hi guys
We are configuring Vmware Replication latest version 5.5.1,
This is the setup.
We are using a administration network (eth0) you know where web and normal client connects and a replica network (eth1) the idea is to separate loads
So:
Site A. Source.
vCenter: eth0=10.7.111.240, eth1: 10.7.119.150
vReplica: eth0=10.7.110.120, eth1: 10.7.119.151
Site B. Destination
vCenter: eth0=10.12.25.33, eth1: 192.168.122.150
vReplica: eth0=10.12.25.133, eth1: 192.168.122.151
We setup the vReplica in both sites but when pairing Site A with Site B, we get the error below.
VRM Server generic error. Please check the documentation for any troubleshooting information. The detailed exception is: 'org.apache.http.conn.HttpHostConnectException: Connection to http://10.12.25.33:80 refused'
Can you help me with this?
I don't get why the vCenter-vReplica are trying to contact 10.12.25.33 when it can contact the vCenter using the 192.168.122.150
thanks a lot
Are you using FQDN's or IP's when pairing? If FQDN, do the appliances and/or vcenter servers have the appropriate DNS records?
Have you tried using literal addresses?
I'm assuming that your network is setup appropriately. Can you ping across the networks to the appropriate IP's?
I am using IPs for the vCenter Pairing to avoid any DNS issues
I have to google about "literal addresses"
Keep in mind that the actual replication traffic is sent to the replication target via the primary management interface on your ESXi host. You may want to check out this article and rethink your configuration, as it may be complicating things unnecessarily while not really achieving the separation you're after.
Using Secondary Management Network for vSphere Replication | VMware Consulting Blog - VMware Blogs
A literal address = IP address
thanks blabarbera
interesting article
well in my configuration below
Site A eth0 does not reach Site B eth0, this is by design we were thinking all the communication was going to be done in the eth1, so eth1 in Site A would pair with eth1 in Site B, that's when thins get weird since when just trying to pair the vCenter in the web client I get the error and in the client says a little more
VRM Server generic error. Please check the documentation for any troubleshooting information. The detailed exception is: 'org.apache.http.conn.HttpHostConnectException: Connection to http://10.12.25.33:80 refused'
Site A. Source.
vCenter: eth0=10.7.111.240, eth1: 10.7.119.150
vReplica: eth0=10.7.110.120, eth1: 10.7.119.151
Site B. Destination
vCenter: eth0=10.12.25.33, eth1: 192.168.122.150
vReplica: eth0=10.12.25.133, eth1: 192.168.122.151
Hi,
The VR UI plugin uses the value at https://vcenter_address/mob/?moid=ExtensionManager&doPath=extensionList[%22com.vmware.vcHms%22].serv...
The vCenter address is taken by default from the VirtualCenter.FQDN setting.
You need to have access to the vCenter reverse proxy of the remote site for the replication management commands.
Network traffic isolation for vSphere Replication using separate NICs is on the roadmap for a future release. There is no official support for multiple NICs of the VR appliance in current releases. For current releases you might be able to isolate replication data traffic using ports 31031 and 44046.