|
On Fri, 3 Aug 2007, Dan Baughman wrote:
> The more hops out
the further away the client, the worse the problem?
> the more the issue presents itself. I am trying to by
> pass the loadbalancer in the problem case to see if it is even the issue.
>>> Just to follow up on this, the packet loss appeared to
>>> be an unrelated issue. If I connect directly to a
>>> realserver, the page loads fairly fast. If I connect
>>> through the loadbalancer ( and get the same real server)
>>> I see a lot of tcp retranmissions and such (as sniffed
>>> from the real server). I can't figure it out. The
>>> delays appear to be from the RTO (retrans timeout). I
>>> saw a patch from miccy soft about the mtu discovery and
>>> have installed, I also tried disabling mtu discovery.
following up on Graeme's suggestion (if it turns out not to
be a problem with the internet)...
turn the director into a passthrough router to one
realserver (turn ip_forward on, take down the VIP, have the
director advertise a route to the VIP eg use a host route).
That way you'll have the director hardware, cables in the
packet path. (I'm assuming the direct path
client<->realserver is OK, since direct connection to the
realserver has no problems).
is it with all realservers?
Joe
--
Joseph Mack NA3T EME(B,D), FM05lw North Carolina
jmack (at) wm7d (dot) net - azimuthal equidistant map
generator at http://www.wm7d.net/azproj.shtml
Homepage http://www.austintek.com/ It's GNU/Linux!
|
