Hi,
First of all I thank all for your suggestions. However more and more
questions take shape in my head...
> Normal/Expected, yes. Some loads don't allow the processes
> to work, you need a kernel space redundancy solution or faster
> hardware :)
What do you mean on "kernel space redundancy solution"?
> > 2. The measurement values were also interesting: on the distributor
> > machine I measured 26000 packets/sec, as long as on the RSs I measured
>
> 26,000KP/s is typical performance result with LVS
> when PCI is 33MHz. With 66MHz PCI you can achieve far more.
Does the scipt show_traffic.sh measure speed in packet/sec or Kpacket/sec?
(It writes out packets/sec.)
> 500K is too high value, it creates high load for the
> routing cache which is usually tuned for ... max 16K entries,
> there are some parameters to tune the cache management but
> I'm not sure it can help for such load. Try with small number
> of spoofed client IPs, 10, 1000, 10000.
I admit. I wanted to have a "real pressure probe". :-)
> How much you need? There are no explicit limits. May be
> only in the hash table rows (there are 2 tables for virtual
> services):
32 perhaps. It depends on many things.
Thank you,
Peter Stefan
|