[Devel] OpenSER 1.1 on NetBSD 3.0 + SEMS

[Bogdan-Andrei Iancu]

Hi Piyush,

the error is not related to the memory used by the openser processes. 
It's about the buffer which is associated to the unix socket. The error 
reports a full buffer which can happen because:
    1) nobody is reading from that socket
    2) the buffer is a little bit small and it cannot handle large write 
operations - this may explain why it works on Linux; maybe the default 
value for this buffer (which is sytem controlled) is larger than on netBSD.

regards,
bogdan

Piyush Devnani wrote:

>Hi Bogdan,
>
>Thanks, now it compiles properly.
>
>Once I tested the same with latest SEMS, The IPC between
>SER <---> SEMS seems to run out of memory. I get the 
>following error multiple times.
>"
>ERROR:tm:tsend_datagram_ev:failed to send: (55) No buffer space available
>ERROR:tm:write_to_unixsock: writev failed: No buffer space available
>ERROR:tm:t_write_unix: write_to_unixsock failed
>"
>I tried the same on SuSe Linux 9.1, and it worked fine. This problem 
>might be specific to NetBSD (3.0)
>
>regards,
>piyush
>
>  
>
>>----- Original Message -----
>>From: "Bogdan-Andrei Iancu" <bogdan at voice-system.ro>
>>To: "Piyush Devnani" <piyush_dev at mail.com>
>>Subject: Re: [Devel] OpenSER 1.1 on NetBSD 3.0
>>Date: Mon, 17 Jul 2006 16:55:49 +0300
>>
>>
>>Hi Piyush,
>>
>>this problem was reported and fixed couple of days ago on the devel 
>>version - I just did a backport to 1.1.0 on CVS. Please update and 
>>the problem should go away.
>>
>>regards,
>>bogdan
>>
>>Piyush Devnani wrote:
>>
>>    
>>
>>>Hi,
>>>
>>>I tried to compile the latest copy of OpenSER from cvs on NetBSD 
>>>3.0(i386) with gnu make 3.81. The compilation
>>>errors were::
>>>"
>>>In file included from socket_info.h:41
>>>                from cfg.y:92
>>>ut.h:In function 'strlower':
>>>ut.h:424: warning: subscript has type 'char'
>>>cfg.y: In function 'yyparse' :
>>>cfg.y:859: error: 'IPTOS_MINCOST' undeclared (first use in this function)
>>>cfg.y:859: error: (Each undeclared identifier is reported only once
>>>cfg.y:859: error: for each function it appears in.)
>>>"
>>>Did any one else faced similar problems ?
>>>
>>>Thanks in advance,
>>>Piyush
>>>
>>>
>>>
>>>
>>>
>>>      
>>>
>
>  
>
>
>
>  
>