Swapper crash / Kernels Out Of Memoy(OOM) killer Problem ?
Hi.
Why I wanted to force OOM killer to send SIGTERM is that -
When my application receives SIGKILL from OOM killer, the application (and it's related threads) are getting killed. But after some time the swapper process is getting crashed and system goes for a reboot. This I feel is due to the cleanup handler (signal handler) not being called when the application receives SIGKILL. Could you please through some light on this issue.
Below is the stack dump from /var/log/messages -
CPU: 0
EIP: 0010:[<d086da5a>] Tainted: PF
EFLAGS: 00010246
eax: 0000007c ebx: 00000042 ecx: 00000000 edx: 0000007c
esi: d0849579 edi: c0377e5b ebp: c0377dcc esp: c0377da0
ds: 0018 es: 0018 ss: 0018
Process swapper (pid: 0, stackpage=c0377000)
Stack: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 0000007c
00427e10 35000080 00000042 c0377e2c d086db60 00000035 00000000 00000000
0000007c 00000042 cfe29800 c0377e08 c0377e0c 00000001 000000a0 05000000
Call Trace: [<d086db60>] [<d0849560>] [<d0871850>] [<d0874830>] [<d0874830>
Code: 8b 40 08 3b 42 10 0f 94 c0 0f b6 c0 89 45 e0 e9 92 00 00 00
<0>Kernel panic: Aiee, killing interrupt handler!In interrupt handler - not syncing
Quoted 103 lines
-----Original Message-----
From: vinay [mailto:[email protected]]
Sent: Tuesday, August 9, 2005 09:23 AM
To: 'Xavier Roche', [email protected]
Subject: Re: Kernels Out Of Memoy(OOM) killer Problem ?Hello Xavier.
Thanks for replying.
I checked that the
/proc/sys/vm/overcommit_memoryis already set to 0.In my case the problem is that I don't have many options like changing
the
overcommit_memoryetc.Only thing I need to do is, have a proper cleanup will exiting the application. As the application is receiving SIKILL from OMM killer the required signal handler is not getting called and no cleanup is happening.
So could you please suggest me that what could be done in this regard.
Thanks and Regards
Vinay.
Quoted 52 lines
-----Original Message-----
From: Xavier Roche [mailto:[email protected]]
Sent: Tuesday, August 9, 2005 08:06 AM
Subject: Re: Kernels Out Of Memoy(OOM) killer Problem ?
Quoted 3 lines · vinay
I have a problem with linux kernel's Out Of Memory (OOM) killer.
I wanted to know, is there any way that we can force OOM killer to send a signal other than SIGKILL to kill a process when ever OOM detects a system memory crunch.
As far as I understand the kernel, oom is called when the system has no
memory pages left, and MUST get one to continue normal (ie. kernel)
processing. The kernel just do not have the time to execute some
user-space code, it MUST get free pages where they are (and hence, kill
immediately some innocent process).
This condition should not occur without using overcommit. Are you sure
you are not using overcommit ? (cat
/proc/sys/vm/overcommit_memory)To dasable it:
echo 0 >
/proc/sys/vm/overcommit_memoryOvercommit is quite dangerous on production systems, because it leads to
oom kills on heavy loads (at least, this is what I experienced).
-
Quoted 9 lines
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Hello,
1) What is your kernel version?
2) It is tainted; Why?
3) Make sure your app/kernel combo works with SIGKILL. If it doesn't, then you have a serious bug.
cheers,
Masoud
Quoted 151 lines · vinay
Hi.
Why I wanted to force OOM killer to send SIGTERM is that -
When my application receives SIGKILL from OOM killer, the application (and it's related threads) are getting killed. But after some time the swapper process is getting crashed and system goes for a reboot. This I feel is due to the cleanup handler (signal handler) not being called when the application receives SIGKILL. Could you please through some light on this issue. Below is the stack dump from/var/log/messages-CPU: 0 EIP: 0010:[<d086da5a>] Tainted: PF EFLAGS: 00010246 eax: 0000007c ebx: 00000042 ecx: 00000000 edx: 0000007c esi: d0849579 edi: c0377e5b ebp: c0377dcc esp: c0377da0 ds: 0018 es: 0018 ss: 0018 Process swapper (pid: 0, stackpage=c0377000) Stack: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 0000007c 00427e10 35000080 00000042 c0377e2c d086db60 00000035 00000000 00000000 0000007c 00000042 cfe29800 c0377e08 c0377e0c 00000001 000000a0 05000000 Call Trace: [<d086db60>] [<d0849560>] [<d0871850>] [<d0874830>] [<d0874830> Code: 8b 40 08 3b 42 10 0f 94 c0 0f b6 c0 89 45 e0 e9 92 00 00 00<0>Kernel panic: Aiee, killing interrupt handler!
In interrupt handler - not syncingQuoted 105 lines
-----Original Message-----
From: vinay [mailto:[email protected]]
Sent: Tuesday, August 9, 2005 09:23 AM
To: 'Xavier Roche', [email protected]
Subject: Re: Kernels Out Of Memoy(OOM) killer Problem ?Hello Xavier.
Thanks for replying.
I checked that the
/proc/sys/vm/overcommit_memoryis already set to 0.In my case the problem is that I don't have many options like changing
the
overcommit_memoryetc.Only thing I need to do is, have a proper cleanup will exiting the application. As the application is receiving SIKILL from OMM killer the required signal handler is not getting called and no cleanup is happening.
So could you please suggest me that what could be done in this regard.
Thanks and Regards
Vinay.
Quoted 69 lines
-----Original Message-----
From: Xavier Roche [mailto:[email protected]]
Sent: Tuesday, August 9, 2005 08:06 AM
Subject: Re: Kernels Out Of Memoy(OOM) killer Problem ?
Quoted 6 lines · vinay
I have a problem with linux kernel's Out Of Memory (OOM) killer.
I wanted to know, is there any way that we can force OOM killer to send a signal other than SIGKILL to kill a process when ever OOM detects a system memory crunch.
As far as I understand the kernel, oom is called when the system has no
memory pages left, and MUST get one to continue normal (ie. kernel)
processing. The kernel just do not have the time to execute some
user-space code, it MUST get free pages where they are (and hence, kill
immediately some innocent process).
This condition should not occur without using overcommit. Are you sure
you are not using overcommit ? (cat
/proc/sys/vm/overcommit_memory)To dasable it:
echo 0 >
/proc/sys/vm/overcommit_memoryOvercommit is quite dangerous on production systems, because it leads to
oom kills on heavy loads (at least, this is what I experienced).
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 11 lines
Quoted 3 lines
Please read the FAQ at http://www.tux.org/lkml/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/