qnx_name_locate

I have a server type component on a node that talks to other processes on
other nodes. If the Send() fails -1, the next time, I do a qnx_name_locate
on the expected process (global name). The problem is that this takes too
much time to return (1-2 seconds). Is there some other method that I could
use to see if that name exists OR see if the node actually exists?

Thanks

“Doug Rixmann” <rixmannd@rdsdata.com> wrote in message
news:ahrl1h$eu6$1@inn.qnx.com

I have a server type component on a node that talks to other processes on
other nodes. If the Send() fails -1, the next time, I do a qnx_name_locate
on the expected process (global name). The problem is that this takes too
much time to return (1-2 seconds). Is there some other method that I could
use to see if that name exists OR see if the node actually exists?

No there is no real simple solution. Method I know is to either play with
netpool value but this can create other problem. You could write a netraw
application or use UDP to make your own watchdog thingy

Thanks

Mario,

Phil Wilson (Thomson Multimedia) here
I’m trying to reach you but your email has changed…
Will you please contact me?


“Mario Charest” postmaster@127.0.0.1 wrote in message
news:ahrum8$le4$1@inn.qnx.com

“Doug Rixmann” <> rixmannd@rdsdata.com> > wrote in message
news:ahrl1h$eu6$> 1@inn.qnx.com> …
I have a server type component on a node that talks to other processes
on
other nodes. If the Send() fails -1, the next time, I do a
qnx_name_locate
on the expected process (global name). The problem is that this takes
too
much time to return (1-2 seconds). Is there some other method that I
could
use to see if that name exists OR see if the node actually exists?


No there is no real simple solution. Method I know is to either play with
netpool value but this can create other problem. You could write a netraw
application or use UDP to make your own watchdog thingy

Thanks
\