Persistance and LVS
horms at verge.net.au
Fri Apr 30 03:36:51 BST 2004
On Thu, Apr 29, 2004 at 04:12:38PM -0400, Joseph Mack wrote:
> Peter Mueller wrote:
> > Also for network:
> > Since you are doing NAT,
> if you're doing LVS-NAT then you have a smaller limit for
> ports, since all ports are coming from the director.
> when NAT'ing you only have ports 61000-65xxx, ie 4000 ports to
> choose from.
> Do the (small) number of ports limit LVS-NAT? If the
> LVS has 40,000connections/hr, with 1hr persistence, can the
> LVS-NAT director hold 40000 connections once (for discussion,
> let's say these are all to different IPs, ie 40,000 IPs).
I don't think that is the case. Even when you use LVS-NAT
the source IP address of the packets is that of the end-user.
(Actually, I used to think this was the case, but I have
subsequently convinced myself otherwise)
That asside, this conversation started as a discussion of
persistance templetes. LVS does not use the source port at all
in persistance templates. In fact it is set to 0 internally
so LVS can differentiate between a persistance template and
a connection entry.
More information about the lvs-users