On Linuxy platforms, you can log and analyze native allocations with GCC's mtrace, which I've exposed to node here: https://github.com/Jimbly/node-mtrace
Unfortunately, with any C++ allocations (which includes Buffers), they're all attributed to a single call site, "operator::new", so your mileage may vary. If it's a rogue addon which is doing it's own memory allocations, it may help, or you may get lucky and something will jump out at you as suspicious. On Tuesday, September 11, 2012 1:51:40 PM UTC-7, wavded wrote: > > We have a large application that continues to grow in memory usage as the > day goes on (we restart it each day). I would like to not have to do that. > We have been optimizing the crap out of the JS code and continue to do so > and memory "seems" to be under control (or at least understood) in JS land. > > We have been using Ben's `heapdump` ( > https://github.com/bnoordhuis/node-heapdump) module to take snapshots in > production and analyze them offline. Right now one of our servers is > sitting at 1.1 GB (RSS) and I just took a heap snapshot and the dump size > is 75 MB.. I don't understand where this extra memory is coming from > (Buffers?, we do a lot of requests, a bad addon module?). Not sure how to > better debug this as the V8 tools don't seem to be helping. Any help > appreciated. > -- Job Board: http://jobs.nodejs.org/ Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines You received this message because you are subscribed to the Google Groups "nodejs" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/nodejs?hl=en?hl=en
