nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

How to find out memory usage

Shakti_ThePower

How to findout memory usage of each brach and sub branch in Solaris.

TIA
Richa.

Find more posts tagged with

Comments

Migrateduser

My understanding is that there is no way to get a precise measure of this, but there are formulas that can be applied. I believe that information is available through the Support site, but you could START by doing a "du -ks" on each branch, but that does not give you the TS overhead, of course.

Dave

Shakti_ThePower

If I use df -ks BranchName then it is calculating each edition has full memory. Example if workarea memory has 100MB and 100 Editions then it is give as 100X100 MB. I think it is not correct,am I right?

TIA
Richa,

nipper

I assume you mean disk rather than memory. Memory per branch makes no sense.

That being said, if you use multistore, you can figure out disk per branch,
but there is no good way of figuring it out.

I have done the du -s for the Staging area to get a snapshot of one instance,
but that does not help if you have many versions, esp of large files.

HTH

Andy

nipper

>If I use df -ks BranchName then it is calculating each edition has full memory. Example if workarea memory has 100MB and 100 Editions then it
>is give as 100X100 MB. I think it is not correct,am I right?

correct that is totally meaningless

Adam Stoller

The attached script will, I believe, come *close* to providing you a sense of how much diskspace is being used by each branch.

I whipped this up quickly - and haven't thoroughly tested it - but ... hey it's free ;-)

--fish
Senior Consultant, Quotient Inc.
http://www.quotient-inc.com

nipper

Good code, does not work for an end branch, but for the higher level branches it is great.

i.e. /iwmnt/default/main/company/internal/emp_portal will return 0

run as:

./brusage.ipl -b /iwmnt/default/main/company/internal/emp_portal

but will work with:

./brusage.ipl -b /iwmnt/default/main/company/internal/

I will play with that.

Andy

nipper

Update, been running this for 2 hours, still counting the external site (1 1/2 years worth, 40K objects,
using most of the 20 GB backingstore)

Andy

herald10

Andy,
how long did the script take to finish? At our site we have two branches and around 12GB content. So just wanted to get an idea. Its been running for the past 20 minutes.

Thanks
-H

nipper

One ran for a 1 GB branch in a few minutes. The large branch never completed (after about 4 hours), I ran it again 1st thing this
am. Will see.

Andy

Adam Stoller

The script stores everything in memory and perhaps it's getting too big.
You might need to modify the script dump periodic data into a temporary file so as to not do everything in memory.

Like I said, I whipped it up and didn't have a real-life full system to test it on.

--fish
Senior Consultant, Quotient Inc.
http://www.quotient-inc.com

nipper

I was wondering that but it is not chewing up too much memory:

PID USERNAME THR PRI NICE SIZE RES STATE TIME CPU COMMAND
17103 rc572c 1 18 4 17M 16M cpu0 73:21 12.01% brusage.ipl

Been running 2 1/2 hours & I cannot imagine that it would be faster if I dumped the data to a file.

Andy

Adam Stoller

Then feel free to add some debugging to the code - at one point I had it printing out all sorts of stuff so I could track what was happening.

--fish
Senior Consultant, Quotient Inc.
http://www.quotient-inc.com