Re: recv() problem?
- From: "FLY135" <fly_135@xxxxxxxxxxx>
- Date: 10 Jan 2006 06:55:58 -0800
David Schwartz wrote:
> "Hans-Juergen Lange" <Hans-Juergen.Lange@xxxxxx> wrote in message
> news:42fm62F1ia7hjU1@xxxxxxxxxxxxxxxx
>
> > in the original post the server closes the communication after sending the
> > data.
>
> The original post didn't say that, and I don't think it's true.
>
> > This is the indication on the client side for completion.
>
> No, the indiciation on the client side for completion is whatever the
> protocol says it is. That's not a TCP issue, it's an issue for the
> higher-level protocol.
I use a timeout on my recv because in my application there is no way to
know when the connection will be closed (or simply disappear). Loss of
connectivity or device reboot (it's embedded) may cause a loss of
connection. You can send some dummy data back as a check to see if the
connection is still alive after a timeout.
In a way it's funny that for as long as TCP has been used the most
common problems are still left for the user to discover and solve over
and over.
> > There are lots of advantages in using select and/or poll.
>
> I agree, but none of them have to do with the original poster's issue.
Other than seeming to be only able to interact by reposting the same
question over and over, nobody really knows what really is the OP's
issue.
.
- References:
- recv() problem?
- From: michal . shmueli
- Re: recv() problem?
- From: Hans-Juergen Lange
- Re: recv() problem?
- From: David Schwartz
- Re: recv() problem?
- From: Hans-Juergen Lange
- Re: recv() problem?
- From: David Schwartz
- recv() problem?
- Prev by Date: Re: linux 2.6.15 breaks route to outside world?
- Next by Date: Recommendations for powerline ethernet adapters
- Previous by thread: Re: recv() problem?
- Next by thread: recv() problem?
- Index(es):
Relevant Pages
|
Loading