Replication configuration/ Replica server

[SModugu]

What actually is Replica server ? this is same as actual production server that has application and copy to another physical servers?

then what is replica vaults? how the connection is made between Main site and the replica server?

I am just confused between Replica server and replica vaults. can you please explain me in detail.

@BenLoosli 

Or do you have any document that has the details of replication configuration and precisely explain the Master site and replica site?

[TomU]

Read this:

https://www.ptc.com/support/-/media/support/refdocs/Windchill_PDMLink/11,-d-,0/WindchillVaultReplication_123336.pdf?sc_lang=en&source=search

[TomU]

Think of it like this:

• All web communication from the remote client (Creo, stand-alone web browser, etc.) goes directly to the main Windchill server.
• Depending on a users preferences, some files can automatically be saved to or retrieved from the replica server, minimizing the amount of time it takes for these same files to travel from the remote client to the main server.
• New files that are saved to the replica (remote) server are automatically replicated (copied back to) the main server on some schedule.
• You cannot log directly into a replica server.

[SModugu]

• But where do we set user preferences to connect to the Replica or Main Windchill server?
• Does replica server would also have the same set up and hardware same as Main Windchill server?
• Replica server and Replica vault are same? @TomU 

[HelesicPetr]

Hi @SModugu 

preference is in a user Preference management or product /org / site level

The Replica server stores the replica vaults. 

PetrH 

[TomU]

To add to what @HelesicPetr said,

Hardware

I'm sure this is dependent on how many people are connected, but our replica servers don't have nearly the same level of hardware or load placed on them as our production server.  The Windchill process running on a replica server are minimal, and the CPU usage is almost nothing.

For comparison:

Production Server

Replica Server

Vault

This is configurable of course, but by default the only things in the replica vault are what the users at the remote site have requested, or what the users at the remote site have uploaded.  Think of the remote vault as being in between the users and the main server.  The first time someone at a remote site requests something from the main server, it's added to the remote vault and then delivered to the user.  All subsequent requests for that file from that user or any other user at that same site will be served directly from the replica vault.  It's essentially a file cache.  If every single file on the main server was requested by the people at the remote site, then yes, the replica vault would get quite large, nearing the size of the master vault.

[SModugu]

@TomU @HelesicPetr 

I have set up a new Replica server in india location where 70% of the clients requests come from.

so how do I make a connection between Master site and Replica site, what files need to be updated on what site?

it obvious that there should be some communication between Master site and Replica site for replicating the data.

[TomU]

---

@SModugu wrote:

...how do I make a connection between Master site and Replica site, what files need to be updated on what site?

---

Follow these steps in the help documentation:

https://support.ptc.com/help/windchill/r12.1.2.0/en/#page/Windchill_Help_Center/WCInstallConfigGuide/WCInstall_PostInstallFileServer.html#

Once the remote file server is configured and operational, change your personal 'Vaulting and Replication' preference to use this new replica server and test it out.  Make sure that when you add files to your workspace, or upload new files from your workspace, that you see them appear in the replica vault folders or the remote file server.

[SModugu]

Thanks! that's actually served the purpose.

I would also need clarity on below.'

What is Synchronization replication.

Is that some thing:

1. synchronization happens between Master site and Replica site 
2. Replication happens between Master vault and replica vault within each site?

[TomU]

From that same PDF:

Replication is the 'push' process from the server to a user upon their request.  It happens on demand, immediately.  (See pages 8 - 9.)

File synchronization is the 'pull' process of collecting all the objects that have been uploaded to the cache vaults and moving them to their correct destination.  This happens on a schedule.  (See pages 17 - 20.)

[SModugu]

1. Replication happens between Master site and Replica site 
2. synchronization happens between Master vault and replica vault within each site?

the above statements are somewhat justify what you were saying?

[TomU]

Not necessarily.  Don't get hung up on which vault or cache something is in.

Replication is simply the process of getting something from wherever it currently is to where the user who is asking for it is located.  It could be from the master to a remote file server, but it could also be from the remote file server back to the master (if someone is requesting it there and synchronization hasn't occurred yet.)

Synchronization is simply the process of taking new stuff that's been uploaded someplace other than where it's supposed to be mastered and moving it to wherever the master location is.  (Usually this is the 'main' site, but it doesn't necessarily have to be.)

Again, replication is on demand copying from place to another, and synchronization is scheduled copying from one place to another, each for different reasons.

[SModugu]

Thanks a tonne @TomU 

the document you have shared is utmost helpful to me.

would you also be kind enough to provide any document for cluster configuration that gives the detailed step step procedure for cluster environment set up

[HelesicPetr]

Hi @SModugu 

Check the help WCAdvDepClusterConfig

It can help.

PetrH