[Xen-users] Force dom0 to use specific cpu?

["Jonathon Jones" <doobla@xxxxxxxxxx>]

I am trying to troubleshoot an odd problem I am having and my main question is this:  How can I force dom0 to use cpu #1 not cpu #0?  So far I have only seen that setting the flag dom0-cpus allows you to control how many cpus are used, but not which ones specifically.  Please point me in the right direction (file and settings).

 

Also, here is the symptom of my problem in case any of you can help with this.  1 week ago I started having a vps randomly about every day or day and a half stop responding to apache requests as well as some other services but ssh would still run fine.  I would reboot the vps using ssh and all was good for another day or two.  Tonight this happened in the dom0 and both the dom0 and that vps were unresponsive at that point to any services.  In the domU I was able to load the hosting control panel but when I tried to login it too stopped responding.

 

Here are samples of the log entries that happen when this problem shows up.

 

From the dom0 failure:

 

May 21 21:57:21 balder007 kernel:  ------------[ cut here ]------------
May 21 21:57:21 balder007 kernel: kernel BUG at <bad filename>:63723!
May 21 21:57:21 balder007 kernel: invalid operand: 0000 [#9]
May 21 21:57:21 balder007 kernel: SMP
May 21 21:57:21 balder007 kernel: Modules linked in: ipt_physdev iptable_filter ip_tables bridge ipv6 autofs4 sunrpc af_packet binfmt_misc quota_v2 dm_mirror dm_mod video thermal processor fan button battery ac md i2c_nforce2 i2c_core shpchp pci_hotplug tg3 sata_sil unix ext3 jbd sata_via sd_mod sata_nv libata scsi_mod
May 21 21:57:21 balder007 kernel: CPU:    1
May 21 21:57:21 balder007 kernel: EIP:    0061:[<c01182b6>]    Tainted: GF     VLI
May 21 21:57:21 balder007 kernel: EFLAGS: 00010282   (2.6.12.6-xen)
May 21 21:57:21 balder007 kernel: EIP is at pgd_ctor+0x26/0x30
May 21 21:57:21 balder007 kernel: eax: fffffff4   ebx: 00000001   ecx: f577e040   edx: 00000000
May 21 21:57:21 balder007 kernel: esi: c066de00   edi: ec6e6818   ebp: ec6e6800   esp: d77eddc4
May 21 21:57:21 balder007 kernel: ds: 007b   es: 007b   ss: 0069
May 21 21:57:21 balder007 kernel: Process anacron (pid: 6349, threadinfo=d77ec000 task=ed712020)
May 21 21:57:21 balder007 kernel: Stack: d7008000 00000000 00000020 c014dcc1 d7008000 c066de00 00000001 ec6e6800
May 21 21:57:21 balder007 kernel:        d7008000 c066de00 00000000 c014de3d c066de00 ec6e6800 00000001 000000d0
May 21 21:57:21 balder007 kernel:        c066de60 00000001 000000d0 c066b600 0000000c 000000d0 00000000 c014e04b
May 21 21:57:21 balder007 kernel: Call Trace:
May 21 21:57:21 balder007 kernel:  [<c014dcc1>] cache_init_objs+0x71/0x80
May 21 21:57:21 balder007 kernel:  [<c014de3d>] cache_grow+0x10d/0x1a0
May 21 21:57:21 balder007 kernel:  [<c014e04b>] cache_alloc_refill+0x17b/0x220
May 21 21:57:22 balder007 kernel:  [<c014e30f>] kmem_cache_alloc+0x7f/0x90
May 21 21:57:22 balder007 kernel:  [<c011833d>] pgd_alloc+0x1d/0x310
May 21 21:57:22 balder007 kernel:  [<c011c840>] activate_task+0x90/0xb0
May 21 21:57:22 balder007 kernel:  [<c01216fe>] mm_init+0xce/0x100
May 21 21:57:22 balder007 kernel:  [<c0121766>] mm_alloc+0x36/0x50
May 21 21:57:22 balder007 kernel:  [<c017410c>] do_execve+0x7c/0x270
May 21 21:57:22 balder007 kernel:  [<c0109066>] sys_execve+0x46/0xa0
May 21 21:57:22 balder007 kernel:  [<c010a65d>] syscall_call+0x7/0xb
May 21 21:57:22 balder007 kernel: Code: 00 f3 ab 5f c3 83 ec 0c b8 20 00 00 00 89 44 24 08 31 c0 89 44 24 04 8b 44 24 10 89 04 24 e8 d2 2b 00 00 85 c0 75 04 83 c4 0c c3 <0f> 0b eb f8 8d b6 00 00 00 00 83 ec 08 b8 f8 1e 36 c0 89 5c 24

Fron the domU failure:

 

May 21 22:32:57 secure kernel: ------------[ cut here ]------------
May 21 22:32:57 secure kernel: kernel BUG at <bad filename>:63723!
May 21 22:32:57 secure kernel: invalid operand: 0000 [#1]
May 21 22:32:57 secure kernel: SMP
May 21 22:32:57 secure kernel: Modules linked in: ipv6 ipt_TOS iptable_mangle ip_conntrack_ftp ip_conntrack_irc ipt_REJECT ipt_LOG ipt_limit iptable_filter ipt_multiport ipt_state ip_conntrack ip_tables i2c_dev i2c_core af_packet binfmt_misc quota_v2 ext3 jbd dm_mod unix
May 21 22:32:57 secure kernel: CPU:    0
May 21 22:32:57 secure kernel: EIP:    0061:[<c01182b6>]    Tainted: GF     VLI
May 21 22:32:57 secure kernel: EFLAGS: 00210282   (2.6.12.6-xen)
May 21 22:32:57 secure kernel: EIP is at pgd_ctor+0x26/0x30
May 21 22:32:57 secure kernel: eax: fffffff4   ebx: 00000001   ecx: f577e000   edx: 00000000
May 21 22:32:57 secure kernel: esi: c0626e00   edi: eb01f6d8   ebp: eb01f6c0   esp: e9489dc4
May 21 22:32:57 secure kernel: ds: 007b   es: 007b   ss: 0069
May 21 22:32:57 secure kernel: Process httpd (pid: 32157, threadinfo=e9488000 task=ea19c060)
May 21 22:32:57 secure kernel: Stack: e9b2d000 00000000 00000020 c014dcc1 e9b2d000 c0626e00 00000001 eb01f6c0
May 21 22:32:57 secure kernel:        e9b2d000 c0626e00 00000000 c014de3d c0626e00 eb01f6c0 00000001 000000d0
May 21 22:32:57 secure kernel:        c0626e60 00000001 000000d0 c0624980 0000000c 000000d0 00000000 c014e04b
May 21 22:32:57 secure kernel: Call Trace:
May 21 22:32:57 secure kernel:  [<c014dcc1>] cache_init_objs+0x71/0x80
May 21 22:32:57 secure kernel:  [<c014de3d>] cache_grow+0x10d/0x1a0
May 21 22:32:57 secure kernel:  [<c014e04b>] cache_alloc_refill+0x17b/0x220
May 21 22:32:57 secure kernel:  [<c014e30f>] kmem_cache_alloc+0x7f/0x90
May 21 22:32:57 secure kernel:  [<c011833d>] pgd_alloc+0x1d/0x310
May 21 22:32:57 secure kernel:  [<c01809c3>] dput+0x33/0x1d0
May 21 22:32:57 secure kernel:  [<c01216fe>] mm_init+0xce/0x100
May 21 22:32:57 secure kernel:  [<c0121766>] mm_alloc+0x36/0x50
May 21 22:32:57 secure kernel:  [<c017410c>] do_execve+0x7c/0x270
May 21 22:32:57 secure kernel:  [<c0109066>] sys_execve+0x46/0xa0
May 21 22:32:57 secure kernel:  [<c010a65d>] syscall_call+0x7/0xb
May 21 22:32:57 secure kernel: Code: 00 f3 ab 5f c3 83 ec 0c b8 20 00 00 00 89 44 24 08 31 c0 89 44 24 04 8b 44 24 10 89 04 24 e8 d2 2b 00 00 85 c0 75 04 83 c4 0c c3 <0f> 0b eb f8 8d b6 00 00 00 00 83 ec 08 b8 f8 1e 36 c0 89 5c 24

 

Any ideas at all?  I feel over my head here....

 

Thanks,

 

Jon

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users