[OpenSER-Users] Dispatcher round robin....

[Borislav Ivandic]

Hi Daniel,

Thaks for your quick help. It works like a charm now, and round robin is
robust now. I tested against many servers, and it really balances doing its
best.

By the way, you mentioned new features for dispatcher. Is there any way I
can learn and test them?

Thanks again and regards,

Borislav Ivandic


007/8/21, Daniel-Constantin Mierla <daniel at voice-system.ro>:
>
> Hello,
>
> the bug was related to some new features added in dispatcher. Hopefully
> I fixed that, please take the latest svn trunk and try again. Thank you,
>
> Daniel
>
>
> On 08/21/07 19:49, Borislav Ivandic wrote:
> > Hi Daniel,
> >
> > Please see the backtrace:
> >
> > Core was generated by `openser'.
> > Program terminated with signal 11, Segmentation fault.
> > #0  0x00000000 in ?? ()
> > (gdb) bt
> > #0  0x00000000 in ?? ()
> > #1  0xb7d154b5 in ds_connect_db () at dispatch.c:353
> > #2  0xb7d1e3d7 in mi_child_init () at dispatcher.c:312
> > #3  0x080e355d in init_mi_child () at mi/mi.c:88
> > #4  0xb7cf6b96 in fifo_process (rank=0) at mi_fifo.c:197
> > #5  0x0809f49c in start_module_procs () at sr_module.c:561
> > #6  0x080736b9 in main (argc=1, argv=0x0) at main.c:695
> >
> >
> > Let me know if you need any other thing. By the way, originally I
> > tested Dispatcher round robin in 1.2.1 version.
> >
> > Thanks and regards,
> >
> > Borislav
> >
> >
> > 2007/8/21, Daniel-Constantin Mierla <daniel at voice-system.ro
> > <mailto:daniel at voice-system.ro>>:
> >
> >     Hello,
> >
> >     On 08/20/07 22:09, Borislav Ivandic wrote:
> >     > Hi Daniel,
> >     >
> >     > Apologize me for the delay, but it wasn't easy setting-up the lab
> >     > environment. Anyway, downloaded the devel version, and, after
> >     > installing, found a "Segmentation fault (core dumped)" log
> message.
> >     > Maybe it's my short experience working with devel versions, but
> >     wasn't
> >     > able to run openser. Please see below the last part of log:
> >     please send the backtrace, you got a coredump.
> >
> >     gdb /path/to/openser /path/to/corefile
> >     bt
> >
> >     Thanks,
> >     Daniel
> >
> >     >
> >     > Thanks in advance,
> >     >
> >     > Borislav Ivandic
> >     >
> >     > Log:
> >     >
> >     >
> >     > DEBUG:db_free_rows: Freeing 0 rows
> >     > DEBUG:init_mod_child: CHILD , rank=1, module=registrar
> >     > DEBUG:init_mod_child: CHILD , rank=1, module=nathelper
> >     > DEBUG:init_mod_child: CHILD , rank=1, module=uri_db
> >     > db_init: Connection 'mysql://root:.acantoflextel@localhost
> /openser'
> >     > found in pool
> >     > DEBUG:init_mod_child: CHILD , rank=1, module=dispatcher
> >     > Aug 20 17:40:11 [3897] DBG:dispatcher:child_init:  #1 / pid <3897>
> >     > Segmentation fault (core dumped)
> >     >
> >     >
> >     >
> >     >
> >     > 2007/8/15, Daniel-Constantin Mierla <daniel at voice-system.ro
> >     <mailto:daniel at voice-system.ro>
> >     > <mailto: daniel at voice-system.ro <mailto:daniel at voice-system.ro>>>:
> >     >
> >     >     Hello,
> >     >
> >     >     can you get latest devel version (trunk) and give it a try?
> >     I did some
> >     >     fix, hopefully. Please report the results.
> >     >
> >     >     Cheers,
> >     >     Daniel
> >     >
> >     >
> >     >     On 08/14/07 23:32, Francisco Ivandic wrote:
> >     >     > Hi All,
> >     >     >
> >     >     > I'm using dispatcher succesfully, with round robin alg
> (option
> >     >     4), and
> >     >     > also use ds_mark_dst and ds_next_dst to select other proxy
> in
> >     >     case of
> >     >     > failure. It balances great when all servers are active,
> >     but when
> >     >     there
> >     >     > are inactive servers, round robin alg selects the next in
> >     the round,
> >     >     > even if it's marked as inactive. This results in going
> through
> >     >     > ds_next_dst, and overloads the first active server in the
> >     >     dispatcher
> >     >     > file until completing the round. Then, other active
> >     servers are
> >     >     > called, and when passing to a group of inactive servers,
> >     starts
> >     >     going
> >     >     > through the first active (as much times as inactive
> >     servers there
> >     >     > are), and again completing the round.
> >     >     >
> >     >     > Does anyone know a way to implement a round robin aware of
> >     inactive
> >     >     > servers?
> >     >     >
> >     >     >
> >     >     > Thanks in advance,
> >     >     >
> >     >     > Borislav Ivandic
> >     >     >
> >     >
> >
> ------------------------------------------------------------------------
> >     >     >
> >     >     > _______________________________________________
> >     >     > Users mailing list
> >     >     > Users at openser.org <mailto:Users at openser.org>
> >     <mailto:Users at openser.org <mailto:Users at openser.org>>
> >     >     > http://openser.org/cgi-bin/mailman/listinfo/users
> >     <http://openser.org/cgi-bin/mailman/listinfo/users>
> >     >     >
> >     >
> >     >
> >
> >
> > ------------------------------------------------------------------------
> >
> > _______________________________________________
> > Users mailing list
> > Users at openser.org
> > http://openser.org/cgi-bin/mailman/listinfo/users
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sip-router.org/pipermail/sr-users/attachments/20070821/a6a13b04/attachment.htm>