Hi All, I need to prepare a SCOM scope and LLD. As per the suggested architecture we have two sites Prod site and DR site. We plan to keep the SCOM RMS and SCOM MS server at Production site and one SCOM server at the DR site which will act as a Backup for the same. Kindly confirm if my understanding is correct and suggest how should I distribute the roles like RP, ACS etc... Regards, Gaurav

Hi Jain, Check first what kind of connection you have with the DR site where you place a managenment server. The connection between ANY management server and the database should be very good and quick and with little latency. Reporting and ACS etc etc, place all in production site. If the line is good enough (at some companies the line between datacenters seems to be better than the one between two racks of servers inside the same one...) you can place a management server in DR. More interesting first is the DR of the SQL databases (opsdb and opdbDW have most prio I imagine, possibly if you have very strong auditing rules to adhere to also the ACS database). Without those all things stop. You can promote the management server in DR to RMS, but it still needs a database to connect to. ACS you can use a passive secondary collector in the other site (as long as you have a plan to bring the database to that site as well in case of disaster). Report server and web console are as I see them a little less interesting. Quickly enough to re-install later. As long as you have access to the scom console you can monitor. Creating reports is less important during disaster times. Windows scom Agents will automatically fail over to another management server when their primary is down, so if your prod-MS boxes are down they will eventually find the one in DR. Of course the MS and agents both have queues and can stack data for a number of hours at least in their queues. So creating an LLD or giving tips really involves much more information to be honest. also needs info on other DR methods you ahve already. Perhaps your RS/Web server is a virtual machine and you have a full snapshot backup of that thing which you can place back running and all on the other site. Perhaps you have QSL log shipping or a streched cluster in place for the SQL boxes between the two sites. Perhaps you have a need to get monitoring up again within 1 hour. Perhaps its 2 days. Perhaps you must retain data, perhaps you can just start fresh with an empty database. It all depends on the high level design and the company requirements on how you will incorporate all the components into the LLD and into the building stages of the solution (and a test run *wink*). If I only go for your first question.. yes the MS in the other site will receive windows agent traffic when the other one goes down (although also the RMS will get a number of agents as it is an MS as well). If both go down in the primary site you can still promote the MS in DR, but your main dependency is the database! So it covers a crash of the RMS or MS, but it doesnt as easy cover a datacenter failure where you lost your database. Important is to make sure that during normal operation that MS in DR site does have low latency with the database in production site. Otherwise do not place an MS there and find another backup/restore method. Let me know if that helps you.Bob Cornelissen - BICTT (My Blog about SCOM) - MVP 2012 and Microsoft Community Contributor 2011 Recipient