[Spread-users] Sudden freezes in message delivery using C API
Doug.Palmer at csiro.au
Mon May 28 22:41:14 EDT 2007
On Wed, 2007-05-23 at 11:25 -0400, John Schultz wrote:
> The quickest diagnosis then would be to simply lower the Hurry_timeout and
> see if that affects your hiccup timings.
Sure enough, lowering Hurry_timeout lowers the freezes to the same
amount of time as the timeout.
> Another potentiality is that you have a bad network hardware that is
> causing lots of loss on a particular link. You can look at the output of
> the spmonitor prgram when it is reporting status on all the machines and
> if you see lots of retrans, then you might have a network problem that
> Spread is overcoming with some work.
The spmonitor program shows increasing numbers of s-retrans on whatever
machine is sending the messages. Only certain combinations of
send-receive pairs and network setups seem to produce this effect and
I'm not sure what the relationship is (for example, sending A->WAN->B
produces freezes, sending A->WAN->Switch->B doesn't).
I'm not sure what the difference between u-, s- and b-retrans is.
More information about the Spread-users