[uml-devel] init_new_context_skas - new_mm failed, errno = -24

[uml-devel] init_new_context_skas - new_mm failed, errno = -24

[Christopher S. Aker]

2.6.22.6-um

init_new_context_skas - new_mm failed, errno = -24
init_new_context_skas - new_mm failed, errno = -24
init_new_context_skas - new_mm failed, errno = -24
Kernel panic - not syncing: switch_mm_skas - PTRACE_SWITCH_MM failed, 
errno = 9


EIP: 0000:[<00000000>] CPU: 0 Not tainted EFLAGS: 00000000
     Not tainted
EAX: 00000000 EBX: 00000000 ECX: 00000000 EDX: 00000000
ESI: 00000000 EDI: 00000000 EBP: 00000000 DS: 0000 ES: 0000
085dbdfc:  [<080818fb>] show_regs+0xeb/0x110
085dbe18:  [<0806b41c>] panic_exit+0x2c/0x50
085dbe28:  [<080973fe>] notifier_call_chain+0x3e/0x60
085dbe48:  [<080974dd>] __atomic_notifier_call_chain+0x2d/0x40
085dbe64:  [<0809751e>] atomic_notifier_call_chain+0x2e/0x40
085dbe80:  [<08087fd2>] panic+0x72/0x110
085dbe9c:  [<0807e85c>] switch_mm_skas+0x5c/0x60
085dbeb4:  [<084b98f4>] schedule+0x2b4/0x5a0
085dbf24:  [<0806837d>] default_idle+0x2d/0x30
085dbf38:  [<0806bc78>] init_idle_skas+0x28/0x30
085dbf48:  [<08068388>] cpu_idle+0x8/0x10
085dbf50:  [<084b92fb>] rest_init+0x5b/0x60
085dbf64:  [<0804889d>] start_kernel+0x1bd/0x220
085dbf84:  [<0806bcac>] start_kernel_proc+0x2c/0x40
085dbf90:  [<0807a190>] run_kernel_thread+0x50/0x60
085dbfcc:  [<0806b9fe>] new_thread_handler+0x7e/0xd0
085dbffc:  [<00000000>] _stext+0xf7fb7f4c/0xc

Is this cause by a still-too-low max_map_count value?

-Chris

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel

Re: [uml-devel] init_new_context_skas - new_mm failed, errno = -24

[Jeff Dike]

On Thu, Sep 06, 2007 at 11:03:52AM -0400, Christopher S. Aker wrote:
> 2.6.22.6-um
> 
> init_new_context_skas - new_mm failed, errno = -24
> init_new_context_skas - new_mm failed, errno = -24
> init_new_context_skas - new_mm failed, errno = -24
> Kernel panic - not syncing: switch_mm_skas - PTRACE_SWITCH_MM failed, 
> errno = 9

> Is this cause by a still-too-low max_map_count value?

Could be - what was the host's max_map_count?

				Jeff

-- 
Work email - jdike at linux dot intel dot com

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel

Re: [uml-devel] init_new_context_skas - new_mm failed, errno = -24

[Christopher S. Aker]

Jeff Dike wrote:
> On Thu, Sep 06, 2007 at 11:03:52AM -0400, Christopher S. Aker wrote:
>> 2.6.22.6-um
>>
>> init_new_context_skas - new_mm failed, errno = -24
>> init_new_context_skas - new_mm failed, errno = -24
>> init_new_context_skas - new_mm failed, errno = -24
>> Kernel panic - not syncing: switch_mm_skas - PTRACE_SWITCH_MM failed, 
>> errno = 9
> 
>> Is this cause by a still-too-low max_map_count value?
> 
> Could be - what was the host's max_map_count?
> 
> 				Jeff

root@atlanta20:~# cat /proc/sys/vm/max_map_count
500000
root@atlanta20:~# free -g
              total       used       free     shared    buffers     cached
Mem:            23          8         15          0          0          7
-/+ buffers/cache:          0         23
Swap:            1          0          1

-Chris



-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel

Re: [uml-devel] init_new_context_skas - new_mm failed, errno = -24

[Blaisorblade]

[-- Attachment #1.1: Type: text/plain, Size: 1948 bytes --]

On martedì 11 settembre 2007, Christopher S. Aker wrote:
> Jeff Dike wrote:
> > On Thu, Sep 06, 2007 at 11:03:52AM -0400, Christopher S. Aker wrote:
> >> 2.6.22.6-um
> >>
> >> init_new_context_skas - new_mm failed, errno = -24
> >> init_new_context_skas - new_mm failed, errno = -24
> >> init_new_context_skas - new_mm failed, errno = -24
> >> Kernel panic - not syncing: switch_mm_skas - PTRACE_SWITCH_MM failed,
> >> errno = 9
> >>
> >> Is this cause by a still-too-low max_map_count value?
> >
> > Could be - what was the host's max_map_count?
> >
> > 				Jeff

errno = 24 means EMFILE (too many open files).

And I read, in mm/mmap.c: do_mmap_pgoff():

        /* Too many mappings? */
        if (mm->map_count > sysctl_max_map_count)
                return -ENOMEM;

While open_proc_mm only returns 0 or -ENOMEM.

So I'm 99% sure that you have indeed too many open files. There is a 
differnence between EMFILE and ENFILE - I vaguely remember that the 2nd is 
about ulimit while the first is about the total number of fd's. (man proc 
should says how to increase it - it is under /proc/sys).

> root@atlanta20:~# cat /proc/sys/vm/max_map_count
> 500000
> root@atlanta20:~# free -g
>               total       used       free     shared    buffers     cached
> Mem:            23          8         15          0          0          7
> -/+ buffers/cache:          0         23
> Swap:            1          0          1

Btw, note this:
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2005.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
SourceForge has not implemented AdSense at all, it seems...
-- 
"Doh!" (cit.), I've made another mistake!
Paolo Giarrusso, aka Blaisorblade
Linux registered user n. 292729
http://www.user-mode-linux.org/~blaisorblade

[-- Attachment #1.2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

[-- Attachment #2: Type: text/plain, Size: 228 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

[-- Attachment #3: Type: text/plain, Size: 194 bytes --]

_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel

Re: [uml-devel] init_new_context_skas - new_mm failed, errno = -24

[Jeff Dike]

On Tue, Sep 11, 2007 at 04:00:58PM +0200, Blaisorblade wrote:
> errno = 24 means EMFILE (too many open files).

Whoops, I mentally turned that into -ENOMEM.

Maybe we should resurrect os_print_error as you suggested and start
using it.  The current habit of using raw errnos dates from very
early, when perror (or strerror - I don't remember which) gave me
undebuggable segfaults deep inside libc.  Probably, that's not a
problem any more.

				Jeff

-- 
Work email - jdike at linux dot intel dot com

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel

Re: [uml-devel] init_new_context_skas - new_mm failed, errno = -24

[Blaisorblade]

[-- Attachment #1.1: Type: text/plain, Size: 1077 bytes --]

On martedì 11 settembre 2007, Jeff Dike wrote:
> On Tue, Sep 11, 2007 at 04:00:58PM +0200, Blaisorblade wrote:
> > errno = 24 means EMFILE (too many open files).
>
> Whoops, I mentally turned that into -ENOMEM.
>
> Maybe we should resurrect os_print_error as you suggested and start
> using it.  The current habit of using raw errnos dates from very
> early, when perror (or strerror - I don't remember which) gave me
> undebuggable segfaults deep inside libc.  Probably, that's not a
> problem any more.

I surely agree. I also think I had introduced os_print_error and exactly for 
this purpose, even if I can't recall well.

strerror() is not reentrant so it can't be used, but there is strerror_r.

There is no strsignal_r (using strsignal(3) would clear messages as "Kernel 
mode signal 8"), but I'd suggest to follow the unrecommended but documented 
practice of using sys_siglist[signr].
-- 
"Doh!" (cit.), I've made another mistake!
Paolo Giarrusso, aka Blaisorblade
Linux registered user n. 292729
http://www.user-mode-linux.org/~blaisorblade

[-- Attachment #1.2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

[-- Attachment #2: Type: text/plain, Size: 228 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

[-- Attachment #3: Type: text/plain, Size: 194 bytes --]

_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel

Re: [uml-devel] init_new_context_skas - new_mm failed, errno = -24

[Jeff Dike]

On Tue, Sep 11, 2007 at 04:00:58PM +0200, Blaisorblade wrote:
> So I'm 99% sure that you have indeed too many open files. 

And this could happen if you have a ton of processes inside the UML,
each with its own /proc/mm descriptor.

				Jeff

-- 
Work email - jdike at linux dot intel dot com

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel