hdisk0 at 100% , system very slow , commands have 10-15 second delay

[strikelit]

Good evening, I wanted to know what settings can I check for slow system response. I enter a command and it takes about 10-15 seconds to come up. The system is very slow. I typed a command "topas" and seen that the rootvg disks hdisk0 and hdisk7 were at 100% all the time, What can I do ??

 

[mrn]

If your not up to speed with tunning get a copy of sarcheck. Evaluations are available, just google and apply.

Mike

"A foolproof method for sculpting an elephant: first, get a huge block of marble, then you chip away everything that doesn't look like an elephant."

http://www.airport-parking-site.co.uk/

 

[mrn]

Rod / P5

I know there isn't a magic wand approach to tuning, but am interested in methods / formulas you both use?

Trail and error?

I've been doing a lot of investigation into tuning over the past few weeks and have read about a formula, which was discussed at a IBM university conference recently but have googled and failed to find it.


Mike

"A foolproof method for sculpting an elephant: first, get a huge block of marble, then you chip away everything that doesn't look like an elephant."

http://www.airport-parking-site.co.uk/

 

[p5wizard]

There are SAP OSS notes about vmtune/vmo settings for an AIX ORACLE/SAP DB server. If I remember right, there's a 5/10/10% or 3/8/8% rule for minperm/maxperm/maxclient depending on the server's RAM size (but the ORA table buffer in the SGA needs to be large enough and have a good hitratio because you're taking away memory from LVM's file IO buffer, which is normally there to limit the amount of diskreads). But if your server uses memory for fileIO buffer and causes excessive paging for computational pages (like you experienced) then that's not doing your server perfomrance a lot of good.

As to pre-5.3 tuning or not, that's just for ease of use. With pre-5.3, you're on your own for making sure that the settings you apply will survive a reboot. (e.g. /etc/rc.tuning and an inittab entry you create yourself). With the new way of setting tuning parameters, you can specify NOW and/or NEXTBOOT, and the system will take care of setting params on next boot like you specify.



HTH,

p5wizard

 

[RodKnowlton]

vmo -p -o lru_file_repage=0
vmo -p -o maxperm%=80 -o maxclient%=80

Those are the values of maxperm% and maxclient% that you were already using. Even with lru_file_repage off and things working, you should set the max(perm/client)% values low.

From the IBM Technote on lru_file_repage:

Remember this is a “hint” to VMM and other conditions
can change the determination of what memory to steal;
therefore, don’t forget the other VMM tuning parameters
(most notable: set maxclient and maxperm to low values
and set strict_maxperm and strict_maxclient to soft
limits).

Rod Knowlton

IBM Certified Advanced Technical Expert pSeries and AIX 5L
CompTIA Linux+
CompTIA Security+