[Linux-cluster] LVS, not so fun today...

David Brieck Jr. dbrieck at gmail.com
Thu Nov 9 20:38:09 UTC 2006


On 11/9/06, isplist at logicore.net <isplist at logicore.net> wrote:
> Hi there,
>
> > Do the ARP ping from anywhere. It might not be related at all to your
> > problem but ARP problems will cause you no end of headaches with
> > direct routing. The output should look something like this:
>
> Testing from ANY machine including one of the real web servers works but not
> from the load balancer, 192.168.1.52 (LB0);
>
> #arping 192.168.1.150
> ARPING 192.168.1.150 from 192.168.1.56 eth0
> Unicast reply from 192.168.1.150 [00:20:94:10:44:A5]  1.193ms
> Unicast reply from 192.168.1.150 [00:20:94:10:44:A5]  0.666ms
>
> > Also, from your Load Balancer, please post the output of "curl -i
> > http://192.168.1.92" and the other real server IPs you have (93 and
> > 94).
>
> They all output the same, as they should, so I'll just post the one;
>
> 192.168.1.52# curl -i http://192.168.1.92/
>
> HTTP/1.1 200 OK
> Date: Thu, 09 Nov 2006 20:17:03 GMT
> Server: Apache
> Last-Modified: Mon, 28 Aug 2006 01:03:47 GMT
> ETag: "4ef781-1679-7f8be2c0"
> Accept-Ranges: bytes
> Content-Length: 5753
> Connection: close
> Content-Type: text/html; charset=UTF-8
>
> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
> <html>
> <head>
> <title>Companions.com</title>
> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
> </head>
>
> <body bgcolor="666699" text="ffffff" link="666699" vlink="666699"
> alink="666699">
> <table width="100%" border="0" cellpadding="0">
>   <tr>
>     <td><div align="left"><font size="6" face="Verdana, Arial, Helvetica,
> sans-serif"><strong>Who
>         Are <font color="#99FF33">You</font> Looking
> For?</strong></font></div></td>
>     <td> </td>
>     <td> </td>
>   </tr>
>   <tr>
>

I went back and looked at this:

eth0      Link encap:Ethernet  HWaddr 00:20:94:10:44:A5
         inet addr:192.168.1.52  Bcast:192.168.1.255  Mask:255.255.255.0
         inet6 addr: fe80::220:94ff:fe10:44a5/64 Scope:Link
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
         RX packets:91631 errors:0 dropped:0 overruns:0 frame:0
         TX packets:8142 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:1000
         RX bytes:6004157 (5.7 MiB)  TX bytes:1402617 (1.3 MiB)

eth0:1    Link encap:Ethernet  HWaddr 00:20:94:10:44:A5
         inet addr:192.168.1.150  Bcast:192.168.1.150  Mask:255.255.255.255
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

and just wanted to make sure you are not setting up eth0:1 manually.
It will be added automatically by lvs. From the looks of that output
compared to what you have in your lvs config it looks like that is the
case.




More information about the Linux-cluster mailing list