# Clustering **URL:** \ **Category:** Archived Topics\ **Created:** [October 20, 2009, 12:37pm UTC](https://forum.inductiveautomation.com/t/clustering/1343 "2009-10-20T12:37:40Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![Step7](https://avatars.discourse-cdn.com/v4/letter/s/45deac/32.png) [@Step7](https://forum.inductiveautomation.com/u/Step7)\ **Post date:** [October 20, 2009, 12:37pm UTC](https://forum.inductiveautomation.com/t/clustering/1343/1 "2009-10-20T12:37:40Z")
Quick clustering question- In the docs, I found this statement: [quote] If the MASTER node of a cluster crashes, another node will become the new MASTER node. When the crashed node comes back online, it will become a MEMBER node. It isn’t important which node is the MASTER node. All Designers communicate only with the MASTER node. Clients communicate with either the MASTER or MEMBER nodes.[/quote] If I am working on a project that is still in continuing development, do I still use the designer on the original master PC even when the other PC becomes the master for some reason? Or if I am still in development, would I have to make sure the orginal master is still the master if I needed to run the designer? ---
**Author:** ![Carl.Gould](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/carl.gould/32/7262_2.png) [@Carl.Gould](https://forum.inductiveautomation.com/u/Carl.Gould)\ **Post date:** [October 20, 2009, 5:05pm UTC](https://forum.inductiveautomation.com/t/clustering/1343/2 "2009-10-20T17:05:04Z")
No, you can launch the designer from wherever you’d like. It will automatically talk to the Master, whichever machine that happens to be at the time. ---
**Author:** ![Step7](https://avatars.discourse-cdn.com/v4/letter/s/45deac/32.png) [@Step7](https://forum.inductiveautomation.com/u/Step7)\ **Post date:** [October 20, 2009, 5:18pm UTC](https://forum.inductiveautomation.com/t/clustering/1343/3 "2009-10-20T17:18:56Z")
Ok, I’m still confused. If I copy a project on Cluster Node 1 through Configuration\>Manage Projects, and then add Node 2 three days later, I assume that Node 2 will then grab the project from Node 1, right? So, if Node 1 then goes down, Node 2 becomes the Master, and then Node 1 comes up again as the passive node, do I have to copy the project to Node 2 to make changes from the designer? ---
**Author:** ![Carl.Gould](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/carl.gould/32/7262_2.png) [@Carl.Gould](https://forum.inductiveautomation.com/u/Carl.Gould)\ **Post date:** [October 20, 2009, 5:27pm UTC](https://forum.inductiveautomation.com/t/clustering/1343/4 "2009-10-20T17:27:09Z")
No, you’re thinking too hard. Clustering means the configuration is shared automatically and synched. There is no manual copying of projects involved. When you have the designer open, it loads the projects from the master, saves to the master, and the master replicates that save to all members. ---
**Author:** ![Step7](https://avatars.discourse-cdn.com/v4/letter/s/45deac/32.png) [@Step7](https://forum.inductiveautomation.com/u/Step7)\ **Post date:** [October 20, 2009, 9:50pm UTC](https://forum.inductiveautomation.com/t/clustering/1343/5 "2009-10-20T21:50:07Z")
Ok, gotcha. Ultimately, there is always only one version of a project, and the clustering takes care of everything. Btw, I just did a clustered config at a customer today, and it works great. The IT dept was impressed. Another quick question- is there a way to get to the FactorySQL Front End from a client? It’s easy to get to the FPMI designer, but I happened to be out on the floor at a client and wanted to make a change to a group, but I couldn’t see how to get to it quickly (not that I tried real hard; I was under a certain amount of duress and ended up just going to the server room in person). ---
**Author:** ![Carl.Gould](https://sea1.discourse-cdn.com/inductiveautomation/user_avatar/forum.inductiveautomation.com/carl.gould/32/7262_2.png) [@Carl.Gould](https://forum.inductiveautomation.com/u/Carl.Gould)\ **Post date:** [October 20, 2009, 9:52pm UTC](https://forum.inductiveautomation.com/t/clustering/1343/6 "2009-10-20T21:52:39Z")
You have 2 options: 1. Install FactorySQL and open the frontend, but point it to the remote service 2. Remote desktop