No direct way to do it.

If you "know" when the memory usage rises, you already don't need to use 
dtrace. Utility tools like vmstat, prstat and pmap can do the job for you.

You may be just fine running "prstat -s rss -c 1 > prstat.out &" in a shell and 
leaving it there for few days, and then a little perl to find out who were top 
hoggers in last 2-3 days.

After you have found the suspects, use libumem to find out if there is anything 
wrong with their memory allocation/deallocation or they are just need to use 
that much memory.

Keep in mind that reported memory usage figures may be highly misleading. You 
need to account for things like shared pages and reservations.


--
This message posted from opensolaris.org
_______________________________________________
dtrace-discuss mailing list
[email protected]

Reply via email to