Hello! What attitude should be taken about the following? fatal trap 12, page fault while in kernel mode wp_lock=00000002 cpuid=0 lapic.id=00000000 fault virtualaddr=0x154 fault code=supervisor read, page not present instruction pointer=0x8>0xc0271e58 stack pointer=0x10:0xc6607be8 frame pointer=0x10:0xc6607bf0 code segment=base 0x0, limit 0xfffff, type 0x1b =DPL 0, pres 1, def32 1, gram 1 processor eflags=interrupt enabled, resume, IDPL=0 current process=369 (smbd) interrupt mask=bio <- SMP:XXX trap number=12 panic: page fault mp_lock=00000002, cpuid=0, lapic.id=00000000 boot() called on cpu#0 Using samba 2.0.10, FreeBSD 4.4 Multiprocessor Kernel. Is this my problem or Samba's problem and is this fixed in further Sambas, i.e. 2.2 or so? Yours thankfully, Indrek Zolk, Estonia Estonian Literary Museum
On Thursday 25 October 2001 13:22, Indrek Zolk wrote:> Hello! > > What attitude should be taken about the following? > > fatal trap 12, page fault while in kernel mode > wp_lock=00000002 cpuid=0 lapic.id=00000000 > fault virtualaddr=0x154 > fault code=supervisor read, page not present > instruction pointer=0x8>0xc0271e58 > stack pointer=0x10:0xc6607be8 > frame pointer=0x10:0xc6607bf0 > code segment=base 0x0, limit 0xfffff, type 0x1b > =DPL 0, pres 1, def32 1, gram 1 > processor eflags=interrupt enabled, resume, IDPL=0 > current process=369 (smbd) > interrupt mask=bio <- SMP:XXX > trap number=12 > panic: page fault > mp_lock=00000002, cpuid=0, lapic.id=00000000 > boot() called on cpu#0 > > Using samba 2.0.10, FreeBSD 4.4 Multiprocessor Kernel. Is this my problem > or Samba's problem and is this fixed in further Sambas, i.e. 2.2 or so?Could be a lot of things. Did you happen to catch a crash dump from the kernel? That would be very useful to pinpointing the problem. If you don't have a crash dump, follow the FreeBSD docs to set up a debug kernel and capture crash dumps, then look into it and find out where exactly the problem came from. -- Bill Moran Potential Technology technical services (412) 793-4257
I don't know here are some more clues. the wp_lock = 000000002 seems to be to be a semaphore so you may have a race condition. Nevertheless the fault occurred in the kernel under the smbd process with trap 12, which is probably a fault DPL=0 means that the segment descriptors priledge was 0 so this did not fault while in a system call ( DPL = 3 ), but I believe the DPL numbers are the same for BSD and Linux, I don't know if they are the same. So it's probably not in a system call the interrupt enabled flag tells me that it was not in an interrupt handler. So it's likely it's in BSD's equivalent of bottom halves or Tasklets. Aah the fault virtualaddr = 0x154 indicates it was trying to write to userspace that had been swapped out maybe. Check your smbfs module ( or bsd equivalent ) is bug free which is should be. It likely was writing data from the network to the user space of the smbd process Tyler -------------- next part -------------- HTML attachment scrubbed and removed