Originally posted by: sajid1

Hi,

hdisk3 shows 100% most of the time and performence go downn.Find below for your information.All data files and oracle home are located at vol01 and archive location is at vol02.I want some of the datafiles that are using in reading more I/O, will shift on vol02,can you suggest me if i do it performence will go up?. As per our AIX team it will not be effect they says vol01,vol02 and vol03 are logical volume.

$ lspv
hdisk0 00c33aa0621dd148 rootvg active
hdisk1 00c33aa0a5823689 rootvg active
hdisk2 none None
hdisk3 00c33aa02b95a0ab datavg active
hdisk4 00c33aa0c8ec5953 datavg active
$ lspv -l hdisk3
hdisk3:
LV NAME LPs PPs DISTRIBUTION MOUNT POINT
fslv00 553 553 00..217..83..100..153 /vol01
fslv01 360 360 218..00..00..94..48 /vol02
loglv00 1 1 00..01..00..00..00 N/A
fslv02 175 175 00..00..134..24..17 /vol03
$ lspv -l hdisk4
hdisk4:
LV NAME LPs PPs DISTRIBUTION MOUNT POINT
fslv00 332 332 00..191..141..00..00 /vol01
fslv01 5 5 00..00..05..00..00 /vol02
fslv02 25 25 00..00..25..00..00 /vol03
$ lspv -l hdisk0
hdisk0:
LV NAME LPs PPs DISTRIBUTION MOUNT POINT
hd10opt 1 1 00..00..01..00..00 /opt
hd1 4 4 00..00..04..00..00 /home
lg_dumplv 8 8 00..08..00..00..00 N/A
hd5 1 1 01..00..00..00..00 N/A
hd8 1 1 00..00..01..00..00 N/A
hd6 77 77 00..77..00..00..00 N/A
hd2 20 20 00..00..20..00..00 /usr
hd4 4 4 00..00..04..00..00 /
hd3 24 24 00..00..24..00..00 /tmp
hd9var 8 8 00..00..08..00..00 /var
$ lspv -l hdisk1
hdisk1:
LV NAME LPs PPs DISTRIBUTION MOUNT POINT
hd10opt 1 1 00..00..01..00..00 /opt
hd1 4 4 00..00..04..00..00 /home
lv02 7 7 00..07..00..00..00 /mkcd/cd_images
loglv01 1 1 00..01..00..00..00 N/A
hd5 1 1 01..00..00..00..00 N/A
hd8 1 1 00..00..01..00..00 N/A
hd6 77 77 00..77..00..00..00 N/A
hd2 20 20 00..00..20..00..00 /usr
hd4 4 4 00..00..04..00..00 /
hd3 24 24 00..00..24..00..00 /tmp
hd9var 8 8 00..00..08..00..00 /var
$ lspv hdisk3
PHYSICAL VOLUME: hdisk3 VOLUME GROUP: datavg
PV IDENTIFIER: 00c33aa02b95a0ab VG IDENTIFIER 00c33aa000004c00000001172
b95a130
PV STATE: active
STALE PARTITIONS: 0 ALLOCATABLE: yes
PP SIZE: 1024 megabyte(s) LOGICAL VOLUMES: 4
TOTAL PPs: 1089 (1115136 megabytes) VG DESCRIPTORS: 2
FREE PPs: 0 (0 megabytes) HOT SPARE: no
USED PPs: 1089 (1115136 megabytes) MAX REQUEST: 1 megabyte
FREE DISTRIBUTION: 00..00..00..00..00
USED DISTRIBUTION: 218..218..217..218..218
$ lsvg -l datavg
datavg:
LV NAME TYPE LPs PPs PVs LV STATE MOUNT POINT
loglv00 jfs2log 1 1 1 open/syncd N/A
fslv00 jfs2 885 885 2 open/syncd /vol01
fslv01 jfs2 365 365 2 open/syncd /vol02
fslv02 jfs2 200 200 2 open/syncd /vol03

Thanks

Originally posted by: alethad

Have you reviewed over the attached document to be sure you have your Oracle & AIX tuned to each other? Look at page 12 specifically. You don't need to change the last parameter unless you're using Oracle RAC.

I would highly recommend you do this also if you haven't already.

Originally posted by: sajid1

Many Thanks,

Find below output of vmstat -v

$ vmstat -v
4079616 memory pages
3870177 lruable pages
493358 free pages
1 memory pools
697573 pinned pages
80.0 maxpin percentage
20.0 minperm percentage
80.0 maxperm percentage
54.0 numperm percentage
2090021 file pages
0.0 compressed percentage
0 compressed pages
54.0 numclient percentage
80.0 maxclient percentage
2090016 client pages
0 remote pageouts scheduled
2039 pending disk I/Os blocked with no pbuf
691171 paging space I/Os blocked with no psbuf
2484 filesystem I/Os blocked with no fsbuf
0 client filesystem I/Os blocked with no fsbuf
160998 external pager filesystem I/Os blocked with no fsbuf
0 Virtualized Partition Memory Page Faults
0.00 Time resolving virtualized partition memory page faults
---It is not same as provided pdf say.Which command i have to run for checking rest of the parameters that is not mentioned in above as per pdf example strict_maxperm,strict_maxclient,lru_file_repage.
--Kindly update me how to change it?

Originally posted by: sajid1

Waiting your response,

Thanks,

Originally posted by: dukessd

What response are you expecting?

alethad already told you to check the link and tune your system as per page 12 of the link and you then posted back showing you have not.....go tune it!

Please make the changes and then after waiting a while for the system to settle down let us know if you still have the same problem - which you may well have because you are working hdisk3 very hard because of the single lp / pp loglv00.

Originally posted by: sajid1

Hi,

Sorry for delay.

I have performed just now as per your action plan. The parameters have been changed. Find below output.
As I read in below link documents that “ If numperm% is less than or equal to minperm%, file pages will not be stolen when RAM is required” Is there any issue with it because minperm% is less than numperm%.Suggest me.

The issue is almost same what I monitored since yesterday it is 100% but not frequently as before it was.

http://stix.id.au/wiki/Tuning_the_AIX_file_caches
$ vmstat -v
4194304 memory pages
3978334 lruable pages
1512024 free pages
8 memory pools
1003017 pinned pages
80.0 maxpin percentage
3.0 minperm percentage
90.0 maxperm percentage
6.5 numperm percentage
262463 file pages
0.0 compressed percentage
0 compressed pages
6.5 numclient percentage
90.0 maxclient percentage
262463 client pages
0 remote pageouts scheduled
854 pending disk I/Os blocked with no pbuf
1317943 paging space I/Os blocked with no psbuf
2484 filesystem I/Os blocked with no fsbuf
106543 client filesystem I/Os blocked with no fsbuf
4495 external pager filesystem I/Os blocked with no fsbuf
0 Virtualized Partition Memory Page Faults
0.00 Time resolving virtualized partition memory page faults
Thanks,

Originally posted by: sajid1

waiting your reply