Re: kernel BUG with 2.4.27-om20041102-tab on SMP when doing heavy I/O

Peter Cordes <[email protected]>
Newsgroups gmane.linux.cluster.openmosix.devel
Message-ID <[email protected]>
On Tue, Sep 27, 2005 at 03:13:32PM -0300, Peter Cordes wrote:
>  I have been running  2.4.27-om20041102-tab (peter-OPrYhD/TZC/[email protected])
> (gcc version 3.3.5 (Debian 1:3.3.5-13)) #1 SMP Wed Sep 7  on my dual Opteron
> cluster.  The master node has 4GB of RAM, and the other 7 nodes have 2GB.
> All with gigabit ethernet (tg3).  The master has an Intel e100 built in to the
> mobo too, which is used to connect to the outside world.

 It seems that channel bonding didn't change things after all (or at least
didn't completely avoid the problem).  I had two more BUG messages
yesterday, which come from sshd.  From my other logs, I see that there were
ssh logins at the times of those two BUGs.  (One of them was my login, and I
didn't notice anything strange, so it didn't actually break sshd!)  I
happened to have sshd's DEBUG logging turned on, and I don't see anything
odd in it, even.

Oct  3 19:06:49 nyneve kernel: kernel BUG at page_alloc.c:274!
Oct  3 19:06:49 nyneve kernel: invalid operand: 0000
Oct  3 19:06:49 nyneve kernel: CPU:    0
Oct  3 19:06:49 nyneve kernel: EIP:    0010:[rmqueue+534/576]    Not tainted
Oct  3 19:06:49 nyneve kernel: EFLAGS: 00010202
Oct  3 19:06:49 nyneve kernel: eax: 0200084c   ebx: c33643d8   ecx: 00038000   edx: 000a1ca3
Oct  3 19:06:49 nyneve kernel: esi: c03c9758   edi: c03c9794   ebp: c03c9758   esp: ebe15e40
Oct  3 19:06:49 nyneve kernel: ds: 0018   es: 0018   ss: 0018
Oct  3 19:06:49 nyneve kernel: Process sshd (pid: 13039, stackpage=ebe15000)
Oct  3 19:06:49 nyneve kernel: Stack: 00038000 c471bfa8 00069ca3 00000282 00000000 c03c9758 c03c9758 c03c9858 
Oct  3 19:06:49 nyneve kernel:        00000002 00000001 c01484b5 00000119 f79961d4 00000128 f7793dc0 f7793e84 
Oct  3 19:06:49 nyneve kernel:        f6b8bb40 c03c9758 c03c9854 00000000 000001d2 c013b2f0 db19c440 00000000 
Oct  3 19:06:49 nyneve kernel: Call Trace:    [__alloc_pages+101/688] [do_no_page+128/592] [do_wp_page+117/736] [handle_mm_fault+271/304] [do_page_fault+872/1289] [sock_sendmsg+104/160] [sock_write+158/192] [local_pre_usermode_actions+21/192] [do_page_fault+0/1289] [error_code+52/60]
Oct  3 19:06:49 nyneve kernel: Code: 0f 0b 12 01 34 e2 37 c0 8b 43 18 a8 80 74 08 0f 0b 14 01 34 


>>ebx; c33643d8 <_end+2ee5bec/386e1894>
>>esi; c03c9758 <contig_page_data+1b8/3b8>
>>edi; c03c9794 <contig_page_data+1f4/3b8>
>>ebp; c03c9758 <contig_page_data+1b8/3b8>
>>esp; ebe15e40 <_end+2b997654/386e1894>

Code;  00000000 Before first symbol
00000000 <_EIP>:
Code;  00000000 Before first symbol
   0:   0f 0b                     ud2a   
Code;  00000002 Before first symbol
   2:   12 01                     adc    (%ecx),%al
Code;  00000004 Before first symbol
   4:   34 e2                     xor    $0xe2,%al
Code;  00000006 Before first symbol
   6:   37                        aaa    
Code;  00000007 Before first symbol
   7:   c0 8b 43 18 a8 80 74      rorb   $0x74,0x80a81843(%ebx)
Code;  0000000e Before first symbol
   e:   08 0f                     or     %cl,(%edi)
Code;  00000010 Before first symbol
  10:   0b 14 01                  or     (%ecx,%eax,1),%edx
Code;  00000013 Before first symbol
  13:   34 00                     xor    $0x0,%al

Oct  3 21:43:51 nyneve kernel:  kernel BUG at page_alloc.c:274!
Oct  3 21:43:51 nyneve kernel: invalid operand: 0000
Oct  3 21:43:51 nyneve kernel: CPU:    0
Oct  3 21:43:51 nyneve kernel: EIP:    0010:[rmqueue+534/576]    Not tainted
Oct  3 21:43:51 nyneve kernel: EFLAGS: 00010202
Oct  3 21:43:51 nyneve kernel: eax: 02000850   ebx: c37a5a48   ecx: 00038000   edx: 000b53e5
Oct  3 21:43:51 nyneve kernel: esi: c03c9758   edi: c03c9794   ebp: c03c9758   esp: f4cc1e40
Oct  3 21:43:51 nyneve kernel: ds: 0018   es: 0018   ss: 0018
Oct  3 21:43:51 nyneve kernel: Process sshd (pid: 14942, stackpage=f4cc1000)
Oct  3 21:43:51 nyneve kernel: Stack: 00038000 c471bfa8 0007d3e5 00000282 00000000 c03c9758 c03c9758 c03c9858 
Oct  3 21:43:51 nyneve kernel:        00000002 00000001 c01484b5 00000119 f79961d4 00000128 f7793dc0 f7793e84 
Oct  3 21:43:51 nyneve kernel:        f6b8bb40 c03c9758 c03c9854 00000000 000001d2 c013b2f0 f56cc840 00000000 
Oct  3 21:43:51 nyneve kernel: Call Trace:    [__alloc_pages+101/688] [do_no_page+128/592] [do_wp_page+117/736] [handle_mm_fault+271/304] [do_page_fault+872/1289] [sock_sendmsg+104/160] [sock_write+158/192] [local_pre_usermode_actions+21/192] [do_page_fault+0/1289] [error_code+52/60]
Oct  3 21:43:51 nyneve kernel: Code: 0f 0b 12 01 34 e2 37 c0 8b 43 18 a8 80 74 08 0f 0b 14 01 34 


>>ebx; c37a5a48 <_end+332725c/386e1894>
>>esi; c03c9758 <contig_page_data+1b8/3b8>
>>edi; c03c9794 <contig_page_data+1f4/3b8>
>>ebp; c03c9758 <contig_page_data+1b8/3b8>
>>esp; f4cc1e40 <_end+34843654/386e1894>

Code;  00000000 Before first symbol
00000000 <_EIP>:
Code;  00000000 Before first symbol
   0:   0f 0b                     ud2a   
Code;  00000002 Before first symbol
   2:   12 01                     adc    (%ecx),%al
Code;  00000004 Before first symbol
   4:   34 e2                     xor    $0xe2,%al
Code;  00000006 Before first symbol
   6:   37                        aaa    
Code;  00000007 Before first symbol
   7:   c0 8b 43 18 a8 80 74      rorb   $0x74,0x80a81843(%ebx)
Code;  0000000e Before first symbol
   e:   08 0f                     or     %cl,(%edi)
Code;  00000010 Before first symbol
  10:   0b 14 01                  or     (%ecx,%eax,1),%edx
Code;  00000013 Before first symbol
  13:   34 00                     xor    $0x0,%al


 My guess is that a some data structure in kernel space is corrupt.  If the
memtester results are any indication (see my other message in
openmosix-general), something is corrupting memory, be it hardware or
software.

-- 
#define X(x,y) x##y
Peter Cordes ;  e-mail: X(peter@cor , des.ca)

"The gods confound the man who first found out how to distinguish the hours!
 Confound him, too, who in this place set up a sundial, to cut and hack
 my day so wretchedly into small pieces!" -- Plautus, 200 BC


-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.