Home » Infrastructure » Unix » CPU in Solaris 10 achieved 100% with Oracle Database (Sun Solaris 10, Oracle Database 10.2.0.1)
CPU in Solaris 10 achieved 100% with Oracle Database [message #386241] Fri, 13 February 2009 04:32 Go to next message
trantuananh24hg
Messages: 744
Registered: January 2007
Location: Ha Noi, Viet Nam
Senior Member
Hi all!

I've got one problem with CPU in Solaris 10.
Only one database, nothing except it running in Sun Sparc V890, 4 CPU, 16gb memory.

The SGA_TARGET parameter was set to 10gb, pga=2gb.
sys@VNP> show parameter sga

NAME                                 TYPE                             VALUE
------------------------------------ -------------------------------- ------
--------------------
lock_sga                             boolean                          FALSE
pre_page_sga                         boolean                          FALSE
sga_max_size                         big integer                      10G
sga_target                           big integer                      10G
sys@VNP> show parameter pga

NAME                                 TYPE                             VALUE
------------------------------------ -------------------------------- ------
--------------------
pga_aggregate_target                 big integer                      2G
sys@VNP>


The shm-max-memory kernel was set to 12gb
sys@VNP> ! cat /etc/system
*ident  "@(#)system     1.18    97/06/27 SMI" /* SVR4 1.5 */
*
* SYSTEM SPECIFICATION FILE
*

* moddir:
*
....................
set shmsys:shminfo_shmmax=12884901888
set shmsys:shminfo_shmmin=1
set shmsys:shminfo_shmseg=10
set shmsys:shminfo_shmmni=100
set semsys:seminfo_semmni=800
set semsys:seminfo_semmsl=256
set semsys:seminfo_semmns=204800
set noexec_user_stack=1

sys@VNP>


Of course, I limit shm-max-memory in project for oracle
sys@VNP> ! cat /etc/project
system:0::::
user.root:1::::
noproject:2::::
default:3::::
group.staff:10::::
oracle:100::::project.max-shm-memory=(priv,12884901888,deny)

sys@VNP> ! cat /etc/user_attr
#
# Copyright (c) 2003 by Sun Microsystems, Inc. All rights reserved.
#
# /etc/user_attr
#
# user attributes. see user_attr(4)
#
#pragma ident   "@(#)user_attr 1.1      03/07/09 SMI"
#
adm::::profiles=Log Management
lp::::profiles=Printer Management
root::::auths=solaris.*,solaris.grant;profiles=Web Console Management,All;lock_a
fter_retries=no
oracle::::project=oracle

sys@VNP> ! prctl -n project.max-shm-memory -i project oracle
project: 100: oracle
NAME    PRIVILEGE       VALUE    FLAG   ACTION                       RECIPIENT
project.max-shm-memory
        privileged      12.0GB      -   deny                                 -
        system          16.0EB    max   deny                                 -

sql@VNP> ! prctl -i project oracle
project: 100: oracle
NAME    PRIVILEGE       VALUE    FLAG   ACTION                       RECIPIENT
project.max-contracts
        privileged      10.0K       -   deny                                 -
        system          2.15G     max   deny                                 -
project.max-device-locked-memory
        privileged      1002MB      -   deny                                 -
        system          16.0EB    max   deny                                 -
project.max-port-ids
        privileged      8.19K       -   deny                                 -
        system          65.5K     max   deny                                 -
project.max-shm-memory
        privileged      12.0GB      -   deny                                 -
        system          16.0EB    max   deny                                 -
project.max-shm-ids
        privileged        128       -   deny                                 -
        system          16.8M     max   deny                                 -
project.max-msg-ids
        privileged        128       -   deny                                 -
        system          16.8M     max   deny                                 -
project.max-sem-ids
        privileged        800       -   deny                                 -
        system          16.8M     max   deny                                 -
project.max-crypto-memory
        privileged      3.92GB      -   deny                                 -
        system          16.0EB    max   deny                                 -
project.max-tasks
        system          2.15G     max   deny                                 -
project.max-lwps
        system          2.15G     max   deny                                 -
project.cpu-shares
        privileged          1       -   none                                 -
        system          65.5K     max   none                                 -
zone.max-lwps
        system          2.15G     max   deny                                 -
zone.cpu-shares
        privileged          1       -   none                                 -


Within Oracle DB, I set the Database Resource Manager

-- Detail Resource Manager
system@VNP> @C:\DBA_Scripts\Resource_Manager\rsrc_plan_status_detail_9i.sql

                    # of Plan  CPU       Active Sess               Parallel Degr
ee                Queueing                            Run
Resource Plan Name  Directives Method    Pool Method               Limit Method
                  Method        Status     Mandatory? Status
------------------- ---------- --------- ------------------------- -------------
----------------- ------------- ---------- ---------- -----------
CCBS_PLAN                    4 EMPHASIS  ACTIVE_SESS_POOL_ABSOLUTE PARALLEL_DEGR
EE_LIMIT_ABSOLUTE FIFO_TIMEOUT             NO         Running
NIGHT_WKND                   4 EMPHASIS  ACTIVE_SESS_POOL_ABSOLUTE PARALLEL_DEGR
EE_LIMIT_ABSOLUTE FIFO_TIMEOUT             NO         Not Running
DAYTIME                      4 EMPHASIS  ACTIVE_SESS_POOL_ABSOLUTE PARALLEL_DEGR
EE_LIMIT_ABSOLUTE FIFO_TIMEOUT             NO         Not Running
INTERNAL_PLAN                1 EMPHASIS  ACTIVE_SESS_POOL_ABSOLUTE PARALLEL_DEGR
EE_LIMIT_ABSOLUTE FIFO_TIMEOUT             YES        Not Running
SYSTEM_PLAN                  3 EMPHASIS  ACTIVE_SESS_POOL_ABSOLUTE PARALLEL_DEGR
EE_LIMIT_ABSOLUTE FIFO_TIMEOUT             NO         Not Running
INTERNAL_QUIESCE             2 EMPHASIS  ACTIVE_SESS_POOL_ABSOLUTE PARALLEL_DEGR
EE_LIMIT_ABSOLUTE FIFO_TIMEOUT             YES        Not Running

6 rows selected.


-- Summary Resource Manager
system@VNP> @C:\DBA_Scripts\Resource_Manager\rsrc_plan_status_summary_9i.sql

                                                                       # of Plan
                        Run
Resource Plan Name  Comments                                           Directive
s Status     Mandatory? Status
------------------- -------------------------------------------------- ---------
- ---------- ---------- -----------
CCBS_PLAN           Simple plan
4            NO         Running
NIGHT_WKND          Plan for nights and weekend processing activities
4            NO         Not Running
DAYTIME             Plan for daytime processing activities
4            NO         Not Running
INTERNAL_PLAN       Default Plan
1            YES        Not Running
SYSTEM_PLAN         Plan to give system sessions priority
3            NO         Not Running
INTERNAL_QUIESCE    Plan to internally quiesce system
2            YES        Not Running

6 rows selected.

system@VNP>


With ulimit, I set by default
sys@VNP> ! ulimit -a
time(seconds)        unlimited
file(blocks)         unlimited
data(kbytes)         unlimited
stack(kbytes)        8192
coredump(blocks)     unlimited
nofiles(descriptors) 15000
vmemory(kbytes)      unlimited

sys@VNP>


However, whenever some client retrieve data from report which was running by Oracle Application Server, both of OAS & Database 's CPU achieved to 100%. The reports statements, of course, was tuning again and again, if I ran them from sqlplus, they'd taken about 0.03% CPU per statement.

May you guide me how to limit CPU usage for oracle?
The purpose just I want to set Oracle Database will achieve only 70%-75% total CPU.

Thank you very much!

[Updated on: Fri, 13 February 2009 04:41]

Report message to a moderator

Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #386262 is a reply to message #386241] Fri, 13 February 2009 06:26 Go to previous messageGo to next message
Mahesh Rajendran
Messages: 10707
Registered: March 2002
Location: oracleDocoVille
Senior Member
Account Moderator
Database and OAS hosted in same server?
Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #386299 is a reply to message #386262] Fri, 13 February 2009 08:52 Go to previous messageGo to next message
trantuananh24hg
Messages: 744
Registered: January 2007
Location: Ha Noi, Viet Nam
Senior Member
Thank you Mahesh Rajendran, for your reply!

Database and OAS are not in same server.
I brief them as following:
- Database in Sun Sparc V890, Solaris 10, 4 CPUs, 16gb Memory, SAN Storage 1TB, Raid 1 and Single Instance.
- 5 OAS separate instances with OC4J in 5 servers running Windows Server 2003, Linux RedHat 4 - HP Proliant ML 350, 4 CPUs, 8gb Memory per server machine.

First time, when clients retrieve data for report monthly through application in OAS, the process & virtual memory in OAS increased more and more. When them (OAS machine server's CPU) achieved 100%, the Database Machine Server's CPU will increase concurrently to 100%.

I'll give to you the log file (in both of OAS and Database) as soon as possible.

By any way, thank you!

[Updated on: Fri, 13 February 2009 08:54]

Report message to a moderator

Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #386301 is a reply to message #386299] Fri, 13 February 2009 09:05 Go to previous messageGo to next message
Mahesh Rajendran
Messages: 10707
Registered: March 2002
Location: oracleDocoVille
Senior Member
Account Moderator
This is a well known issue with Oracle Application servers.
Check metalink. There are too many cases with too many scenarios.
Best to open an SR.
Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #386346 is a reply to message #386301] Fri, 13 February 2009 21:03 Go to previous messageGo to next message
trantuananh24hg
Messages: 744
Registered: January 2007
Location: Ha Noi, Viet Nam
Senior Member
Thank you, Mahesh Rajendran!
Yesterday, we tested the report module with ~ 100 concurrent sessions in Database, ~ 750-820 concurrent session cached in OAS. And, they're .. death, both of Database, AS & Machine. In Database, I've got

$ cat /var/adm/messages
.............
Feb 13 14:14:17 database /usr/dt/bin/ttsession[11485]: [ID 676140 daemon.error] gettransient(): 0
Feb 13 14:14:27 database telnetd[11569]: [ID 970222 daemon.error] libpkcs11: Unable to contact kcfd: Bad file number
Feb 13 14:14:27 database telnetd[11569]: [ID 523743 daemon.error] libpkcs11: /usr/lib/security/pkcs11_softtoken.so unexpected failure in ELF signature verification. System may have been tampered with. Cannot continue parsing /etc/crypto/pkcs11.conf
Feb 13 14:15:08 database telnetd[11788]: [ID 970222 daemon.error] libpkcs11: Unable to contact kcfd: Bad file number
Feb 13 14:15:08 database telnetd[11788]: [ID 523743 daemon.error] libpkcs11: /usr/lib/security/pkcs11_softtoken.so unexpected failure in ELF signature verification. System may have been tampered with. Cannot continue parsing /etc/crypto/pkcs11.conf
Feb 13 14:21:59 database /usr/dt/bin/ttsession[12742]: [ID 304705 daemon.error] Error: could not contact rpcbind
Feb 13 14:21:59 database /usr/dt/bin/ttsession[12742]: [ID 676140 daemon.error] gettransient(): 0
Feb 13 14:26:40 database nfs: [ID 455708 kern.warning] WARNING: lockd: cannot contact statd (I/O error), continuing
Feb 13 14:31:43 database telnetd[13162]: [ID 970222 daemon.error] libpkcs11: Unable to contact kcfd: Bad file number
Feb 13 14:31:43 database telnetd[13162]: [ID 523743 daemon.error] libpkcs11: /usr/lib/security/pkcs11_softtoken.so unexpected failure in ELF signature verification. System may have been tampered with. Cannot continue parsing /etc/crypto/pkcs11.conf
Feb 13 14:31:47 database telnetd[13185]: [ID 970222 daemon.error] libpkcs11: Unable to contact kcfd: Bad file number
Feb 13 14:31:47 database telnetd[13185]: [ID 523743 daemon.error] libpkcs11: /usr/lib/security/pkcs11_softtoken.so unexpected failure in ELF signature verification. System may have been tampered with. Cannot continue parsing /etc/crypto/pkcs11.conf
Feb 13 14:31:51 database telnetd[13195]: [ID 970222 daemon.error] libpkcs11: Unable to contact kcfd: Bad file number
Feb 13 14:31:51 database telnetd[13195]: [ID 523743 daemon.error] libpkcs11: /usr/lib/security/pkcs11_softtoken.so unexpected failure in ELF signature verification. System may have been tampered with. Cannot continue parsing /etc/crypto/pkcs11.conf
Feb 13 14:31:55 database telnetd[13197]: [ID 970222 daemon.error] libpkcs11: Unable to contact kcfd: Bad file number
.........................


The alert_log file
SMON: Parallel transaction recovery tried
Fri Feb 13 14:30:21 2009
Shutting down archive processes
Fri Feb 13 14:30:41 2009
Process startup failed, error stack:
Fri Feb 13 14:30:41 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_psp0_12072.trc:
ORA-27300: OS system dependent operation:fork failed with status: 11
ORA-27301: OS failure message: Resource temporarily unavailable
ORA-27302: failure occurred at: skgpspawn3
Fri Feb 13 14:30:44 2009
Process J001 died, see its trace file
Fri Feb 13 14:30:44 2009
kkjcre1p: unable to spawn jobq slave process 
Fri Feb 13 14:30:44 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_cjq0_12086.trc:

Fri Feb 13 14:31:36 2009
ARCH shutting down
ARC2: Archival stopped
Fri Feb 13 14:31:44 2009
Process startup failed, error stack:
Fri Feb 13 14:31:44 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_psp0_12072.trc:
ORA-27300: OS system dependent operation:fork failed with status: 12
ORA-27301: OS failure message: Not enough space
ORA-27302: failure occurred at: skgpspawn3
Fri Feb 13 14:31:44 2009
Process q002 died, see its trace file
Fri Feb 13 14:31:44 2009
ksvcreate: Process(q002) creation failed
Fri Feb 13 14:31:50 2009
Process startup failed, error stack:
Fri Feb 13 14:31:50 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_psp0_12072.trc:
ORA-27300: OS system dependent operation:fork failed with status: 11
ORA-27301: OS failure message: Resource temporarily unavailable
ORA-27302: failure occurred at: skgpspawn5
Fri Feb 13 14:31:50 2009
Process m000 died, see its trace file
Fri Feb 13 14:31:50 2009
ksvcreate: Process(m000) creation failed
Fri Feb 13 14:31:54 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_j000_12171.trc:
ORA-04030: out of process memory when trying to allocate 8072 bytes (joxcx callheap,f:CDUnscanned)
Fri Feb 13 14:31:57 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_j000_12171.trc:
ORA-07445: exception encountered: core dump [joe_stack_access_control_context()+304] [SIGSEGV] [Address not mapped to object] [0x000000008] [] []
ORA-04030: out of process memory when trying to allocate 8072 bytes (joxcx callheap,f:CDUnscanned)
Fri Feb 13 14:36:51 2009
Errors in file /u02/app/oracle/admin/VNP/udump/vnp_ora_13559.trc:
ORA-07445: exception encountered: core dump [qcplgnt_multibyte()+20] [SIGSEGV] [Address not mapped to object] [0xFFFFFFFF7FFE7720] [] []


Of course, I searched the error in google and metalink, some article like
Quote:

The EAGAIN error means the fork(2) failed because of a temporary resource outage. Perhaps this user had hit the limit for the maximum number of child processes. Check your:

$ ulimit -a

as this user. I would have expected a system call that returned EAGAIN to be retried until it succeeds or the process decides it has tried enough, but I don't know what the RDBMS does here.



Source link: http://forums.oracle.com/forums/thread.jspa?threadID=697339

Source metalink: https://metalink2.oracle.com/metalink/plsql/f?p=130:14:5668155891328437684::::p14_database_id,p14_docid,p14_show_header,p14_show_help ,p14_black_frame,p14_font:NOT,392006.1,1,0,1,helvetica
Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #386348 is a reply to message #386241] Fri, 13 February 2009 21:11 Go to previous messageGo to next message
BlackSwan
Messages: 26766
Registered: January 2009
Location: SoCal
Senior Member
http://www.dbasupport.com/forums/showthread.php?t=53556

above found via GOOGLE below

Results 1 - 10 of about 56,300 for fork failed with status: 11
Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #386352 is a reply to message #386348] Fri, 13 February 2009 21:31 Go to previous messageGo to next message
trantuananh24hg
Messages: 744
Registered: January 2007
Location: Ha Noi, Viet Nam
Senior Member
Thank you, Swan!

I've read the article, more and more yesterday. It looks like a bug of Oracle, but still now, I've not understood why system downed.

May be I have to search continue!
Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #386355 is a reply to message #386241] Fri, 13 February 2009 22:13 Go to previous messageGo to next message
BlackSwan
Messages: 26766
Registered: January 2009
Location: SoCal
Senior Member
It appears the actual error is an OS error.
It could be induced by Oracle bug or by Oracle init.ora value which you particular system can't correctly handle.

The various trace files
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_psp0_12072.trc
may provide clues; then again they may be a wild goose chase.
Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #389685 is a reply to message #386355] Tue, 03 March 2009 03:24 Go to previous messageGo to next message
trantuananh24hg
Messages: 744
Registered: January 2007
Location: Ha Noi, Viet Nam
Senior Member
@BlackSwan & Mahesh Rajendran:

Yesterday, my database had got problem such as those above.
And I found something, especially. However, I separate one by one:

-- In alert_log_file:
Mon Mar  2 18:20:18 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_psp0_964.trc:
ORA-27300: OS system dependent operation:fork failed with status: 11
ORA-27301: OS failure message: Resource temporarily unavailable
ORA-27302: failure occurred at: skgpspawn3
Mon Mar  2 18:20:20 2009
Process J001 died, see its trace file
Mon Mar  2 18:20:21 2009
kkjcre1p: unable to spawn jobq slave process 
Mon Mar  2 18:20:21 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_cjq0_978.trc:

Mon Mar  2 18:20:42 2009
Process startup failed, error stack:
Mon Mar  2 18:20:43 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_psp0_964.trc:
ORA-27300: OS system dependent operation:fork failed with status: 11
ORA-27301: OS failure message: Resource temporarily unavailable
ORA-27302: failure occurred at: skgpspawn3
Mon Mar  2 18:20:43 2009
Process J001 died, see its trace file
Mon Mar  2 18:20:43 2009
kkjcre1p: unable to spawn jobq slave process 
Mon Mar  2 18:20:43 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_cjq0_978.trc:
Mon Mar  2 18:03:53 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_psp0_26217.trc:
ORA-27300: OS system dependent operation:fork failed with status: 12
ORA-27301: OS failure message: Not enough space
ORA-27302: failure occurred at: skgpspawn3
Mon Mar  2 18:03:53 2009
Process m000 died, see its trace file
Mon Mar  2 18:03:53 2009
ksvcreate: Process(m000) creation failed
Mon Mar  2 18:04:54 2009
Process startup failed, error stack:
Mon Mar  2 18:04:54 2009
Errors in file /u02/app/oracle/admin/VNP/bdump/vnp_psp0_26217.trc:
ORA-27300: OS system dependent operation:fork failed with status: 12
ORA-27301: OS failure message: Not enough space
ORA-27302: failure occurred at: skgpspawn3


OK, take a keen on eyes to this bdump_trace_file
/u02/app/oracle/admin/VNP/bdump/vnp_psp0_26217.trc
Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - 64bit Production
With the Partitioning, OLAP and Data Mining options
ORACLE_HOME = /u02/app/oracle/oracle/product/10.2.0/NEO
System name:	SunOS
Node name:	database
Release:	5.10
Version:	Generic_118833-33
Machine:	sun4u
Instance name: VNP
Redo thread mounted by this instance: 1
Oracle process number: 3
Unix process pid: 26217, image: oracle@database (PSP0)

*** SERVICE NAME:(SYS$BACKGROUND) 2009-03-02 17:09:20.507
*** SESSION ID:(1098.1) 2009-03-02 17:09:20.507
*** 2009-03-02 17:09:20.507
Process startup failed, error stack:
ORA-27300: OS system dependent operation:fork failed with status: 11
ORA-27301: OS failure message: Resource temporarily unavailable
ORA-27302: failure occurred at: skgpspawn3
*** 2009-03-02 17:09:48.609
Process startup failed, error stack:
ORA-27300: OS system dependent operation:fork failed with status: 11
ORA-27301: OS failure message: Resource temporarily unavailable
ORA-27302: failure occurred at: skgpspawn3
*** 2009-03-02 17:10:40.258


No such need content information, and then, try to search in metalink.

Quote:

On 10.2.0.2 in Production:
the following error occurs in the alert.log

ERROR
Errors in file /u01/oracle/admin/MFACTOR/bdump/mfactor1_psp0_954436.trc:
ORA-27300: OS system dependent operation:fork failed with status: 11
ORA-27301: OS failure message: Resource temporarily unavailable



BUSINESS IMPACT
The issue has the following business impact:
Due to this issue, users cannot connect to the database.


Verified the issue in the log file as noted below:

LOG FILE
Filename = alert_MFACTOR1.log
See the following error:

Process startup failed, error stack:
Mon Sep 18 18:10:34 2006
Errors in file /u01/oracle/admin/MFACTOR/bdump/mfactor1_psp0_954436.trc:
ORA-27300: OS system dependent operation:fork failed with status: 11
ORA-27301: OS failure message: Resource temporarily unavailable
ORA-27302: failure occurred at: skgpspawn3
Mon Sep 18 18:10:35 2006
Process P073 died, see its trace file

Cause
The issue is caused by the following setup: maximum number of processes allowed per user.

The error messages indicating that oracle has problem in forking more process.

Checked the "maxuproc", it is only set to 128.



Please remove the metalink content after Swan & Mash read, Thank you..

OK, now, I will show to you about max_uproc:
$ su
Password:
# echo maxuprc/D | adb -k
physmem 1fd745
maxuprc:
maxuprc:        29995
# sysdef | grep v_maxup
   29995        maximum processes per user id (v.v_maxup)
# echo maxusers/D | adb -k
physmem 1fd745
maxusers:
maxusers:       2048


In Solaris:

max_nprocs = (10 + 16 * maxusers)
maxuprc = (max_nprocs - 5)


Where:

* max_nprocs = Total number of system wide processes

* maxuprc = Total number of processes a non-root user can have occupying a process table slot at any point in time

Quote:

maxusers is a tunable parameter at the kernel level that is by default set to the lesser of the amount of memory in Mbytes or 2048. Range: 1 to 2048, based on physical memory if not set in the /etc/system file; and 1 to 4096, if set in the /etc/system file.



Therefore, I've got:
logvnp@VNP> select (10+(16*2048)-5) max_uprc from dual;

  MAX_UPRC
----------
     32773

logvnp@VNP>


There is one big problem in here:
maxuprc_os is only 29995 but really is 32773.

May you suggest me for something?

Thank you very much!

Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #389700 is a reply to message #389685] Tue, 03 March 2009 05:32 Go to previous messageGo to next message
babuknb
Messages: 1736
Registered: December 2005
Location: NJ
Senior Member

>>ORA-27300: OS system dependent operation:fork failed with status: 11

Refer Doc ID: 579365.1

>>Process J001 died, see its trace file
Mon Mar 2 18:20:21 2009
kkjcre1p: unable to spawn jobq slave process

You may need to increase your JOB_QUEUE_PROCESSES

Refer Doc ID: 416244.1
Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #389709 is a reply to message #389700] Tue, 03 March 2009 06:10 Go to previous messageGo to next message
trantuananh24hg
Messages: 744
Registered: January 2007
Location: Ha Noi, Viet Nam
Senior Member
@gentlebabu: Thank you for your reply!

It's not simple to do with only increasing job_queue_process initialization parameter. I know about my database, and I am sure how many jobs are running concurrently. In my database, there're 4 jobs was submitted by DBMS_JOBS, 5 jobs submitted by DBMS_SCHEDULER, and job_queue_procees=14.

So that, according to OS max_uprocs & max_users, there are
: 29995 max processes, 2048 max users. In Solaris, how many users are default? There is about 18 users, I'm sorry if it's not exactly, but ~18-22, except oracle users. In one server, how many program for each users? Usually, no much, except oracle's application. Trying to execute "vmstat", "prstat", "sar -u x x", "swap -l" to get the list of memory%, cpu%, vm%, swap% for all users's programming whose are concurrently connect to.

What's about oracle processes? OK, using "ps -ef | grep oracle" and pmap -id in which, id is the PID of oracle's process. I often use those commands to calculate what's size of each process, may be DBWr0, J000, ARC0...

And in oracle database, what's the memory size of each process? Oh, in Dedicated Server, PGA & UGA are not separate components, my database was configured to be dedicated server. Select statement's result come from v$statname, v$sesstat, v$session, v$process will show memory. According to v$pgastat, v$sgastat I can tell you that "The problem is in here". Why can I tell that? Because of statspack. The coding of application had got problem, big one! hmm, what's the problem? Tracing bdump, udump, and ash_report, I found that 2 big problems.

One is INSERT statement: 2972118 buffer_get for only 16 executions.
One is one job scheduled by DBMS_SCHEDULER, in which, scan one very big table ~ 120gb, to execute DML Interval=1 mins.

May be they're really problems, may be the really one is other, by anyway, I said that "Thank you". Because you tell me about jobs.

In note 579365.1, Oracle suggests the DBA to increase max_uprocs & max_nprocs, and you see the post below, in my server, max_upros=29995, and for the 11, ORA-27300: OS system dependent operation:fork failed with status: 11, metalink site has one note: https://metalink2.oracle.com/metalink/plsql/f?p=130:14:2428737725030416143::::p14_database_id,p14_docid,p14_show_header,p14_show_help ,p14_black_frame,p14_font:NOT,392006.1,1,0,1,helvetica

I'll feed back to you when I find the really problem, of course, if the problem are not belonged to our coding. Very Happy

Thank you!

Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #389814 is a reply to message #389709] Tue, 03 March 2009 12:05 Go to previous messageGo to next message
babuknb
Messages: 1736
Registered: December 2005
Location: NJ
Senior Member


Hello,

Yes; Your correct; You may need to increase the processes value; How your going to increase? What's your calculation...

Babu
Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #389875 is a reply to message #389814] Tue, 03 March 2009 20:29 Go to previous messageGo to next message
trantuananh24hg
Messages: 744
Registered: January 2007
Location: Ha Noi, Viet Nam
Senior Member
Thank you for your reply, gentlebabu!
I am going to increase pga (pga_aggregate_target) from 2gb to 2.5gb.

I'm sorry because I removed some thing, it's not the referent table to calculate pga

$ su
Password:
# ps -ef | grep oracle
  oracle 19505     1   0 09:36:12 ?           2:01 oracleVNP (LOCAL=NO)
  oracle   526     1   0   Mar 02 ?           2:29 /u02/app/oracle/oracle/produc
t/10.2.0/NEO/bin/tnslsnr LISTENER -inherit
  oracle 18038     1   0 09:05:18 ?           0:03 oracleVNP (LOCAL=NO)
  oracle 16868     1   0 08:17:43 ?           2:23 oracleVNP (LOCAL=NO)
  oracle  3073  3072   0   Mar 02 ?           0:00 sh -c /u02/app/oracle/oracle/
product/10.2.0/NEO/bin/nmccollector DSGA '(DESCRIP
  oracle  3451     1   0   Mar 02 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 26375     1   1 14:40:59 ?           7:15 oracleVNP (LOCAL=NO)
  oracle 16735     1   0 08:12:36 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 18426     1   0 09:24:03 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 27348     1   0 15:14:54 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 16695     1   0 08:11:41 ?           0:10 oracleVNP (LOCAL=NO)
  oracle 16830     1   0 08:16:47 ?           0:42 oracleVNP (LOCAL=NO)
  oracle  3017     1   0   Mar 02 ?           0:34 oracleVNP (LOCAL=NO)
  oracle 17346     1   0 08:37:14 ?           1:52 oracleVNP (LOCAL=NO)
  oracle  2857     1   0   Mar 02 ?           1:39 ora_arc0_VNP
  oracle  2849     1   0   Mar 02 ?           2:30 ora_mmnl_VNP
  oracle  2845     1   0   Mar 02 ?           1:33 ora_cjq0_VNP
  oracle  2839     1   0   Mar 02 ?           2:39 ora_ckpt_VNP
  oracle  2841     1   0   Mar 02 ?           1:02 ora_smon_VNP
  oracle  2837     1   0   Mar 02 ?           5:22 ora_lgwr_VNP
  oracle  2847     1   0   Mar 02 ?           0:22 ora_mmon_VNP
  oracle  2843     1   0   Mar 02 ?           0:00 ora_reco_VNP
  oracle  2917     1   0   Mar 02 ?           0:01 ora_q001_VNP
  oracle 17707     1   0 08:51:41 ?           0:09 oracleVNP (LOCAL=NO)
  oracle 17814     1   0 08:56:52 ?           0:00 oracleVNP (LOCAL=NO)
  oracle  3557     1   0   Mar 02 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 18422     1   0 09:23:46 ?           0:00 oracleVNP (LOCAL=NO)
  oracle  2867     1   0   Mar 02 ?           0:02 ora_qmnc_VNP
  oracle  2891     1   1   Mar 02 ?          79:01 ora_j000_VNP
  oracle  1097     1   0   Mar 02 ?           2:19 /u02/app/oracle/oracle/produc
t/10.2.0/NEO/jdk/bin/java -Djava.awt.headless=true
  oracle 29917     1   0 17:00:58 ?           1:31 oracleVNP (LOCAL=NO)
  oracle 18370     1   0 09:20:11 ?           0:04 oracleVNP (LOCAL=NO)
  oracle  2859     1   1   Mar 02 ?           2:11 ora_arc1_VNP
  oracle  2851     1   0   Mar 02 ?           0:00 ora_d000_VNP
  oracle  2853     1   0   Mar 02 ?           0:00 ora_s000_VNP
  oracle 18040     1   0 09:05:27 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 29403     1   0 16:36:17 ?           0:00 oracleVNP (LOCAL=NO)
  oracle  3022     1   0   Mar 02 ?           0:22 oracleVNP (LOCAL=NO)
  oracle 17972     1   0 09:01:29 ?           0:16 oracleVNP (LOCAL=NO)
  oracle  3100     1   0   Mar 02 ?           0:04 oracleVNP (LOCAL=NO)
  oracle 16769     1   1 08:13:50 ?           2:59 oracleVNP (LOCAL=NO)
  oracle  2907     1   0   Mar 02 ?           0:02 ora_q000_VNP
  oracle 18920     1   0   Mar 03 ?           2:49 oracleVNP (LOCAL=NO)
  oracle 17191     1   1 08:31:40 ?           1:09 oracleVNP (LOCAL=NO)
  oracle 16279     1   0 07:51:55 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 20606     1   0 10:24:01 ?           5:39 oracleVNP (LOCAL=NO)
  oracle  3030     1   0   Mar 02 ?           1:44 oracleVNP (LOCAL=NO)
  oracle 17647     1   0 08:48:16 ?           0:01 oracleVNP (LOCAL=NO)
  oracle  2829     1   0   Mar 02 ?           1:46 ora_pmon_VNP
  oracle 16689     1   0 08:11:39 ?           0:02 oracleVNP (LOCAL=NO)
  oracle  1470     1   0   Mar 02 ?           2:31 /u02/app/oracle/oracle/produc
t/10.2.0/NEO/perl/bin/perl /u02/app/oracle/oracle/
  oracle  2735  1470   0   Mar 02 ?          70:55 /u02/app/oracle/oracle/produc
t/10.2.0/NEO/jdk/bin/java -server -Xmx256M -XX:Max
  oracle 18080     1   0 09:07:19 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 17138     1   0   Mar 03 ?           6:38 oracleVNP (LOCAL=NO)
  oracle  3048     1   0 19:11:52 ?           0:03 oracleVNP (LOCAL=NO)
  oracle 20822     1   0 10:31:44 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 16777     1   0 08:14:25 ?           0:04 oracleVNP (LOCAL=NO)
  oracle  3072     1   0   Mar 02 ?           0:00 /u02/app/oracle/oracle/produc
t/10.2.0/NEO/perl/bin/perl /u02/app/oracle/oracle/
  oracle 16747     1   0 08:13:26 ?           0:02 oracleVNP (LOCAL=NO)
  oracle  2938     1   0   Mar 02 ?           3:07 oracleVNP (LOCAL=NO)
  oracle 16820     1   0 08:16:03 ?           0:36 oracleVNP (LOCAL=NO)
  oracle 18424     1   0 09:23:50 ?           0:02 oracleVNP (LOCAL=NO)
  oracle 16956     1   2 08:23:10 ?           1:41 oracleVNP (LOCAL=NO)
  oracle  2019  1470   0   Mar 02 ?          10:51 /u02/app/oracle/oracle/produc
t/10.2.0/NEO/bin/emagent
  oracle  2930     1   0   Mar 02 ?           1:13 oracleVNP (LOCAL=NO)
  oracle 17859     1   0 08:59:42 ?           0:08 oracleVNP (LOCAL=NO)
    root 18444 18443   0 09:25:32 pts/1       0:00 grep oracle
  oracle 16657     1   0 08:10:34 ?           0:20 oracleVNP (LOCAL=NO)
  oracle 16745     1   0 08:13:26 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 16950     1   0 08:22:28 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 18433     1   0 09:24:45 ?           0:00 oracleVNP (LOCAL=NO)
  oracle  2835     1   0   Mar 02 ?          14:00 ora_dbw0_VNP
  oracle 23073     1   1 12:20:23 ?           4:57 oracleVNP (LOCAL=NO)
  oracle  3491     1   0   Mar 02 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 16714     1   0 08:11:57 ?           0:03 oracleVNP (LOCAL=NO)
  oracle 16890     1   1 08:19:25 ?           0:28 oracleVNP (LOCAL=NO)
  oracle  2833     1   0   Mar 02 ?           0:07 ora_mman_VNP
  oracle 16767     1   0 08:13:44 ?           0:03 oracleVNP (LOCAL=NO)
  oracle 17602     1   0 08:45:40 ?           0:14 oracleVNP (LOCAL=NO)
  oracle  3074  3073   1   Mar 02 ?          34:09 /u02/app/oracle/oracle/produc
t/10.2.0/NEO/bin/nmccollector DSGA (DESCRIPTION=(A
  oracle   855     1   1 17:37:18 ?           1:30 oracleVNP (LOCAL=NO)
  oracle  3062     1   0 19:12:01 ?           1:01 oracleVNP (LOCAL=NO)
  oracle 16763     1   0 08:13:39 ?           0:10 oracleVNP (LOCAL=NO)
  oracle 16912     1   0 08:20:42 ?           0:21 oracleVNP (LOCAL=NO)
  oracle  3519     1   0   Mar 02 ?           0:07 oracleVNP (LOCAL=NO)
  oracle  2831     1   0   Mar 02 ?           0:04 ora_psp0_VNP
  oracle 16697     1   0 08:11:48 ?           0:18 oracleVNP (LOCAL=NO)
  oracle 17350     1   0 08:37:29 ?           0:08 oracleVNP (LOCAL=NO)
  oracle 16910     1   0 08:20:31 ?           0:05 oracleVNP (LOCAL=NO)
  oracle 20681     1   0 10:27:45 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 17258     1   0 08:34:14 ?           0:02 oracleVNP (LOCAL=NO)
  oracle  3227     1   0   Mar 02 ?           1:15 oracleVNP (LOCAL=NO)
  oracle 18410     1   0 09:23:15 ?           0:02 oracleVNP (LOCAL=NO)
  oracle  2934     1   0   Mar 02 ?           0:24 oracleVNP (LOCAL=NO)
  oracle  2863     1   0   Mar 02 ?           0:46 ora_ctwr_VNP
  oracle 16858     1   0 08:17:27 ?           1:01 oracleVNP (LOCAL=NO)
  oracle  6472     1   0   Mar 02 ?           0:01 oracleVNP (LOCAL=NO)
  oracle   396     1   0 17:18:38 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 18338     1   0 09:18:23 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 16775     1   0 08:14:22 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 16749     1   2 08:13:28 ?           0:59 oracleVNP (LOCAL=NO)
  oracle 17553     1   0   Mar 03 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 16773     1   0 08:14:08 ?           0:23 oracleVNP (LOCAL=NO)
  oracle  3050     1   0 19:11:52 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 17563     1   0   Mar 03 ?           6:32 oracleVNP (LOCAL=NO)
  oracle 16906     1   0 08:20:27 ?           0:26 oracleVNP (LOCAL=NO)
  oracle 17728     1   0   Mar 03 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 17593     1   0   Mar 03 ?           0:12 oracleVNP (LOCAL=NO)
  oracle 29768     1   0 16:54:40 ?           3:30 oracleVNP (LOCAL=NO)
  oracle 17710     1   0   Mar 03 ?           0:37 oracleVNP (LOCAL=NO)
  oracle 16716     1   0 08:11:57 ?           0:32 oracleVNP (LOCAL=NO)
  oracle 17179     1   0 08:31:36 ?           0:24 oracleVNP (LOCAL=NO)
  oracle 18549     1   0   Mar 03 ?           2:14 oracleVNP (LOCAL=NO)
  oracle 18036     1   0 09:05:18 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 18551     1   0   Mar 03 ?           0:09 oracleVNP (LOCAL=NO)
  oracle 16976     1   0 08:24:37 ?           0:05 oracleVNP (LOCAL=NO)
  oracle 17116     1   0 08:27:58 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 16757     1   2 08:13:32 ?           1:44 oracleVNP (LOCAL=NO)
  oracle 17384     1   0 08:39:14 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 16974     1   0 08:24:37 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 16733     1   0 08:12:24 ?           0:44 oracleVNP (LOCAL=NO)
  oracle 20170     1   1 10:02:28 ?          10:13 oracleVNP (LOCAL=NO)
  oracle 17731     1   0 08:52:10 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 16908     1   0 08:20:28 ?           0:02 oracleVNP (LOCAL=NO)
  oracle 22106     1   0 11:32:09 ?           0:01 oracleVNP (LOCAL=NO)
  oracle 17853     1   1 08:58:56 ?           0:16 oracleVNP (LOCAL=NO)
  oracle 25207     1   0 13:44:00 ?           0:00 oracleVNP (LOCAL=NO)
  oracle   857     1   0 17:37:18 ?           0:00 oracleVNP (LOCAL=NO)
  oracle 16759     1   1 08:13:36 ?           0:13 oracleVNP (LOCAL=NO)
  oracle 17322     1   0 08:36:45 ?           1:04 oracleVNP (LOCAL=NO)
  oracle 18439 18436   0 09:25:17 pts/1       0:00 -ksh
  oracle 16761     1   0 08:13:39 ?           0:51 oracleVNP (LOCAL=NO)
  oracle 17816     1   2 08:56:52 ?           0:24 oracleVNP (LOCAL=NO)
  oracle 29921     1   0 17:01:22 ?           0:18 oracleVNP (LOCAL=NO)
  oracle   859     1   0 17:37:18 ?           0:03 oracleVNP (LOCAL=NO)
  oracle 16811     1   0 08:15:40 ?           0:05 oracleVNP (LOCAL=NO)
#


Calculation memory for each process:
Example: 2981/ job_process
# pmap 2891
2891:   ora_j000_VNP
0000000100000000      98304K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/bi
n/oracle
0000000106000000       1824K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/bi
n/oracle
00000001062C6000        808K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/bi
n/oracle
0000000106390000       1160K rwx--    [ heap ]
0000000380000000   10485784K rwxsR    [ ism shmid=0x24 ]
FFFFFFFF79660000         64K rw---    [ anon ]
FFFFFFFF79AF0000         64K rw---    [ anon ]
FFFFFFFF79B00000         64K rw---    [ anon ]
FFFFFFFF79B10000        128K rw---    [ anon ]
FFFFFFFF79B30000         64K rw---    [ anon ]
FFFFFFFF79B40000        128K rw---    [ anon ]
FFFFFFFF79B60000         64K rw---    [ anon ]
FFFFFFFF79B70000         64K rw---    [ anon ]
FFFFFFFF79B80000         64K rw---    [ anon ]
FFFFFFFF79C70000         64K rw---    [ anon ]
FFFFFFFF79C80000         64K rw---    [ anon ]
FFFFFFFF79C90000         64K rw---    [ anon ]
FFFFFFFF79CA0000         64K rw---    [ anon ]
FFFFFFFF79CB0000         64K rw---    [ anon ]
FFFFFFFF79CC0000         64K rw---    [ anon ]
FFFFFFFF79CD0000         64K rw---    [ anon ]
FFFFFFFF79CE0000         64K rw---    [ anon ]
FFFFFFFF79D00000         64K rw---    [ anon ]
FFFFFFFF79D10000         64K rw---    [ anon ]
FFFFFFFF79D20000         64K rw---    [ anon ]
FFFFFFFF79D30000         64K rw---    [ anon ]
FFFFFFFF79D40000         64K rw---    [ anon ]
FFFFFFFF79D50000         64K rw---    [ anon ]
FFFFFFFF79D60000         64K rw---    [ anon ]
FFFFFFFF79D70000         64K rw---    [ anon ]
FFFFFFFF79DC4000          8K rwxs-    [ anon ]
FFFFFFFF79E00000          8K r-x--  /platform/sun4u-us3/lib/sparcv9/libc_psr.so.
1
FFFFFFFF79F00000         24K rwx--    [ anon ]
FFFFFFFF7A000000          8K rwx--    [ anon ]
FFFFFFFF7A100000          8K r-x--  /lib/sparcv9/libmd5.so.1
FFFFFFFF7A202000          8K rwx--  /lib/sparcv9/libmd5.so.1
FFFFFFFF7A300000          8K rwx--    [ anon ]
FFFFFFFF7A400000        640K r-x--  /lib/sparcv9/libm.so.2
FFFFFFFF7A59E000         40K rwx--  /lib/sparcv9/libm.so.2
FFFFFFFF7A600000         32K r-x--  /lib/sparcv9/librt.so.1
FFFFFFFF7A708000          8K rwx--  /lib/sparcv9/librt.so.1
FFFFFFFF7A800000         32K r-x--  /lib/sparcv9/libaio.so.1
FFFFFFFF7A908000          8K rwx--  /lib/sparcv9/libaio.so.1
FFFFFFFF7AA00000        920K r-x--  /lib/sparcv9/libc.so.1
FFFFFFFF7ABE6000         64K rwx--  /lib/sparcv9/libc.so.1
FFFFFFFF7ABF6000          8K rwx--  /lib/sparcv9/libc.so.1
FFFFFFFF7AD00000          8K rwx--    [ anon ]
FFFFFFFF7AE00000          8K r-x--  /usr/lib/sparcv9/libsched.so.1
FFFFFFFF7AF00000         32K r-x--  /lib/sparcv9/libgen.so.1
FFFFFFFF7B008000          8K rwx--  /lib/sparcv9/libgen.so.1
FFFFFFFF7B100000         56K r-x--  /lib/sparcv9/libsocket.so.1
FFFFFFFF7B20E000         16K rwx--  /lib/sparcv9/libsocket.so.1
FFFFFFFF7B300000        688K r-x--  /lib/sparcv9/libnsl.so.1
FFFFFFFF7B4AC000         64K rwx--  /lib/sparcv9/libnsl.so.1
FFFFFFFF7B4BC000         32K rwx--  /lib/sparcv9/libnsl.so.1
FFFFFFFF7B500000          8K r-x--  /lib/sparcv9/libkstat.so.1
FFFFFFFF7B602000          8K rwx--  /lib/sparcv9/libkstat.so.1
FFFFFFFF7B700000       2176K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libnnz10.so
FFFFFFFF7BA1E000        240K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libnnz10.so
FFFFFFFF7BA5A000          8K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libnnz10.so
FFFFFFFF7BB00000         64K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libdbcfg10.so
FFFFFFFF7BC0E000         16K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libdbcfg10.so
FFFFFFFF7BD00000       1056K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libclsra10.so
FFFFFFFF7BF06000         56K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libclsra10.so
FFFFFFFF7BF14000          8K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libclsra10.so
FFFFFFFF7C000000       8192K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libjox10.so
FFFFFFFF7C800000       1064K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libjox10.so
FFFFFFFF7CA08000        552K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libjox10.so
FFFFFFFF7CB00000          8K rwx--    [ anon ]
FFFFFFFF7CC00000          8K rwx--    [ anon ]
FFFFFFFF7CD00000         40K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libocrutl10.so
FFFFFFFF7CE08000         16K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libocrutl10.so
FFFFFFFF7CF00000        304K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libocrb10.so
FFFFFFFF7D04A000         24K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libocrb10.so
FFFFFFFF7D100000       1472K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libocr10.so
FFFFFFFF7D36E000         72K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libocr10.so
FFFFFFFF7D380000          8K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libocr10.so
FFFFFFFF7D400000          8K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libskgxn2.so
FFFFFFFF7D500000          8K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libskgxn2.so
FFFFFFFF7D600000          8K rwx--    [ anon ]
FFFFFFFF7D700000       1728K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libhasgen10.so
FFFFFFFF7D9AE000         80K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libhasgen10.so
FFFFFFFF7D9C2000          8K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libhasgen10.so
FFFFFFFF7DA00000        192K r-x--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libskgxp10.so
FFFFFFFF7DB2E000         16K rwx--  /u02/app/oracle/oracle/product/10.2.0/NEO/li
b/libskgxp10.so
FFFFFFFF7DC00000          8K r-x--  /lib/sparcv9/libdl.so.1
FFFFFFFF7DD00000          8K rwx--    [ anon ]
FFFFFFFF7DE00000        176K r-x--  /lib/sparcv9/ld.so.1
FFFFFFFF7DF2C000         16K rwx--  /lib/sparcv9/ld.so.1
FFFFFFFF7FFBE000        264K rw---    [ stack ]
         total     10610264K
#


You see 0000000380000000 10485784K rwxsR [ ism shmid=0x24 ]

Take total=10610264 - 10485784 =

logvnp@VNP> select (10610264 - 10485784)/1024 Mb from dual;

        MB
----------
  121,5625


121MB, the process take it from physical memory. Try to capture each other to calculate.

May you correct me if i'm wrong?
Thank you very much!

[Updated on: Tue, 03 March 2009 21:12]

Report message to a moderator

Re: CPU in Solaris 10 achieved 100% with Oracle Database [message #390150 is a reply to message #389875] Thu, 05 March 2009 03:45 Go to previous message
babuknb
Messages: 1736
Registered: December 2005
Location: NJ
Senior Member

Hi,

Sorry for delay response.

You going to increase process parameter in oracle database??

Babu
Previous Topic: Running sql script on UNIX
Next Topic: Resurrect a backup corrupted by FTP (ASCII)
Goto Forum:
  


Current Time: Thu Mar 28 09:24:33 CDT 2024