Pumplerod, This is not normal, please send here the code (only what is needed no private data) to reproduce the problem so that it can be investigated. What is needed is: model controller view
mic 2013/2/5 pumplerod <pumple...@gmail.com>: > As a more complete test I did 3 things... > > 1) replicated what my function through web2py was doing in a separate script > and ran that thousands of times on the same server. This did not severely > impact memory usage. > > 2) set up a web2py function in my controller to simply return a json obect. > I am using the protocol: domain/app/controller/func.json No processing, > just a return statement. I then ran this script from a separate host. > Calling it 1000 times on multiple processors. The memory shot up right > away. > > 3) reran, from the separate host the same script only this time not > requesting a json object response. domain/app/controller/func This also > made the memory peak to a point of crashing the machine. > > is this a leak or do I need to specify some sort of cache for web2py? > > > On Monday, February 4, 2013 12:19:37 PM UTC-8, pumplerod wrote: >> >> To further elaborate. I am running my web2py instance on a VPS at >> dreamhost.com with a memory allocation of 4Gig. It is running under >> passenger. >> >> >> >> On Monday, February 4, 2013 12:16:38 PM UTC-8, pumplerod wrote: >>> >>> I had been having critical memory overflow problems and finally saw a >>> post about a memory leak in the version I was using (2.1.x) so I have >>> updated to the latest stable version (2.3.2), however I am still seeing my >>> memory climb until the machine crashes. >>> >>> I'm using web2py as my interface which is accepting data, parsing it, and >>> then appending to a neo4j database. I thought, for a while, that neo4j was >>> the culprit, however, after writing a script to emulate what web2py was >>> doing and running it on a separate host I saw no memory jump. It only >>> happens while running web2py, and only when I'm receiving many hundreds of >>> requests per min. There is nothing else running on the machine, other than >>> system tools. >>> >>> Within 5min I'm able to watch the memory go from a reasonable 730MB to >>> over 2500MB. >>> >>> I'm using the default SQLite database and am only making 1 or 2 calls to >>> the db upon each request. >>> >>> How can I track this down? Should I be looking for something >>> specifically in the logs? Use particular parameters for the logger? > > -- > > --- > You received this message because you are subscribed to the Google Groups > "web2py-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to web2py+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out. > > -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to web2py+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
