smp hangs with 2.2.19 kernel

smp hangs with 2.2.19 kernel

[Andrea Dell'Amico]

Hallo, I have a problem with a dual processor machine. With Red Hat
2.2.19 kernel, but with every 2.2.1x kernel, from Red Hat or self
compiled, the machine hangs with lot of processes in D state:

vmstat output is

procs       memory    swap          io     system         cpu
 r  b  w swpd free  buff cache si  so    bi    bo   in    cs  us  sy  id
 1 225 2   0  10632  10936  16456   0   0  6   3   14     9   6   3  14


The memory situation:

[root@petra petra]# free
             total       used       free     shared    buffers
cached
Mem:        523800     504688      19112          0      10620
31352
-/+ buffers/cache:     462716      61084
Swap:      2097136      35164    2061972


The box has 5 scsi disks with a aic7xxx controller. 4 disks are in RAID
1, the fifth is only used for the swap partition (2 GB).

When the machine hangs there are no logs.

Is there a way I can debug the cause? 

Thanks in advance,
andrea


-- 
Andrea Dell'Amico
<adellam@link.it> - Link s.r.l. <http://www.link.it>

Re: smp hangs with 2.2.19 kernel

[kees]

Hi,

I'm too chasing this one. I have only one IDE drive and an SCSI controller
which is mostly idle but sometimes used for a scanner. At first I thought
is was hardware but I'm going to believe it is a software problem. 
I replaced most of the hardware even the motherboard.
2.2.19 with PIII-667 Mhz/128MB MSI 594D mobo.

Kees

On 17 May 2001, Andrea Dell'Amico wrote:

> 
> Hallo, I have a problem with a dual processor machine. With Red Hat
> 2.2.19 kernel, but with every 2.2.1x kernel, from Red Hat or self
> compiled, the machine hangs with lot of processes in D state:
> 
> vmstat output is
> 
> procs       memory    swap          io     system         cpu
>  r  b  w swpd free  buff cache si  so    bi    bo   in    cs  us  sy  id
>  1 225 2   0  10632  10936  16456   0   0  6   3   14     9   6   3  14
> 
> 
> The memory situation:
> 
> [root@petra petra]# free
>              total       used       free     shared    buffers
> cached
> Mem:        523800     504688      19112          0      10620
> 31352
> -/+ buffers/cache:     462716      61084
> Swap:      2097136      35164    2061972
> 
> 
> The box has 5 scsi disks with a aic7xxx controller. 4 disks are in RAID
> 1, the fifth is only used for the swap partition (2 GB).
> 
> When the machine hangs there are no logs.
> 
> Is there a way I can debug the cause? 
> 
> Thanks in advance,
> andrea
> 
> 
> -- 
> Andrea Dell'Amico
> <adellam@link.it> - Link s.r.l. <http://www.link.it>
> 
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
> 

Re: smp hangs with 2.2.19 kernel

[kees]

Hi,

I have solid hangs (no log) with SMP machine for some time. I swapped
almost all of the hardware (incl mobo).
Dual PIII 677 MHZ on MSI 694D mobo. I had the hangs with 2.2.19 AND now
also with 2.4.4. I linked the kdb in the kernel and when the lockup
occurred I was able the get some info (serial console).

This is the kdb output for linux-2.4.4

[1]kdb> rd
eax = 0x00000001 ebx = 0xc0c57bc0 ecx = 0x00000020 edx = 0x00000001 
esi = 0xc03de880 edi = 0x00000004 esp = 0xc2137ed0 eip = 0xc0108b6a 
ebp = 0xc2137edc xss = 0x00000018 xcs = 0x00000010 eflags = 0x00000002 
xds = 0xc1220018 xes = 0xc03d0018 origeax = 0x00000001 &regs = 0xc2137e9c
[1]kdb> 

[1]kdb> cpu
Currently on cpu 1
Available cpus: 0*, 1
[1]kdb> 

[1]kdb> bt
    EBP       EIP         Function(args)
0xc2137edc 0xc0108b6a handle_IRQ_event+0x2e (0x4, 0xc2137f10, 0xc0c57bc0, 0xc2a98005, 0x0)
                               kernel .text 0xc0100000 0xc0108b3c 0xc0108bc4
           0xc0108d9c do_IRQ+0xa8
                               kernel .text 0xc0100000 0xc0108cf4 0xc0108df0
           0xc0106e10 ret_from_intr
                               kernel .text 0xc0100000 0xc0106e10 0xc0106e30
Interrupt registers:
eax = 0x0000006b ebx = 0xc2a98005 ecx = 0xc2137fa0 edx = 0xc2a98001 
esi = 0x00000000 edi = 0xc12cc7e0 esp = 0xc2137f44 eip = 0xc013f183 
ebp = 0xc2137f68 xss = 0x00000018 xcs = 0x00000010 eflags = 0x00000216 
xds = 0xc2a90018 xes = 0xc2a90018 origeax = 0xffffff04 &regs = 0xc2137f10
           0xc013f183 path_walk+0x24f (0xc2a98006, 0xc2137fa0)
                               kernel .text 0xc0100000 0xc013ef34 0xc013f7f0
0xc2137f84 0xc013fcd0 __user_walk+0x3c (0x805c808, 0x9, 0xc2137fa0)
                               kernel .text 0xc0100000 0xc013fc94 0xc013fcf0
0xc2137fbc 0xc013c33b sys_newstat+0x17 (0x805c808, 0xbfffed28, 0x805c808, 0xbfffee34, 0x805c808)
                               kernel .text 0xc0100000 0xc013c324 0xc013c394
           0xc0106d37 system_call+0x37
                               kernel .text 0xc0100000 0xc0106d00 0xc0106d40
[1]kdb> 

What can I do next? I have the bt for all processes in a file.

regards

Kees