Re: Need advice on ORA-04030 and pga_aggregate_target parameter

On Tue, 15 May 2007 12:25:45 -0700, harvinder76 wrote:

> free
> total used free shared buffers
> cached
> Mem: 4147184 1474236 2672948 0 22936 1295784
> -/+ buffers/cache: 155516 3991668 Swap: 4620280 0
> 4620280
>
> On May 15, 3:06 pm, Mladen Gogala <mgogala.SPAM..._at_not-at-verizon.net>
> wrote:

>> On Tue, 15 May 2007 11:35:42 -0700, harvinder76 wrote:
>> > Thanks for the info, ulimit looks good and nothing is in the alert
>>
>> How about the swap partition? What do free or swapon -s say?
>> --http://www.mladen-gogala.com

Harvinder, it's hard to know what's going on without debugging. One way of finding out what is going on would be to connect to a dedicated server turn on strace like this: strace -o /tmp/proc.out -p <pid> -etrace=brk

That will leave output in /tmp/proc.out. Here is an example of a procedure, but without the error:
strace -o /tmp/proc.out -p 3650 -etrace=open

This has produced a trace file, tracing all calls to "open". The output file looks like this:

open("/oradata/file_dir/oracle/system01.dbf", O_RDWR|O_SYNC|O_DIRECT| O_LARGEFILE) = 9
open("/oradata/file_dir/oracle/users01.dbf", O_RDWR|O_SYNC|O_DIRECT| O_LARGEFILE) = 10
open("/oracle/product/10g/admin/oracle/udump/10g_ora_3650.trc", O_WRONLY| O_CREAT|O_TRUNC|O_LARGEFILE, 0660) = 5
open("/dev/null", O_RDONLY) = 5 open("/oracle/product/10g/admin/oracle/udump/10g_ora_3650_TEST.trc", O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE, 0660) = 5 open("/dev/shm/ora_10G_1540109", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory)

You don't have a problem with the "open" call, you have a problem with the "brk" call. You can get the list of system calls by through yelp (Manuals->Developemnt->System Calls) I tried with the "brk" but the process didn't use it, even once.

-- 
http://www.mladen-gogala.com