IBM P55A主机宕机dump分析

一客户P550每周六晚上22点后都会出现宕机现象，需要分析原因：
1、检查日志，发现每次宕机都生成了dump日志
p550a:/#sysdumpdev -L
0453-039
Device name:         /dev/hd6
Major device number: 10
Minor device number: 2
Size:                93209600 bytes
Uncompressed Size:   835558845 bytes
Date/Time:           Thu Jul 22 01:17:44 BEIST 2010
Dump status:         0
dump completed successfully
Dump copy filename: /var/adm/ras/vmcore.1.Z
2、分析dump日志
p550a:/tmp/ibmsupt#kdb vmcore.0 /unix
The specified kernel file is a 64-bit kernel
vmcore.0 mapped from @ 700000000000000 to @ 70000003401245f
Preserving 1317350 bytes of symbol table
First symbol __mulh
Component Names:
 1)  minidump [2 entries]
 2)  dmp_minimal [9 entries]
 3)  proc [345 entries]
 4)  thrd [2387 entries]
 5)  rasct [1 entries]
 6)  ldr [2 entries]
 7)  errlg [3 entries]
 8)  mtrc [26 entries]
 9)  lfs [2 entries]
10)  bos [2 entries]
11)  ipc [7 entries]
12)  vmm [13 entries]
13)  alloc_kheap [512 entries]
14)  alloc_other [21 entries]
15)  rtastrc [8 entries]
16)  sisraid [4 entries]
17)  aixpcm [9 entries]
18)  scdisk [19 entries]
19)  lvm [2 entries]
20)  jfs2 [1 entries]
21)  tty [4 entries]
22)  netstat [10 entries]
23)  goent_dd [7 entries]
24)  dump_failures [1 entries]
25)  dump_statistics [1 entries]
Component Dump Table has 3398 entries
           START              END
0000000000001000 0000000003BBA050 start+000FD8
F00000002FF47600 F00000002FFDC920 __ublock+000000
000000002FF22FF4 000000002FF22FF8 environ+000000
000000002FF22FF8 000000002FF22FFC errno+000000
F100070F00000000 F100070F10000000 pvproc+000000
F100070F10000000 F100070F18000000 pvthread+000000
PFT:
PVT:
id....................0002
raddr.....0000000000686000 eaddr.....F200800030000000
size..............00040000 align.............00001000
valid..1 ros....0 fixlmb.1 seg....0 wimg...2
[kdb_read_mem] no real storage @ F1000000107145D8
Dump analysis on CHRP_SMP_PCI POWER_PC POWER_5 machine with 8 available CPU(s)  (64-bit registers)
Processing symbol table...
.......................done
[kdb_read_mem] no real storage @ F1000000106C145B
        ERROR: Unable to acess nfs_syms
Unable to initialize module: /usr/lib/ras/autoload/nfs64.kdb
(2)> stat
SYSTEM_CONFIGURATION:
CHRP_SMP_PCI POWER_PC POWER_5 machine with 8 available CPU(s)  (64-bit registers)
SYSTEM STATUS:
sysname... AIX
nodename.. p550a
release... 3
version... 5
build date Jan 10 2006
build time 10:56:32
label..... 0602A_53E
machine... 000B27ACD600
nid....... 0B27ACD6
time of crash: Thu Jul  8 00:53:14 2010
age of system: 95 day, 19 hr., 57 min., 14 sec.
xmalloc debug: disabled
CRASH INFORMATION:
CPU 2 CSA 018BDE00 at time of crash, error code for LEDs: 30000000
pvthread+000C00 STACK:
[00075FEC]v_delpft+000108 (F200800020000008 [??])
[0010AA88]v_relframe+000464 (??, ??, ??)
[001027E4]v_pageout+0006D0 (??, ??, ??)
[00141A20]v_steal+00043C (??, ??, ??, ??)
[00144EF4]v_fblru_scan+0003B8 (??)
[001403D4]v_lru+00035C (??)  
[001414D0]v_memp_lru+00023C (??)
[00207FEC]v_prememp_lru+000020 (??)
[002A2474].backt+000080 ()   
____ Exception (F000000030017780) ____
iar   : 00000000002A23F4  msr   : 80000000000010B2  cr    : 42000024
lr    : 00000000001408D4  ctr   : 0000000000000000  xer   : 00000000
mq    : 00000000  asr   : 00000000F372A001
r0  : 0000000000207FCC  r1  : 0FFFFFFFF4017E90  r2  : 0000000001491C28
r3  : 0000000000000000  r4  : F10001002CBA1100  r5  : 0000000003B90000
r6  : 0000000000000000  r7  : 0000000000000000  r8  : 0000000000000106
r9  : 0000000000000000  r10 : 00000000001408D4  r11 : F000000030017780
r12 : 80000000000010B2  r13 : F10001002CB82400  r14 : 00000000DEADBEEF
r15 : 000000000101A9C0  r16 : 00000000DEADBEEF  r17 : 00000000DEADBEEF
r18 : 00000000DEADBEEF  r19 : 00000000DEADBEEF  r20 : 00000000DEADBEEF
r21 : 00000000DEADBEEF  r22 : 00000000DEADBEEF  r23 : 00000000DEADBEEF
r24 : 00000000DEADBEEF  r25 : 00000000DEADBEEF  r26 : 00000000DEADBEEF
r27 : 00000000DEADBEEF  r28 : 00000000DEADBEEF  r29 : 00000000DEADBEEF
r30 : 0000000003B90000  r31 : 0000000000000000
                             
prev      0000000000000000 stackfix  0000000000000000 int_ticks 00
kjmpbuf   0000000000000000 excbranch 0000000000000000 no_pfault 00
intpri    0B backt     00 flags     00
fpscr     0000000000000000 fpscrx    00000000 fpowner   00
fpeu      00 fpinfo    00 alloc     F000
o_iar     0000000000000000 o_toc     0000000000000000
o_arg1    0000000000000000 o_vaddr   0000000000000000
krlockp   0000000000000000   
Except :
 csr   0000000000000000 dsisr 0000000000000000
 esid  0000000000000000 dar   0000000000000000 dsirr 0000000000000106
[002A23F4].backt+000000 ()
[kdb_get_memory] no real storage @ FFFFFFFF4017EA0
(2)>
(2)> status
CPU     TID  TSLOT     PID  PSLOT  PROC_NAME
  0     2005      2    2004      2  wait
  1    12025     18    D01A     13  wait
  2     C019     12    4008      4  lrud
  3    1502B     21    F01E     15  wait
  4      12D  32768     120  16384  wait
  5     3133  32771    3126  16387  wait
  6     E01D     14    600C      6  psmd
  7     5137  32773    512A  16389  wait
  8-63   Disabled
这里分析，宕机时有进程一直在做页面交换操作。