On Fri, Feb 13, 2009 at 01:59:14PM -0700, Alex Rousskov wrote: > Sorry, I do not know much about pprof. The graph shows that 200MB out of > 500MB was allocated to close callbacks for deferred reads (the current > suspect), right? Is there an indication that those objects were actually > leaked?
Well, delayRead() and the stuff it calls stands for 493.8MB. 201MB of those do indeed go to close callbacks. > How long did this test run, and does the 200MB/test_time memory > consumption speed correlate with what you have observed in longer tests? This was a few hours. And yes, it seems to run pretty linearly. > Finally, the other 300MB are due to pooled allocations. Those are > tracked and reported by cache manager. Do you see any abnormal memory > stats when you query your longer-running Squids? I'm not entirely sure what stats you're looking for. Almost all of the overconsumption in the pools (300MB or so) originates in delayRead(), though. /* Steinar */ -- Homepage: http://www.sesse.net/
