On Sat, May 15, 2010 at 04:22:54PM -0700, tester wrote: > this is now solved. Jarod in an earlier post suggested a bigger > number of buffer size and string size. I had to go all the way to > jstack(160,10240) to a full stack. My testing increments were power of > 2, so a smaller may have worked. > > Does this mean on a 5440 this becomes a memory hog, since there is a > per CPU buffer allocation.?
These are allocated out of the per-CPU transaction and aggregation buffers, so they won't cause any additional allocations. Since they use more of those buffers per entry, they may induce drops, which you would have to increase the buffer sizes to avoid. Cheers, - jonathan _______________________________________________ dtrace-discuss mailing list dtrace-discuss@opensolaris.org
