Date: Mon, 23 Mar 2009 00:19:57 -0700 From: perryh@pluto.rain.com To: ota@j.email.ne.jp Cc: freebsd-hackers@freebsd.org, rwatson@freebsd.org Subject: Re: 2 uni-directional TCP connection good? Message-ID: <49c7381d.eJH7/fiaDJB9Gr6c%perryh@pluto.rain.com> In-Reply-To: <20090322235253.432874dd.ota@j.email.ne.jp> References: <20090320045319.04484fc5.ota@j.email.ne.jp> <alpine.BSF.2.00.0903201321570.48549@fledge.watson.org> <20090322235253.432874dd.ota@j.email.ne.jp>
next in thread | previous in thread | raw e-mail | index | archive | help
> What I meant was that there were cases when a receiver could not > tell weather no data was coming or communication was interrupted. > Once connection is established, a route is available between a > server and a client. Let's say this route is broken for some > reasons, i.e. someone unplugged a cable or a firewall started > dropping or rejecting between these server and client, a sender > may not notice as soon as it happens but at least, a sender knows > a massages was not delivered right. On the other hand, receiver > side does not have any idea that a message delivery failure has > happened at all or for a while unless using heartbeat messages > in upper layer. KEEP_ALIVE option seems to be implementation > dependent such that you cannot assure TCP connection availability > for every minute. The whole point of TCP (vs IP alone, or UDP) is to establish reliable end-to-end communication over unreliable underlying links. If a packet is corrupted or lost, it gets resent. If a route goes down, and an alternate is available, TCP will -- eventually -- find it and recover. If the last (or only) route goes down, TCP will in principle wait indefinitely for a route to become available, whether by reestablishment of the original or provision of an alternative. So you are correct that a receiver can't tell the difference between a loss of connectivity and the sender having crashed, however the situation is entirely symmetric: the sender can't tell the difference either. It all gets sorted out when communication is reestablished; at that point traffic will resume (if the link had been down) or the uncrashed end will get a connection reset (if its peer had crashed). The practice of sending keep-alive packets simply converts a temporary (thus potentially recoverable) communication loss into what amounts to an unrecoverable crash of whichever end gets impatient first.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?49c7381d.eJH7/fiaDJB9Gr6c%perryh>