2 EPO Servers 1 Master 1 Fallback?

[jatkinson]

Is it possible to have 2 EPO servers on a network, the 2nd server purely to provide a 2nd means of updating the clients?

I was thinking it could be done by creating a distribution repository on EPO_SERVER2 and then creating a fallback repository on EPO_SERVER1 that points to the path where the distribution repository is. Therefore the Repository List would be:

EPO_SERVER1
Distribution repository on EPO_SERVER2
NaiFTP (not used by clients as no access so can be removed)

The only problem I have with this solution is the extra work involved in creating the distribtion repository and the replication tasks. I would prefer a solution where the repository list is simply:

EPO_SERVER1
EPO_SERVER2

So I'd only need to set up pull tasks on both my EPO Servers. Is it possible to do this? I was thinking that it may be possible via rewriting the sitelist.xml file to include details of the 2nd EPO Server.

Has anyone done this or does anyone have any ideas?

Cheers

 

[qtin]

I watched a webcast where somebody asked a question along the same lines - the response was, "Why do you want to do this?"

I assume you are talking about 3.0.x. Why not just setup a distributed repository and be done w/ it? Are you looking for faster response times from wakeup calls?

I'm not saying it can't be done - I'm just not sure if you want to. There are enough smart people out there that we should be able to find an easier solution.

So - what are you looking to gain?

 

[jatkinson]

I could do, I just don't see the point in the extra work if the repository list can be made to simply point at each server.

I don't have a need for a distribution repository if I have just one EPO server so by extension why have one if I have 2 EPO servers? That is, unless there is no other way.

 

[b0pp]

No you can't make an Epo Agent connect to some sort of a Failover Epo Server. The sitelist.xml merely defines where the agent should retrieve it's software packages from, not where the Epo server is. So basically, what you want is not possible.

Furthermore, qtin has a point by asking why you want to have 2 Epo servers? You would have to install Epo all over again on another server, which in my opninion is far more work than defining a Distributed Repository...

A Distributed Repository is just a UNC, HTTP or FTP share. If you define such a share through Repository > Add distributed repository, it will be stated as a repository in the ePolicy Orchestrator Agent automatically, and will be used as such too.

Than, if u define a Scheduled replication task that runs right after your Scheduled pull task, everything will be the same on both servers.

If you're Epo server should crash, you can update the Distributed Repository automatically using Autoupdate Architect.

The NaiFtp server is stated as being the fallback because that's it's default role. You can change this to the NaiHttp server through Repository > Source Repositories.

Really, you don't want another Epo server, just set up your Distributed Repositories the way you want to. Your Epo server can be down forever from that moment, Epo agents will work using the Policies they've saved locally...

GL

Erik

 

[jatkinson]

Hi there, thanks for the advice

I think the point you made re: using Autoupdate Architect (AA) to update the Dist. Rep. , (which I wasn't aware of) has laid to rest my biggest concern, which was that if the EPO Server is down how does my system stay up-to-date.

Can you give more info as to how to use AA or point me somewhere where I can research it further.

The 'new' EPO server is already built anyway as I was originally going to decommission the old EPO Server.