I should point out that I used the Git version of APE for all my tests. Also, I just ran Valgrind, and it reported a large amount of illegal reads. The summary states that several blocsk were in fact definitely lost, and even more were indirectly lost.
Example violation: ==12010== Invalid read of size 4 ==12010== at 0x8058145: get_pipe (in /usr/bin/aped) ==12010== by 0x80581B8: get_pipe_strict (in /usr/bin/aped) ==12010== by 0x80590C4: post_to_pipe (in /usr/bin/aped) ==12010== by 0x804F3EA: cmd_send (in /usr/bin/aped) ==12010== by 0x804FC21: process_cmd (in /usr/bin/aped) ==12010== by 0x805002F: checkcmd (in /usr/bin/aped) ==12010== by 0x804E6FF: checkrecv_websocket (in /usr/bin/aped) ==12010== by 0x805C6AB: parser_ready_websocket (in /usr/bin/aped) ==12010== by 0x8056D37: process_websocket (in /usr/bin/aped) ==12010== by 0x804D98E: sockroutine (in /usr/bin/aped) ==12010== by 0x804CB64: main (in /usr/bin/aped) ==12010== Address 0x4391110 is 32 bytes inside a block of size 33 alloc'd ==12010== at 0x4005903: malloc (vg_replace_malloc.c:195) ==12010== by 0x8057B40: xmalloc (in /usr/bin/aped) ==12010== by 0x805333B: json_callback (in /usr/bin/aped) ==12010== by 0x8054EF0: parse_parse_buffer (in /usr/bin/aped) ==12010== by 0x8055746: JSON_parser_char (in /usr/bin/aped) ==12010== by 0x8052FD9: init_json_parser (in /usr/bin/aped) ==12010== by 0x804FFF3: checkcmd (in /usr/bin/aped) ==12010== by 0x804E6FF: checkrecv_websocket (in /usr/bin/aped) ==12010== by 0x805C6AB: parser_ready_websocket (in /usr/bin/aped) ==12010== by 0x8056D37: process_websocket (in /usr/bin/aped) ==12010== by 0x804D98E: sockroutine (in /usr/bin/aped) ==12010== by 0x804CB64: main (in /usr/bin/aped) ==12010== LEAK SUMMARY: ==12010== definitely lost: 47,473 bytes in 813 blocks ==12010== indirectly lost: 286,831 bytes in 568 blocks ==12010== possibly lost: 11,308 bytes in 13 blocks ==12010== still reachable: 20,850 bytes in 97 blocks ==12010== suppressed: 0 bytes in 0 blocks On Tue, Nov 2, 2010 at 3:42 PM, Felix Filozov <[email protected]> wrote: > I'm encountering problems with APE where memory is not being released. The > first test I ran was having 2000 browsers connect to APE within 5 minutes. > That used up about 20% of memory. After all the clients have been > disconnected, it remained at 20%. I thought, perhaps my custom commands were > at fault. So, I took the Git version (just removed nickname.js) and ran the > same test, and the results were the same. The end result seems to be > independent from the number of browsers accessing APE. However little memory > is allocated, it is not being released. > > Any thoughts? > > - Felix > -- You received this message because you are subscribed to the Google Groups "APE Project" 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/ape-project?hl=en --- APE Project (Ajax Push Engine) Official website : http://www.ape-project.org/ Git Hub : http://github.com/APE-Project/
