After increasing the sessions (as Rene suggested) and then changing code and recompiling (as Alejandro suggested), the issue is still occurring. I am not sure my system is running out of resources, since CPU time is available and memory is 16GB (which is pretty big I think). Here is the fresh piece of log, can someone please help if this could be improved anyhow:
2011-04-18 10:26:51 [31380] [1] PANIC: Too many concurrent allocations. 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox(gw_panic+0xc2) [0x80730e2] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox [0x8066af5] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox(gw_check_malloc+0x84) [0x8066b84] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox(octstr_create_from_data_real+0xed) [0x8076f9d] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox(octstr_duplicate_real+0x50) [0x80771b0] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox(octstr_format_valist_real+0x6a1) [0x8077941] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox(octstr_format+0x1b) [0x8077ffb] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox [0x8051c44] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox(mysql_save_msg+0x1b8) [0x8051ef8] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox [0x8050f64] 2011-04-18 10:26:51 [31380] [1] PANIC: sqlbox [0x8068b8b] 2011-04-18 10:26:51 [31380] [1] PANIC: /lib/libpthread.so.0 [0xd89832] 2011-04-18 10:26:51 [31380] [1] PANIC: /lib/libc.so.6(clone+0x5e) [0xcc8f6e] On Sun, Apr 17, 2011 at 2:32 PM, ha...@aeon.pk <ha...@aeon.pk> wrote: > I increased the max-connections from 1 to 20, but still it aborted on high > traffic load. > > Now I am going to try the method below. > > > On Sat, Apr 16, 2011 at 8:09 PM, Alejandro Guerrieri < > alejandro.guerri...@gmail.com> wrote: > >> Hrm, I think it's actually a gwlib error, on gwmem-check.c >> >> You could try increasing MAX_TAB_SIZE on gwlib/gwmem-check.c >> >> Search for this lines: >> >> /* 100 MB */ >> #define MAX_TAB_SIZE (100*1024*1024L) >> #define MAX_ALLOCATIONS ((long) (MAX_TAB_SIZE/sizeof(struct area))) >> >> Try putting 200 instead of 100 and see if it improves. >> >> You'll have to recompile kannel and sqlbox afterwards. >> >> Anyway, I suspect you're hitting the limits of your system, but can't hurt >> to give it a try ;) >> >> Regards, >> >> Alex >> >> On Sat, Apr 16, 2011 at 5:03 PM, Rene Kluwen <rene.klu...@chimit.nl>wrote: >> >>> Increase max-connections in group = mysql-connection. >>> >>> >>> >>> == Rene >>> >>> >>> >>> *From:* users-boun...@kannel.org [mailto:users-boun...@kannel.org] *On >>> Behalf Of *ha...@aeon.pk >>> *Sent:* Saturday, 16 April, 2011 15:34 >>> *To:* Ram Manohar >>> *Cc:* kannel users >>> *Subject:* Re: Issue with SQLBOX >>> >>> >>> >>> I tried it a lot of time, but same issue comes. I have to manually start >>> everything again. >>> >>> On Sat, Apr 16, 2011 at 6:33 PM, Ram Manohar <nr...@airhub.com> wrote: >>> >>> pls kill the sql box and start the sql box again... >>> >>> >>> >>> On Sat, Apr 16, 2011 at 6:57 PM, ha...@aeon.pk <ha...@aeon.pk> wrote: >>> >>> Hi, >>> >>> >>> >>> My SQLBOX is randomly aborting on high traffic. Following error is thrown >>> in logs: >>> >>> >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: Too many concurrent allocations. >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: sqlbox(gw_panic+0xc2) [0x80730e2] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: sqlbox [0x8066af5] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: sqlbox(gw_check_malloc+0x84) >>> [0x8066b84] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: sqlbox(msg_create_real+0x29) >>> [0x8054099] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: sqlbox(msg_unpack_real+0x2d) >>> [0x805422d] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: >>> sqlbox(read_from_bearerbox_real+0xe2) [0x8055132] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: sqlbox [0x8050619] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: sqlbox [0x805105b] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: sqlbox [0x8068b8b] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: /lib/libpthread.so.0 [0xd89832] >>> >>> 2011-04-16 18:21:47 [12759] [2] PANIC: /lib/libc.so.6(clone+0x5e) >>> [0xcc8f6e] >>> >>> >>> >>> As a result, SMSBOXes also get disconnected, since they are connected to >>> sqlbox. >>> >>> >>> >>> Can someone help in this regard? >>> >>> >>> >>> Hamza >>> >>> >>> >>> -- >>> >>> *Regards,* >>> >>> *RamManohar.N,* >>> >>> *AirHub Technologies,* >>> >>> *Hyderabad* >>> >>> >>> >>> >>> >> >> >
