OAS 10g and HP rp4440?

[Jose Manuel Fager Rojas]

Hello
I got a rp4440 with 4GB ram, we recently got OAS 10g and install it with all the options but we note that the free mem (TOP) was left in 1.2GB! so we stop the database instance and other instances, we only left the j2ee instance and nothing more but the memory still low we got today >1Gb and there are no more process or services in the machine, is this behaviour normal?

I paste the output of /usr/contrib/Q4/bin/kmeminfo:
Physical memory usage summary (in pages):

Physmem = 1310208 Available physical memory:
Freemem = 239221 Free physical memory
Used = 1070987 Used physical memory:
System = 696229 by kernel:
Static = 41091 for text and static data
Dynamic = 0 for dynamic data
Bufcache = 655104 for file-system buffer cache
Eqmem = 34 for equiv. mapped page pool
SCmem = 0 for system critical page pool
User = 149928 by user processes
Uarea = 4488 for thread uareas
Disowned = 32538 disowned pages

And the output of TOP:
Load averages: 0.02, 0.10, 0.14
136 processes: 109 sleeping, 24 running, 3 zombies
Cpu states:
CPU LOAD USER NICE SYS IDLE BLOCK SWAIT INTR SSYS
0 0.03 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
1 0.00 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%
--- ---- ----- ----- ----- ----- ----- ----- ----- -----
avg 0.02 0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 0.0% 0.0%

Memory: 559144K (495624K) real, 1530432K (1448352K) virtual, 956756K free Page# 1/6

CPU TTY PID USERNAME PRI NI SIZE RES STATE TIME %WCPU %CPU COMMAND
0 ? 15545 orainsc 152 24 268M 75096K run 1:38 0.59 0.59 java
0 pty/ttyp1 2652 orainsc 152 24 494M 118M run 1:02 0.56 0.55 java
0 pty/ttyp1 2651 orainsc 152 24 271M 69400K run 0:44 0.53 0.53 emagent
0 ? 36 root 152 20 8448K 8448K run 6:41 0.43 0.43 vxfsd
0 ? 8263 orainsc 152 24 470M 106M run 3:33 0.34 0.34 java
1 ? 8264 orainsc 152 24 22808K 4576K run 2:35 0.29 0.29 opmn
1 ? 1275 root 152 20 23308K 4740K run 0:26 0.12 0.12 vxsvc
1 ? 1644 root 154 20 4328K 700K sleep 0:06 0.06 0.06 dm_stape
1 ? 15128 orainsc 152 24 27352K 3712K run 0:07 0.06 0.06 httpd
0 ? 448 root 152 20 2060K 304K run 0:38 0.04 0.04 syncer
0 pty/ttyp1 2650 orainsc 168 24 8984K 4120K sleep 0:00 0.04 0.04 perl
0 ? 15129 orainsc 152 24 27892K 4332K run 0:00 0.04 0.04 httpd
1 ? 15131 orainsc 152 24 28020K 4460K run 0:00 0.04 0.04 httpd
0 ? 15132 orainsc 152 24 28020K 4460K run 0:00 0.04 0.04 httpd
1 ? 15133 orainsc 152 24 27892K 4332K run 0:00 0.04 0.04 httpd
1 ? 15134 orainsc 152 24 28020K 4460K run 0:00 0.04 0.04 httpd
1 ? 15152 orainsc 152 24 27892K 4332K run 0:00 0.04 0.04 httpd
1 ? 15153 orainsc 152 24 27892K 4332K run 0:00 0.04 0.04 httpd
0 ? 15154 orainsc 152 24 27892K 4332K run 0:00 0.04 0.04 httpd
1 ? 15155 orainsc 152 24 27892K 4332K run 0:00 0.04 0.04 httpd
0 ? 15156 orainsc 152 24 27892K 4332K run 0:00 0.04 0.04 httpd
0 ? 15161 orainsc 152 24 28020K 4460K run 0:00 0.04 0.04 httpd
1 ? 0 root 127 20 32K 0K sleep 0:24 0.02 0.02 swapper
0 ? 1 root 168 20 488K 204K sleep 0:04 0.02 0.02 init
1 ? 2 root 128 20 32K 32K sleep 0:01 0.02 0.02 vhand
1 ? 3 root 128 20 32K 32K sleep 4:41 0.02 0.02 statdaemon
1 ? 4 root 128 20 32K 32K sleep 0:06 0.02 0.02 unhashdaemon

Can someone explain this ?!!!
thanks

Vladimir Ariza
Support Engineer

[Jose Manuel Fager Rojas]

Now I got kmeminfo v4 and this is the output with the -u option:

# ./kmeminfo -u

Kernel TEXT pages not requested in crashconf

Will use an artificial mapping from a.out TEXT pages

kmeminfo (4.00)

unix: /stand/vmunix 11.11 64bit PA2.0

core: /dev/kmem live

link: Fri Jul 23 10:14:00 SAT 2004

boot: Wed Jul 28 16:15:14 2004

dump: Wed Aug 11 12:37:14 2004



Note: resyncing "freemem" with freemem caches: was 228314, now 228345



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

Summary of processes memory usage:



List sorted by physical size, in pages/bytes:



virtual physical swap

pid ppid pages / bytes pages / bytes pages / bytes command

2652 2650 126594 494.5m 36393 142.2m 40292 157.4m java

8263 1 117693 459.7m 29114 113.7m 30117 117.6m java

2651 2650 70603 275.8m 25522 99.7m 25665 100.3m emagent

8972 8264 66614 260.2m 20147 78.7m 22818 89.1m java

1029 1 3991 15.6m 2697 10.5m 1030 4.0m cimserver

1275 1 5828 22.8m 2476 9.7m 3450 13.5m vxsvc

8264 8079 5703 22.3m 1454 5.7m 1643 6.4m opmn

7496 8306 7008 27.4m 1318 5.1m 1316 5.1m httpd

7640 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7495 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7505 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7506 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7498 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7504 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7503 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7497 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7634 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7493 8306 6976 27.2m 1286 5.0m 1284 5.0m httpd

7641 8306 6944 27.1m 1254 4.9m 1252 4.9m httpd

1138 1 2733 10.7m 1189 4.6m 916 3.6m swagentd

8335 1 2247 8.8m 1175 4.6m 1098 4.3m perl

2650 1 2247 8.8m 1175 4.6m 1098 4.3m perl

1026 1 2575 10.1m 1136 4.4m 1021 4.0m dced

7492 8306 6777 26.5m 1067 4.2m 1086 4.2m httpd

7490 8306 6679 26.1m 998 3.9m 984 3.8m httpd

8306 8264 6679 26.1m 998 3.9m 984 3.8m httpd

1738 1 1680 6.6m 827 3.2m 839 3.3m ia64_corehw

2496 1 1418 5.5m 787 3.1m 304 1.2m hpterm

955 1 2893 11.3m 776 3.0m 404 1.6m snmpdm

1768 1 1301 5.1m 351 1.4m 341 1.3m lpmc_em

20604 1 901 3.5m 315 1.2m 336 1.3m rpc.lockd

977 1 1398 5.5m 293 1.1m 288 1.1m mib2agt

8079 1 1285 5.0m 289 1.1m 160 640.0k opmn

1696 1 935 3.7m 274 1.1m 275 1.1m fpl_em

1387 1111 1121 4.4m 263 1.0m 310 1.2m psmctd

1386 1111 1121 4.4m 263 1.0m 326 1.3m memlogd

20598 1 910 3.6m 261 1.0m 360 1.4m rpc.statd

20569 1 843 3.3m 257 1.0m 289 1.1m rpc.mountd

645 1 812 3.2m 252 1008.0k 273 1.1m rpcbind

1352 1305 901 3.5m 252 1008.0k 141 564.0k dtlogin

945 1 910 3.6m 248 992.0k 173 692.0k sendmail

1521 1 1168 4.6m 247 988.0k 299 1.2m dm_core_hw

1440 1 1094 4.3m 243 972.0k 261 1.0m disk_em

1644 1 1083 4.2m 234 936.0k 259 1.0m dm_stape

1807 1 1079 4.2m 232 928.0k 257 1.0m sysstat_em

1111 1 1013 4.0m 229 916.0k 209 836.0k diagmond

1584 1 1061 4.1m 221 884.0k 248 992.0k dm_memory

567 566 977 3.8m 202 808.0k 268 1.0m netfmt

968 1 1213 4.7m 201 804.0k 213 852.0k hp_unixagt

986 1 1187 4.6m 182 728.0k 210 840.0k trapdestagt

1314 1 937 3.7m 160 640.0k 192 768.0k cimserverd

1385 1111 971 3.8m 154 616.0k 311 1.2m diaglogd

1394 672 608 2.4m 146 584.0k 148 592.0k registrar

1011 1 717 2.8m 137 548.0k 131 524.0k fddi4subagt

20586 20582 719 2.8m 129 516.0k 172 688.0k nfsd

20595 20581 719 2.8m 129 516.0k 172 688.0k nfsd

20581 1 719 2.8m 129 516.0k 172 688.0k nfsd

672 1 776 3.0m 129 516.0k 160 640.0k inetd

20592 20581 719 2.8m 129 516.0k 172 688.0k nfsd

20583 20581 719 2.8m 129 516.0k 172 688.0k nfsd

20597 20582 719 2.8m 129 516.0k 172 688.0k nfsd

20582 20581 719 2.8m 129 516.0k 172 688.0k nfsd

20591 20581 719 2.8m 129 516.0k 172 688.0k nfsd

20594 20581 719 2.8m 129 516.0k 172 688.0k nfsd

20587 20581 719 2.8m 129 516.0k 172 688.0k nfsd

20589 20581 719 2.8m 129 516.0k 172 688.0k nfsd

20596 20582 719 2.8m 129 516.0k 172 688.0k nfsd

20590 20582 719 2.8m 129 516.0k 172 688.0k nfsd

20585 20582 719 2.8m 129 516.0k 172 688.0k nfsd

20584 20582 719 2.8m 129 516.0k 172 688.0k nfsd

20588 20582 719 2.8m 129 516.0k 172 688.0k nfsd

1313 1 772 3.0m 116 464.0k 147 588.0k p_client

1183 1 696 2.7m 116 464.0k 131 524.0k emsagent

1068 1 694 2.7m 114 456.0k 147 588.0k pwgrd

8765 8764 688 2.7m 104 416.0k 129 516.0k ksh

8764 672 937 3.7m 100 400.0k 139 556.0k telnetd

10931 672 937 3.7m 100 400.0k 139 556.0k telnetd

20580 1 687 2.7m 97 388.0k 140 560.0k nfsd

7491 8306 509 2.0m 96 384.0k 101 404.0k sh

7488 8306 509 2.0m 96 384.0k 101 404.0k sh

1043 1 485 1.9m 93 372.0k 107 428.0k rbootd

2497 2496 676 2.6m 90 360.0k 117 468.0k ksh

448 1 516 2.0m 82 328.0k 127 508.0k syncer

10932 10931 148 592.0k 81 324.0k 60 240.0k sh

566 1 480 1.9m 78 312.0k 111 444.0k ntl_reader

1309 1 94 376.0k 76 304.0k 45 180.0k sfd

1308 1 94 376.0k 76 304.0k 45 180.0k krsd

1305 1 148 592.0k 75 300.0k 60 240.0k dtrc

537 1 659 2.6m 75 300.0k 115 460.0k syslogd

1085 1 543 2.1m 70 280.0k 96 384.0k cron

2047 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

7295 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

15016 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

1953 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

2163 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

29731 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

7494 7491 795 3.1m 68 272.0k 128 512.0k rotatelogs

8527 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

29904 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

29510 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

6765 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

15125 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

29842 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

7489 7488 795 3.1m 68 272.0k 128 512.0k rotatelogs

7044 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

8759 1 795 3.1m 68 272.0k 128 512.0k rotatelogs

540 1 529 2.1m 64 256.0k 94 376.0k ptydaemon

1307 1 455 1.8m 59 236.0k 86 344.0k getty

1114 1 453 1.8m 59 236.0k 86 344.0k envd

556 1 462 1.8m 58 232.0k 91 364.0k nktl_daemon

1120 1 438 1.7m 42 168.0k 72 288.0k hubd

8337 8335 8 32.0k 8 32.0k 9 36.0k emagent

12104 1738 8 32.0k 8 32.0k 9 36.0k ia64_corehw

8338 8335 8 32.0k 8 32.0k 9 36.0k java



Total: 154591 603.9m 162908 636.4m

[Alzhy]

Looks normal...
OAS, java are really memory hogs.. If you[re wondering where your other memory went - then use glance -m.

Also, chances are your system has dbc_max_pct set to its default of 50%.. so your system probably has allocated about 2GB as your filesystem buffer cache.

So a "sysdef|grep dbc" ...

[Jeff Schussele]

Hi Jose,

I agree w/Nelson i.e. you need to check dbc_max_pct. On a 4GB system you should set it around 10% & then lower dbc_min_pct to around 2%.
Also I see a lot of HTTP,NFS, SNMP stuff running & if you aren't using them then turn off Apache, NFS & the SNMP stuff by stopping them with their /sbin/init.d/ stop scripts & edit their /etc/rc.config.d config files to prevent restarting them on next boot.
Also why do you have so many rotatelogs scripts running? Maybe they're hanging & chewing up memory as well.

Rgds,
Jeff

[Steven E. Protter]

The behavior is normal.

From the 10g docs:

Infrastructure requiers 1 GB of physical Ram

Business Intelligence(reports) and Forms 1 GB

Don't forget the database needs at least 512 MB and the OS a few hundred MB.

SEP

[Steven E. Protter]

My notes apply to the 10g application server, not the database.

J2EE and Webcache need another 512 MB
Portal and Wireless aanother 1 GB.

If its a database only question neither of my posts are relavent.

SEP

[Jose Manuel Fager Rojas]

Hello

The change the dbc_max_pct to 10% had any performance hit-down in the system?

The snmp service is not in use, I will shutdown that. But the others are in use.

Also how can I check that rotatelogs in the system?

Of the OAS 10g installation we left ON only the j2ee instance.

Nelson, Jeff, Steven. Thank you so much

UPDATE:
After the change in the Kernel we got a lot more memory as expected, now the system have 3813568 Kb of free ram !

Thank you again.