|
At 12:40 99-1-2 -0600, you wrote:
>Here's a run down of my network...
>
>all nodes in the farm (192.168.3.201 - 206) have the gw of 192.168.3.1
>
>the gw is 192.168.3.1 on the inside and 24.226.93.88 on the outside...
>
>When I use lynx to try and get at my web-page at 24.226.93.88 I get the
>following on my LAN... (tcpdump output)
>
>12:34:40.663748 arp who-has 192.168.3.206 tell 192.168.3.1
>12:34:40.663748 arp reply 192.168.3.206 is-at 0:20:94:1:9:d6
>12:34:40.663748 24.226.93.88.14492 > 192.168.3.206.80: S
>2090105203:2090105203(0) win 512 <mss 3544>
>12:34:40.663748 192.168.3.206.80 > 24.226.93.88.14492: S
>977916291:977916291(0)ack 2090105204 win 32736 <mss 1460>
>12:34:40.663748 24.226.93.88.14492 > 192.168.3.206.80: R
>2090105204:2090105204(0) win 0
>12:34:43.653748 24.226.93.88.14492 > 192.168.3.206.80: S
>2090105203:2090105203(0) win 31896 <mss 3544>
>12:34:49.653748 24.226.93.88.14492 > 192.168.3.206.80: S
>2090105203:2090105203(0) win 31896 <mss 3544>
>12:35:01.643748 24.226.93.88.14492 > 192.168.3.206.80: S
>2090105203:2090105203(0) win 31896 <mss 3544>
>
>There doesn't seem to be anything happening on the node...
>
In the virtual server via NAT, you cannot view the virtual server on the
real servers or on the load balancer, I mean that you can't lynx
24.226.93.88 on the server node or on the 24.226.93.88, because both
request packets and reply packets need to be rewritten.
Please try "lynx 24.226.93.88" on the others internet hosts except
24.226.93.88, see what happens.
>when I go directly to 192.168.3.201 - 206 (with lynx) they all work...
>
>Is there any other setup that needs to be done with IPmasquerading?
>-----Original Message-----
>From: Wensong Zhang <wensong@xxxxxxxxxxxx>
>To: Ron Harwood <harwoodr@xxxxxxxxxxxxxxxx>;
>linux-virtualserver@xxxxxxxxxxxx <linux-virtualserver@xxxxxxxxxxxx>
>Date: Saturday, January 02, 1999 9:54 AM
>Subject: Re: What am I doing wrong?
>
>
>>At 15:55 99-1-1 -0600, Ron Harwood wrote:
>>>...but it still doesn't work. :(
>>
>>I guess the problem is that you haven't set the default route of server
>>farm (from node1 to node6) to your load balancer. Because in the virtual
>>server via NAT, the reply packets from the real servers needs to be
>>rewritten back to the virtual IP address, otherwise the connection will be
>>broken.
>>
>
>
>
|
