THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.
The following task has a new comment added:
FS#385 - Possible memory leak when processing sht_reload User who did this - i (takeshi)
---------- Just in case you want to see them, I have re-uploaded the logs as: http://wikisend.com/download/342330/kamailio.log.gz
About latest master, I tested with it, but it failed in the first reload attempt. So I checked with cgdb and I can see you changed the code to use pkg mem instead of shm:
493│ nht.entries = (ht_entry_t*)pkg_malloc(nht.htsize*sizeof(ht_entry_t)); 494│ if(nht.entries == NULL) 495│ { 496│ ht_db_close_con(); 497├> return init_mi_tree(500, MI_ERR_RELOAD, MI_ERR_RELOAD_LEN); 498│ }
So, the problem now is that I don't have enough pkg memory as I am using default pkg memory size (4MB) and the reload will require about 25MB: (gdb) p (nht.htsize*sizeof(ht_entry_t)) $6 = 25165824
So I have rebuilt kamailio with 50MB of PKG memory: -#define PKG_MEM_SIZE 4 +#define PKG_MEM_SIZE 50 And after that I don't see the problem anymore:
root@LAB-TEST:sip-router# for i in $(seq 0 10000); do echo $i; kamctl fifo sht_reload domain;done ... abridged ... 9994 9995 9996 9997 9998 9999 10000
So all seems fine now. Thanks a lot. And happy new year! ----------
More information can be found at the following URL: http://sip-router.org/tracker/index.php?do=details&task_id=385#comment12...
You are receiving this message because you have requested it from the Flyspray bugtracking system. If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.