The camera is set to 'Mass storage' mode. 2. The system seems to recognize the camera with some success: Jun 25 16:35:48 bsd root: Unknown USB device: vendor 0x04b0 product 0x020d bus uhub1 Jun 25 16:35:48 bsd kernel: umass0: NIKON NIKON DSC COOLPIX L5, class 0/0, rev 1.10/1.00, addr 2 on uhub1 Jun 25 16:35:48 bsd kernel: da0 at umass-sim0 bus 0 target 0 lun 0 Jun 25 16:35:48 bsd kernel: da0: NIKON DSC COOLPIX L5 Removable Direct Access SCSI-2 device Jun 25 16:35:48 bsd kernel: da0: 1.000MB/s transfers Jun 25 16:35:48 bsd kernel: da0: 485MB (994304 512 byte sectors: 64H 32S/T 485C) However, at that point the machine pretty much stops responding. Clicking any icons in KDE has no effect. I can switch to another tty by pressing for example Alt+F3, but I cannot log in there (the login prompt appears but I can't type my login name). 3. I disconnect the camera and kernel panics. umass0: BBB reset failed, IOERROR umass0: BBB bulk-in clear stall failed, IOERROR umass0: at uhub1, port 2 (addr 2) disconnected (da0:umass-sim0:0:0:0): lost device (da0:dead-sim0):0:0:0): Synchronize cache failed, status == 0x39, scsi status == 0x0 Fatal trap 12: page fault while in kernel mode fault virtual address = 0x0 fault code = supervisor write, page not present instruction pointer = 0x20:0xc043fb4b stack pointer = 0x28:0xd4cd4adc frame pointer = 0x28:0xd4cd4af8 code segment = base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt ennabled, resume, IOPL = 0 current process = 2 (g_event) trap number = 12 panic: page fault Uptime: 1m 20s Instruction pointer and current process are always identical, I haven't compared all the other information. $ nm -n /boot/kernel/kernel | grep c043fb c043fb10 T xpt_done Note that I haven't mounted the camera before disconnecting, so it's probably not the 'well known problem' described on Jeremy Chadwick's wiki page. Any further ideas on how to debug such problem or which mailing list is more appropriate? Thanks a lot, -- Toomas ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
