On Wed, Jun 20, 2001 at 01:48:16PM +0200, Serge Maandag wrote: > both "ping -c5" and "ping -c5 -w5" give an endless supply of: > > 13:43:20.404258 > 217.103.244.253 > 217.103.245.56: icmp: echo request > 13:43:21.400132 > 217.103.244.253 > 217.103.245.56: icmp: echo request > 13:43:22.400118 > 217.103.244.253 > 217.103.245.56: icmp: echo request > 13:43:23.310074 > 217.103.244.253 > 217.103.245.56: icmp: echo request > 13:43:23.400111 > 217.103.244.253 > 217.103.245.56: icmp: echo request > 13:43:24.140069 > 217.103.244.253 > 217.103.245.56: icmp: echo request > 13:43:24.400128 > 217.103.244.253 > 217.103.245.56: icmp: echo request > 13:43:25.590070 > 217.103.244.253 > 217.103.245.56: icmp: echo request > 13:43:33.310086 > 217.103.244.253 > 217.103.245.56: icmp: echo request > > So both options seem to be ignored. > I've tried it on RedHat 7.0 and 6.0 > > Peter, or anyone, got any clues left? Upgrade to iputils-20001110-1 on redhat. Other versions of ping and/or their documentation have 'issues' on redhat. Even then, the man page isn't 100% descriptive about how long ping takes to exit under certain conditions. As an example of this, a ping -c 10 will take about 19 seconds if it gets back no packets at all ... you really need to read the source to ping to figure out how it behaves unfortunately. For the 'no packets received case', ping exits in (n+9) seconds IIRC. YMMV, --C - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org
