HPUX Error: 12: Not enough space

Paul Lavoie

unread,

Feb 10, 2000, 3:00:00 AM2/10/00

to

Platform
HPUX 11.0 HP9000 series K 2CPU 160Mhz 750Mb RAM 3.5Gb swap

At peak periods when approx. 700 processes running, swapinfo
reveals RAM 100% SWAPSPACE 25-50 % USED.

Entire Virtual and Real never at 100% combined.

What else could be causing this error??

Any help greatly appreciated.

* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!

Stéphane Lachapelle

unread,

Feb 10, 2000, 3:00:00 AM2/10/00

to

Hi,

Did you call swapinfo with option -t to report TOTAL. Sometimes, there is
reserved but unused swap.

The second thing is to ensure there is no unswappable programm running. The
sticky bit is used to mark these programms (-rwxr-xr-t the "t").

Stef

Paul Lavoie wrote in message
<111c3cc8...@usw-ex0107-051.remarq.com>...

John Pezzano

unread,

Feb 10, 2000, 3:00:00 AM2/10/00

to

Paul Lavoie wrote:
>
> Platform
> HPUX 11.0 HP9000 series K 2CPU 160Mhz 750Mb RAM 3.5Gb swap
>
> At peak periods when approx. 700 processes running, swapinfo
> reveals RAM 100% SWAPSPACE 25-50 % USED.
>
> Entire Virtual and Real never at 100% combined.
>
> What else could be causing this error??
>
> Any help greatly appreciated.
>
> * Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
> The fastest and easiest way to search and participate in Usenet - Free!

I am not positive but swapinfo might not tell you about swap that can't
be used because kernel parameter maxswapchunks is set too low. The man
page might give that info. Also read the help information in the SAM
section for changing this kernel parameter.

root

unread,

Feb 12, 2000, 3:00:00 AM2/12/00

to

Paul Lavoie wrote:
>
> Platform
> HPUX 11.0 HP9000 series K 2CPU 160Mhz 750Mb RAM 3.5Gb swap
>
> At peak periods when approx. 700 processes running, swapinfo
> reveals RAM 100% SWAPSPACE 25-50 % USED.
>
> Entire Virtual and Real never at 100% combined.
>
> What else could be causing this error??
>
> Any help greatly appreciated.
>
> * Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
> The fastest and easiest way to search and participate in Usenet - Free!

Is this an Oracle error??
If so, you might be running out of semaphores. Check your kernel parms
for max semaphores.

Regs,

Ian O'D

ECStahl

unread,

Feb 13, 2000, 3:00:00 AM2/13/00

to

Paul,

Checking swchunk/maxswapchunk is a good first step.

Was there anything else displayed with the error eg "shmctl, errno 12 Not
enough space"?

Check kernel parameters, and usage of resources related to kernel parameters eg
shm*, maxdsiz...

Are you using memory windows? If yes or no, how much of the related window,
global or otherwise, is in use? Window fragmentation issues? Appropriate VM
patching in place?

Determine what program is returning that error, what memory it is currently
utilizing, what memory it is requesting, and what type eg lockable, shared
memory, memory mapped file? Use Glance, and check with the program developer.

Regards,
Eric Stahl

Paul Lavoie paul.lavo...@nrc.ca.invalid wrote:
>HPUX Error: 12: Not enough space
...

ThomasW154

unread,

Mar 8, 2000, 3:00:00 AM3/8/00

to

I got something similar and increased the value of SHMMAX to 0x40000000 in the
kernel paramters and this seemed to work

ECStahl

unread,

Mar 9, 2000, 3:00:00 AM3/9/00

to

Rebooting your system, not increasing shmmax, resolved the errno 12 ie freed
the resources that led to the error.

Eric Stahl

thoma...@aol.com (ThomasW154) wrote:
Re - HPUX Error: 12: Not enough space