Curphyhost and a secondary backup network

Hi Pier,

By definition "curphyhost" stands for current physical host, meaning that NetWorker will find out which of the nodes is currently the "active" one, and that will be the one sending data to the devices, if these are correctly configured. However this only applies to virtual clients.

Let's say you have a cluster as follows:

Physical node 1=phn1

Physical node 2=phn2

Virtual client=vtclnt

In this case, you should be configuring a client resource for the virtual client called vtclnt, and add curphyhost in the storage node definition:

Note: curphyhost is only applicable to virtual clients and should never be inserted into the clone storage nodes list even for the NetWorker Server’s NSR resource.

In NetWorker Administrator's guide for NW 7.6.2, page 497 and 498 you'll find a good explanation about the usage of curphyhost.

Thank you.

Carlos.

Carlos,

I agree with you, but what about my question? What can I do if I have a secondary backup network ? It is a little bit different from the normal configuration-

Thanks for any suggestion

PierPa

Hi Pier,

I would say you have different options., however the first approach would be to set both names into the alias field:

nodeA

nodeA.domain.com

nodeA_bck

Same with node B.

You can also use routing to send the data over the right interface, so let's say you gather the data from clients through management interface (nodeA) and then send the data to the Storage Nodes through the bck Interface (nodeA_bck)

There are other options, such as using hosts file to set the name resolution.

I think the best approach is always to create client entries with the bck interface, is that is the interface dedicated to backup, however you will probably need to add all aliases into the alias field, and then play around with hosts file and/or DNS.

Is this a correct answer for you or you are more concern about the cluster stuff?