On 10/7/05, Peter Bodik <[EMAIL PROTECTED]> wrote: > > Now, there is one thing to note -- if the packets are being > > sent every second, if one packet it lost, it should be noted > > there'll be another one in the next second. That said, > > because ganglia may send simple differential packets, if one > > gets lost, then the whole results recieved by e.g. the web > > interface may drift off the actual value. > > (That said, the full value is resent every hour or few hours or so, > > IIRC.) > > One lost packet can corrupt the data for several hours? This would make it > really unusable ... > > So does ganglia send differential packets?
If the range doesn't change by x over between when metrics are taken. If you're taking them every second, I doubt they'll change THAT much that it'll be resent, meaning it should use differentials unless I'm mistaken. Which is why (as I said above) if you need reliability, and massive data quantities you should use something like TCP. -- ~Mike - Just my two cents - No man is an island, and no man is unable.
