At 12:41 PM 8/19/02 +0300, Oded Arbel wrote:
yup, we have had this problem at times as well, it happens more if we try and shut bearerbox down from the console (with a break if running in foreground, or a kill).. We never get it happening when we shut bearerbox down via the http interface. When this does happen with us we generally have messages or acks queued in bbox or the smsc modules.
Nisan
Hi list.
I have this problem , that from time to time, bearerbox does not shutdown when receiving a signal or the command to shutdown from the HTTP interface. I think I tracked it down to smsboxc_run(), which is doing this :
<snip>
wait_for_connections(fd, run_smsbox, incoming_sms);
</snip>
now, wait for connections is doing this :
<snip>
/* XXX: if we are being shutdowned, as long as there is
* messages in incoming list allow new connections, but when
* list is empty, exit
*/
if (bb_status == BB_SHUTDOWN) {
ret = list_wait_until_nonempty(waited);
if (ret == -1) break;
}
</snip>
which basicly means that ig the incoming_sms list, for some reason, has a message in it, but no procuders, and bearerbox has been called to shutdown, and so closed all the threads that consume from that list, the smsboxc_run() will look endlessly waiting for someone to extract messages from that list, and thereby cause bearerbox not to shut down.
I'm still not clear on how to resolve this. comments anyone ?
--
Oded Arbel
m-Wise mobile solutions
[EMAIL PROTECTED]
+972-9-9581711 (116)
+972-67-340014
::..
I've found a solution to Fermat's last theorem, but I see I've ran out of room o
