Hi All,

I have two sites, one of the site will be offline for few days and I need to make sure that the mailboxes on that site are live. Here is the deployment at that site:

DAG1 -Mailbox Server 1
DAG1 - Mailbox Server 2
DAG1 - Mailbox Server 3 (at second site)

CASArray of two cas servers
Edge Servers
Exchange Online Protection

My questions are:
1) How to make DAG1 live only by having Mailbox Server 3 online during the time rest two are offline? Is quarom going to five issues?
2) If somehow DAG1 is live with Mailbox3, I will make databases RPCClientProxy to CASArray on Second site and remove EdgeServer IPs from EOP for the site going offline. Is this plan good enough?

I appreciate the community's help!

Hi Abdullah,

Q1)How to make DAG1 live only by having Mailbox Server 3 online during the time rest two are offline? Is quarom going to five issues?

A. There are two ways, either way you need a Witness Server available online and accessible by the MBX3. First i.Dynamic quorum can survive failures if only one server at a time fails till last MBX+Witness(CAS can be if not multirole)

ii.When the quorum is down, you can force quorum using a Witness and MBX3 , by excluding\eviting the MBX1,2 from DAG.

Read this to know more on Dynamic Quorum on Windows Server 2012

Q2) If somehow DAG1 is live with Mailbox3, I will make databases RPCClientProxy to CASArray on Second site and remove EdgeServer IPs from EOP for the site going offline. Is this plan good enough?

Please note Ex2013 doesn't use or have CAS Array, its just a DNS names. RPCClientProxy field is used to generate the XML that is provided by Autodiscover. So if you have a single namespace for CAS using DNS roundrobin at Public site its enough for CAS failover or work without any config changes for clients. And yes if you want you can remove the Offline Site's Egde's IP from EOP.

Just realized your question was for Ex20

Hi,

1) You would need a Witnesses in both sites and manual intervention to activate it for Secondary.

A four-member DAG

https://technet.microsoft.com/en-us/library/dd979781(v=exchg.141).aspx

A four-member DAG in a single datacenter deployment provides greater resilience to failures than a two-member or three-member DAG. Larger DAGs inherently provide greater resilience because they can sustain more failures without an interruption in service. Whereas a two-member or three-member DAG can sustain the loss of only a single voter without losing quorum and compromising service, a four-member DAG, which by definition has five quorum voters, can sustain the loss of two voters without losing quorum and compromising service.