Originally posted by: sajid1

Hi,

AIX 5.3
Oracle Database 10.2.0.2
EBS R12

After cloning Application and database are running very slow.while i am running topas after taking down all services the kernel parameter is showing 90% for sometime.For your information there is a running process "syncd" might be the faulty of performence degrade.Find below topas output while kernel went 98.7% after taking down all services.Nothing is running. Suggest me.

Topas Monitor for host: emgr EVENTS/QUEUES FILE/TTY
Wed Aug 11 11:09:29 2010 Interval: 2 Cswitch 141 Readch 0
Syscall 43 Writech 320
Kernel 98.7 |############################| Reads 0 Rawin 0
User 0.3 |# | Writes 1 Ttyout 320
Wait 0.0 | | Forks 0 Igets 0
Idle 1.0 |# | Execs 0 Namei 0
Physc = 0.20 %Entc= 99.8 Runqueue 2.0 Dirblk 0
Waitqueue 0.0
Network KBPS I-Pack O-Pack KB-In KB-Out
en2 0.6 4.0 2.0 0.2 0.4K PAGING MEMORY
lo0 0.0 0.0 0.0 0.0 0.0K Faults 0 Real,MB 8192
Steals 0 % Comp 34.7
Disk Busy% KBPS TPS KB-Read KB-Writ PgspIn 0 % Noncomp 56.0
hdisk0 1.0 4.0 1.0 0.0 4.0K PgspOut 0 % Client 56.0
hdisk1 0.5 4.0 1.0 0.0 4.0 PageIn 0
hdisk2 0.0 2.0 0.5 0.0 2.0 PageOut 0 PAGING SPACE
Sios 0 Size,MB 19200
Name PID CPU% PgSp Owner % Used 2.1
syncd 86104 16.2 0.5 root NFS (calls/sec) % Free 97.8
topas 1036488 1.1 1.9 oracle ServerV2 0
topas 557212 1.1 1.9 root ClientV2 0 Press:
getty 209036 0.0 0.4 root ServerV3 0 "h" for help
gil 61470 0.0 0.9 root ClientV3 0 "q" to quit
nfsd 176266 0.0 1.8 root

Thanks,
#AIX-Forum

Originally posted by: DjRobin69

syncd could be syncing the data across your mirrored disks if you have them?

Can you also post the output from the following

ps augw | head -1 ; ps augw | sort -rnk 3 | head

vmstat 1 10
#AIX-Forum

Originally posted by: sajid1

Thanks for replying.

syncd could be syncing the data across your mirrored disks if you have them?
---How could i know it.

Find below output as required by you.

This output while application is down and database down It ran by applmgr user that is the
owner of application user.

$ ps augw | head -1 ; ps augw | sort -rnk 3 | head
USER PID %CPU %MEM SZ RSS TTY STAT STIME TIME COMMAND
root 8196 0.1 0.0 384 384 - A 10:20:50 0:03 wait
oracle 495858 0.1 1.0 105880 56208 - A 10:39:30 0:01 ora_arc0_emgr
oracle 381124 0.1 1.0 93352 43680 - A 10:39:23 0:01 ora_cjq0_emgr
oracle 360640 0.1 1.0 94160 44488 - A 10:39:29 0:01 oracleemgr (DES
CRIPTION=(LOCAL=YES)(ADDRESS=(PROTOCOL=beq)))
root 532488 0.0 0.0 688 708 - A 10:40:03 0:00 telnetd -a
root 487662 0.0 0.0 448 448 - A 10:39:30 0:00 aioserver
root 483564 0.0 0.0 448 448 - A 10:39:30 0:00 aioserver
root 471270 0.0 0.0 448 448 - A 10:39:29 0:00 aioserver
root 467172 0.0 0.0 448 448 - A 10:39:29 0:00 aioserver

This is while application and database are up

$ ps augw | head -1 ; ps augw | sort -rnk 3 | head
USER PID %CPU %MEM SZ RSS TTY STAT STIME TIME COMMAND
applmgr 1032230 3.7 1.0 87880 87900 - A 10:56:23 0:18 /vol02/applmgr
/EMGR/apps/tech_st/10.1.3/appsutil/jdk/bin/java -DCLI
applmgr 745642 3.5 1.0 79584 79604 - A 10:56:50 0:15 /vol02/applmgr
/EMGR/apps/tech_st/10.1.3/appsutil/jdk/bin/java -DCLI
oracle 655434 2.6 1.0 105676 56068 - A 11:00:07 0:01 oracleemgr (L
OCAL=NO)
applmgr 983274 1.5 1.0 59652 59672 - A 10:56:36 0:07 /vol02/applmgr
/EMGR/apps/tech_st/10.1.3/appsutil/jdk/bin/java -DCLI
applmgr 958694 1.5 1.0 46492 46512 pts/2 A 10:57:41 0:05 /vol02/applmgr
/EMGR/apps/tech_st/10.1.3/appsutil/jdk/bin/java -DCLI
applmgr 704626 1.5 1.0 59324 59344 - A 10:56:36 0:07 /vol02/applmgr
/EMGR/apps/tech_st/10.1.3/appsutil/jdk/bin/java -DCLI
applmgr 856236 1.2 1.0 44092 44112 pts/2 A 10:57:41 0:04 /vol02/applmgr
/EMGR/apps/tech_st/10.1.3/appsutil/jdk/bin/java -DCLI
applmgr 827544 1.2 1.0 42124 42144 pts/2 A 10:57:41 0:04 /vol02/applmgr
/EMGR/apps/tech_st/10.1.3/appsutil/jdk/bin/java -DCLI
applmgr 839838 0.9 1.0 40380 40400 pts/2 A 10:57:41 0:03 /vol02/applmgr
/EMGR/apps/tech_st/10.1.3/appsutil/jdk/bin/java -DCLI
applmgr 802964 0.9 1.0 41480 41500 pts/2 A 10:57:41 0:03 /vol02/applmgr
/EMGR/apps/tech_st/10.1.3/appsutil/jdk/bin/java -DCLI

---

$ vmstat 1 10

System configuration: lcpu=2 mem=8192MB ent=1.00

kthr memory page faults cpu

---

-----------

---

------------

---

r b avm fre re pi po fr sr cy in sy cs us sy id wa pc ec
5 0 568839 775936 0 0 0 0 0 0 39 6588 833 71 2 27 0 0.73 73.
3
6 0 570050 774724 0 0 0 0 0 0 77 12407 1458 90 3 7 0 0.93 9
3.3
1 0 569375 775386 0 0 0 0 0 0 278 6804 1219 20 19 62 0 0.39 38
.9
2 0 576250 768510 0 0 0 0 0 0 22 35420 1105 84 15 1 0 1.00 9
9.7
3 0 580225 764535 0 0 0 0 0 0 16 24121 1749 95 5 0 0 1.00 10
0.0
3 0 583905 760855 0 0 0 0 0 0 12 7330 989 98 2 0 0 1.00 99.
9
1 0 585275 759485 0 0 0 0 0 0 2 4874 941 98 1 0 0 1.00 100.
2
5 0 586142 758618 0 0 0 0 0 0 10 2484 712 99 1 0 0 1.00 100.
3
2 0 591099 753660 0 0 0 0 0 0 8 15733 1304 95 4 1 0 1.00 9
9.7
4 0 592931 751822 0 0 0 0 0 0 36 6221 904 96 2 2 0 0.99 98.
5
$

Regards,
#AIX-Forum

Originally posted by: alethad

You need to adjust the OS settings on AIX for Oracle.

There is a Red paper on AIX & Oracle Tuning. Do you have that?
If not let me know & I will send it. I have put it out in this forum more than once.
#AIX-Forum

Originally posted by: sajid1

Hi,
There is a Red paper on AIX & Oracle Tuning. Do you have that?
--Is it this book "Oracle_tuning_AIX_Jan09".

Thanks
#AIX-Forum