I'm running Debian 2.2 on a RS/6000 f50, kernel 2.4.8.
I noticed that kswapd was marked zombie in the process list:
root 4 0.0 0.0 0 0 ? Z Sep13 55:27 [kswapd
<defunct>]
Why would it crash, and what can I do? I noticed nothing irregular
yet. The machine has been up since September 13 without problems.
Maybe I should go back to an older, stable kernel? The only problem
is the machine being at 200km distance from here :-(
Running ksymoops on the logfile gives this:
-----
Warning (compare_maps): ksyms_base symbol __VERSIONED_SYMBOL(rtc_lock)
not found in System.map. Ignoring ksyms_base entry
Warning (compare_maps): ksyms_base symbol matrox_G100_R__ver_matrox_G100
not found in System.map. Ignoring ksyms_base entry
Warning (compare_maps): ksyms_base symbol
nfsd_linkage_R__ver_nfsd_linkage not found in System.map. Ignoring
ksyms_base entry
Warning (compare_maps): mismatch on symbol xchg_u32 , ksyms_base says
c000b6d0, System.map says c0005ff0. Ignoring ksyms_base entry
Warning (compare_maps): mismatch on symbol lvm_name , lvm-mod says
e131dd84, /lib/modules/2.4.8/kernel/drivers/md/lvm-mod.o says e131e114.
Ignoring /lib/modules/2.4.8/kernel/drivers/md/lvm-mod.o entry
Warning (compare_maps): mismatch on symbol pde , lvm-mod says e131ddac,
/lib/modules/2.4.8/kernel/drivers/md/lvm-mod.o says e131e13c. Ignoring
/lib/modules/2.4.8/kernel/drivers/md/lvm-mod.o entry
Dec 4 00:04:42 www kernel: Machine check in kernel mode.
Dec 4 00:04:42 www kernel: Oops: machine check, sig: 7
Dec 4 00:04:42 www kernel: NIP: 00000300 XER: 20000000 LR: C00568D8 SP:
C04D7F60 REGS: c04d7eb0 TRAP: 0200
Using defaults from ksymoops -t elf32-powerpc -a powerpc:common
Dec 4 00:04:42 www kernel: MSR: 00081000 EE: 0 PR: 0 FP: 0 ME: 1 IR/DR:
00
Dec 4 00:04:42 www kernel: TASK = c04d6000[4] 'kswapd' Last syscall: -1
Dec 4 00:04:42 www kernel: last math cca8c000 last altivec 00000000
Dec 4 00:04:42 www kernel: GPR00: C0152104 C04D7F60 C04D6000 C04D7F68
000000C0 00000001 DFED98A0 00000000
Dec 4 00:04:42 www kernel: GPR08: C0471000 00000000 C04D7F68 001F0001
44848042 1002BC40 00000000 00000000
Dec 4 00:04:42 www kernel: GPR16: 00000000 00000000 00000000 00000000
003FF000 00000000 C0150000 00000001
Dec 4 00:04:42 www kernel: GPR24: 00000000 C0110000 C0110000 000000F4
C04D7F68 001F4B88 001F4B88 001F4B80
Dec 4 00:04:42 www kernel: Call backtrace:
Dec 4 00:04:42 www kernel: C00568D8 C0056918 C00341F8 C00342A0 C000640C
Warning (Oops_read): Code line not seen, dumping what data is available
>>NIP; 00000300 Before first symbol <=====
Trace; c00568d8 <prune_icache+150/168>
Trace; c0056918 <shrink_icache_memory+28/48>
Trace; c00341f8 <do_try_to_free_pages+3c/70>
Trace; c00342a0 <kswapd+74/fc>
Trace; c000640c <kernel_thread+34/40>
8 warnings issued. Results may not be reliable.
-----
Any help is very much appreciated.
Jens.
