CiviCRM Community Forums (archive)

Hi,

I'm running a CentOS-based LVS 2-node cluster.  The virtual server is the "www" that the public accesses and the two nodes are "web1" and "web2."  Web2 is a varnish cache of web1.  All editing is done using the web1 URL.  CiviCRM is setup to use the www URL.

The public side of CiviCRM is working fine.  The issue is of course with CiviCRM configured as such with "www", the links on the admin side don't always work because I need them to go through "web1."  The problem I'm having makes sense because that's the way I have it configured but I'm wondering if there's a better or preferred way of setting this up in a clustered environment.

Thanks!