#3455 closed defect (fixed)
OpenSolaris Kernel Dump
回報者: | drseek | 負責人: | |
---|---|---|---|
元件: | host support | 版本: | VirtualBox 2.1.4 |
關鍵字: | OpenSolaris Crash Dump | 副本: | |
Guest type: | Windows | Host type: | Solaris |
描述
I'm running VirtualBox 2.1.4 on OpenSolaris 101b. The guest OS is Windows XP SP3. Every time I shut down the guest OpenSolaris crashes. Maybe the cause is the same as described in ticket #2705 but I'm using an actual OpenSolaris and VirtualBox build.
The logs (/var/adm/messages and VirtualBox log) are appended to the ticket.
Here is the SCAT output of the crash:
user: (root:0) release: 5.11 (32-bit) version: snv_101b machine: i86pc node name: cracking system type: i86pc hostid: 87b92e06 dump_conflags: 0x10000 (DUMP_KERNEL) on /dev/dsk/c4d0s1(1.99G) time of crash: Sat Feb 28 12:14:53 CET 2009 age of system: 1 hours 5 minutes 49.06 seconds panic CPU: 1 (2 CPUs, 1022M memory) panic string: BAD TRAP: type=e (#pf Page fault) rp=e4ba8e60 addr=14 occurred in module "genunix" due to a NULL pointer dereference ==== checking for trap information ==== CPU 1 had the panic ==== panic thread: 0xd7b1a600 ==== CPU: 1 ==== ==== panic user (LWP_SYS) thread: 0xd7b1a600 PID: 570 on CPU: 1 ==== cmd: /usr/lib/fm/fmd/fmd fmri: svc:/system/fmd:default t_procp: 0xd9b639b0 p_as: 0xdbb94cb8 size: 35139584 RSS: 11706368 hat: 0xd246b0c8 cpuset: 1 zone: global t_stk: 0xe4ba8fac sp: 0xe4ba8d1c t_stkbase: 0xe4ba7000 t_pri: 59(TS) pctcpu: 0.026998 t_lwp: 0xd7bd9300 lwp_regs: 0xe4ba8fac mstate: LMS_SYSTEM ms_prev: LMS_KFAULT ms_state_start: 0.000020675 seconds earlier ms_start: 1 hours 4 minutes 29.373664131 seconds earlier psrset: 0 last CPU: 1 idle: 386646 ticks (1 hours 4 minutes 26.46 seconds) start: Sat Feb 28 11:10:24 2009 age: 3869 seconds (1 hours 4 minutes 29 seconds) syscall: #255 sys#255(, 0x0) (sysent: genunix:umount2+0x0) tstate: TS_ONPROC - thread is being run on a processor tflg: T_PANIC - thread initiated a system panic T_DFLTSTK - stack is default size tpflg: TP_TWAIT - wait to be freed by lwp_wait TP_MSACCT - collect micro-state accounting information tsched: TS_LOAD - thread is in memory TS_DONT_SWAP - thread/LWP should not be swapped pflag: SMSACCT - process is keeping micro-state accounting SMSFORK - child inherits micro-state accounting pc: unix:vpanic_common+0xab: addl $0x10,%esp unix:vpanic_common+0xab() unix:panic+0x12() unix:die+0x93() unix:trap+0x1422() unix:cmntrap_pushed+0x2f() -- panic trap data type: 0xe (Page fault) addr 0x14 rp 0xe4ba8e60 trapno 0xe (Page fault) err 0 (page not present,read,supervisor) %efl 0x10287 (carry|parity|negative|interrupt enable|resume) savbp 0xe4ba8f04 savip genunix:dnlc_purge_vfsp+0x94: movl 0x14(%ecx),%eax %ebp 0xe4ba8f04 %esp 0xe4ba8e98 %eip genunix:dnlc_purge_vfsp+0x94: movl 0x14(%ecx),%eax %eax 0xfec65a90 %ebx 0xd32b3748 %ecx 0 %edx 0 %edi 0xd3006361 %esi 0 %cs 0x158 (KCS_SEL) %ds 0xd3280160 (GDT unknown,KPL) %es 0xe4ba0160 (GDT unknown,KPL) %fs 0xe4ba0000 (GDT unknown,KPL) %gs 0x1b0 (KGS_SEL) <trap>genunix:dnlc_purge_vfsp+0x94() genunix:dounmount+0x16() genunix:umount2_engine+0x49() genunix:umount2+0x12d() unix:sys_sysenter+0x106 _sysenter_done() -- switch to user thread's user stack -- ==== analyzing panic thread stack for trap frames ==== no trap frames found
附加檔案 (3)
更動歷史 (12)
16 年 前 由 編輯
附檔: | 新增 messages.copy |
---|
comment:1 16 年 前 由 編輯
Have you tried using NAT instead of Host Interface networking to see if it prevents the crash?
Also could please you show output of ifconfig -a from the host?
跟進: 3 comment:2 16 年 前 由 編輯
I tried NAT but crashes again.
Here is the output of ifconfig -a:
lo0: flags=2001000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4,VIRTUAL> mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 vfe0: flags=201000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4,CoS> mtu 1500 index 2 inet 10.0.0.2 netmask ff000000 broadcast 10.255.255.255 ether 0:b:6a:9c:70:f9 sppp0: flags=10010008d1<UP,POINTOPOINT,RUNNING,NOARP,MULTICAST,IPv4,FIXEDMTU> mtu 1492 index 3 inet 92.229.129.40 --> 213.191.64.143 netmask ff000000 ether 0:0:0:0:0:0 lo0: flags=2002000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv6,VIRTUAL> mtu 8252 index 1 inet6 ::1/128
At the moment the Windows guest doesn't have access to the internet.
跟進: 4 comment:3 16 年 前 由 編輯
Replying to drseek:
I tried NAT but crashes again.
Could you please attach log file of crash with NAT, if possible core file.
# ulimit -c unlimited # VirtualBox -startvm [name of your VM]
comment:4 16 年 前 由 編輯
Could you please attach log file of crash with NAT, if possible core file.
If core file available we would provide information where it could be uploaded (please contact me vasily.levchenko [at] sun [dot] com)
comment:5 16 年 前 由 編輯
I attached the log file of crash.
Before the crash occured I leaved the full screen mode of the guest, switch to firefox on the host and browse variouse webpages and suddenly the system crashes. At that time the guest was running in the background.
comment:6 16 年 前 由 編輯
For submitter - could you turn on kmem_flags in /etc/system please:
set kmem_flags=0xf
regenerate your boot archive and then reboot.
Also, if you can supply the actual crash dump files (or a pointer to them) from your host that would be handy.
Finally, are you using a fixed size backing store, or expanding? (Not that I think it matters, but it's always better to have a full picture of the configuration).
/var/adm/messages