DAG Design

I've got 3 database servers in a DAG, 2 of them at my main site, and 1 of them at my DR site. We also have 2 file share witnesses configured, the primary being at the main site, and the alternate at the DR site. We recently had a scenario where
the main site went down completely and that caused the cluster to go down, leaving only the db server and FSW at the DR site up. Since the cluster was down, the databases dismounted on the DR site server and would not mount. It appears that the
servers at the main site going down caused the DR site to think that it was the site that was offline and therefore things didn't fail over properly.
My question is - What can we change to make it so that the DR site will become the active site in the event of an unexpected outage at the main site? I understand that since we have 3 database servers, we're not actually using the FSW at either location
unless we happen to lose one of those 3 servers. Would putting the primary FSW at the DR site remedy this problem?

Failovers are not automatic in your scenario.
You need to implement DAC Mode and do a datacenter switchover:
http://technet.microsoft.com/en-us/library/dd351049(v=exchg.150).aspx
If you want automatic failover, you would need a 3rd datacenter for the File Share Witness and an even amount of DAG members in each data center.
http://blogs.technet.com/b/scottschnoll/archive/2012/11/01/storage-high-availability-and-site-resilience-in-exchange-server-2013-part-3.aspx
Twitter!: Please Note: My Posts are provided “AS IS” without warranty of any kind, either expressed or implied.

Similar Messages

Exchange 2010 /2013 DAG Design

Hi All,
I Have Gone through some Articles and Blogs where i read about the DAG Design and Member server placement in Site Resiliency Scenarios..
I just want to know what and how many DAG Designs we can propose to Customers.
As per my knowledge and study i came across -
(A) Active/Passive Model with Shared Namespace (Single DAG)
(B) Active/Passive model with different namespaces (Single DAG)
(C) Active/Active Model with Different Namespaces (2 DAGs).
I am Just Curious to know is there any other model DAG that can be proposed apart from above mentioned Designs in terms of both Namespace and Database Active copy Location.
Also can you also put some light and clarify whether we can have
Active/Active Model with Shared Namespace (Single DAG). I have read it is supported in exchange 2013, if so can we do the same in Exchange 2010 as well. If not what makes this design different in exchange 2010 v/s exchange 2013. How can we
achieve this design in exchange 2013?
I got a good article on above scenarios
here but I have some terms or paragraphs bit
confusing. Please help me to understand the bold lines in below paragraph.
Since we typically only have active users connecting to the primary datacenter (at least the majority of the users connect to this datacenter unless a site failover occurs),
the autodiscover record (autodiscover.exchangelabs.dk) in external DNS points to the load balancer in the primary datacenter. The internal “AutoDiscoverServiceInternalUri” on the CAS servers in the primary datacenter has been configured with a value of https://mail.exchangelabs.dk/autodiscover/autodiscover.xml,
and the same goes for the CAS servers in the failover datacenter. Now you could choose to point the AutoDiscoverInternalUri on CAS servers in the failover datacenter at “https://failover.exchangelabs.dk/autodiscover/autodiscover.xml” but
you can easily up on in a situation where SCP’s aren’t reachable during a site failover. Also, cross-site traffic caused by Autodiscover have a minor impact on the WAN link since autodiscover requests consists of small XML based text files.
Also in
Active/Passive model with different namespaces (Single DAG) or Active/Active Model with Different Namespaces (2 DAGs) model do we still repoint EWS,OA,EAS
etc. web service records in public DNS. If yes what should be their value (please refer
Complete Site failover Segment in Above URL)
it would be very helpful if you could provide any detailed Design diagram as mentioned in URL to understand the things better. or any Technet Articled with well
description
Thanks in advance
Regards
Aanand Singh
Regards, Aanand Singh

Hello,
The sentence " you
can easily up on in a situation where SCP’s aren’t reachable during a site failover. Also, cross-site traffic caused by Autodiscover have a minor impact on the WAN link since autodiscover requests consists of small XML based text files." I
understand that if site failover occurs,as long as the CAS servers are available in the primary datacenter, clients will continue to connect to the CAS servers in this datacenter. The CAS servers will not be used in the failover datacenter.
Consider WAN outage, we recommend you deploy Active/Active Model with Different Namespaces (2 DAGs).
Besides, I recommend you configure DNS to round-robin between the VIPs of the load balancers in each datacenter to realize single namespace.
Here are some articles for your reference.
http://technet.microsoft.com/en-us/library/dd979781(v=exchg.141).aspx
http://blogs.technet.com/b/scottschnoll/archive/2012/11/01/storage-high-availability-and-site-resilience-in-exchange-server-2013-part-3.aspx
http://blogs.technet.com/b/exchange/archive/2013/01/25/exchange-2013-client-access-server-role.aspx
Cara Chen
TechNet Community Support

Exchange 2013 - Unbound Stretched Active/Active DAG

I'm designing an Active/Active DAG utilizing a 3-site scenario, with the third site for FSW.
I intend to have 3 CAS+MBX servers in each location.
Site A & B are separately co-located and Site C is geographically distant site for the FSW
Should the WAN connections of Site A go down (for whatever reason), do the passive copies of Site A's DB's on Site B automatically mount?
What happens with the copies in Site A when this is going on, and what occurs when the links are restored? Is there any manual intervention required here and is DAC/Split-Brain syndrome a concern at all?
What if the client doesn't want Auto-failover, can blockautoactivation be set for the passive DB's to -SiteOnly?

Hi Hayley,
Q:- Should the WAN connections of Site A go down (for whatever reason), do the passive
copies of Site A's DB's on Site B automatically mount?
Yes it could be automatically mount depends on DAC mode, you should enable it by default it is disable.
Q:- What happens with the copies in Site A when this is going on, and what occurs when the links
are restored? Is there any manual intervention required here and is DAC/Split-Brain syndrome a concern at all?
Site A copy would be like Passive Copy which will get the log shipping from Site B once the WAN link will be up and running, as the DAC mode is enable will not allow
the Site A database to mount its database once it will up and running,
You can later change your active copy to Site A.
Q: - What
if the client doesn't want Auto-failover, can blockautoactivation be set for the passive DB's to -SiteOnly?
Please check the below link for better understanding.
http://technet.microsoft.com/en-us/library/dd979790(v=exchg.150).aspx
Thanks & Regards
Ashish Gaur
Ashish Gaur

Exchange 2013 MBX in DAG along with Hyper-V and Failover Cluster

Hi Guys! I've tried to find out an answer of my question or some kind of solution, but with no luck that's why I am writing here. The case is as follows. I have two powerful server boxes and iSCSI storage and I have to design high availability
solution, which includes SCOM 2012, SC DPM 2012 and exchange 2013 (two CAS+HUB servers and two MBX servers).
Let me tell you how I plan to do that and you will correct me if proposed solution is wrong.
1. On both hosts - add Hyper-V role.
2. On both hosts - add failover clustering role.
3. Create 2 VMs through failover cluster manager, VMs will be stored on a iSCSI LUN, the first one VM for SCOM 2012 and the second one for SCDPM 2012. Both VMs will be added as failover resource.
4. Create 4 VMs - 2 for CAS+HUB role and 2 for MBX role, VMs will be stored on a iSCSI LUN as well.
5. Create a DAG within the two MBX servers.
In general, that's all. What I wonder is whether I can use failover clustering to acheive High Availability for 2 VMs and at the same time to create DAG between MBX-servers and NLB between CAS-servers?
Excuse me for this question, but I am not proficient in this matter.

Hi,
As far as I know, it’s supported to create DAG for mailbox server installed in hyper-v server.
And since load balance has been changed for CAS 2013, it is more worth with DNS round robin instead of NLB. However, you can use NLB in Exchange 2013.
For more information, you can refer to the following article:
http://technet.microsoft.com/en-us/library/jj898588(v=exchg.150).aspx
If you have any question, please feel free to let me know.
Thanks,
Angela Shi
TechNet Community Support

Need to know how to calculate and allocate the server side bandwidth between two exchange sites for DAG replication

Hi All,
Please have a look in to the below mentioned environment.
primary site :SITE 1
server 1
server 2
secondary site :SITE 2
server 3
Note : All the above three servers are in single DAG .All the database are mounted on the server (server 1) which is located in the primary site (I.E SITE 1).
On that 5 databases we are having 3500 users .Based on the user designation we have allocated the mailbox size.
Query : Each and every day we are having the large no copy queue length for all the five databases to the server located in SITE 2.
Please help me out on this and also tell me is there any tool available to exactly get the required bandwidth for the exchange servers located between the active directory sites.
Thanks & Regards S.Nithyanandham

I don't know of any tool that can calculate things for you but latency is the biggest thing you have to worry about. The other issues is looking at the network gear between the 2 sites. I have seen many times were switches get maxed out if they are set to
1 GB or even ports getting maxed out and need to bonded for more throughput. Test your latency from the edge of each site then test it within to see if there is a difference.
Regardless of their geographic location relative to other DAG members, each member of the DAG must have round trip network latency no greater than 500 milliseconds between each other member.
As the round trip latency between two Mailbox servers hosting copies of a database increases, the potential for replication not being up to date also increases. Regardless of the latency of the solution, customers should validate that the networks between
all DAG members is capable of satisfying the data protection and availability goals of the deployment. Configurations with higher latency values may require special tuning of DAG, replication, and network parameters, such as increasing the number of databases
or decreasing the number of mailboxes per database, to achieve the desired goals.
Round trip latency requirements may not be the most stringent network bandwidth and latency requirement for a multi-datacenter configuration. You must evaluate the total network load, which
includes client access, Active Directory, transport, continuous replication, and other application traffic, to determine the necessary network requirements for your environment.
http://technet.microsoft.com/en-us/library/dd638104(v=exchg.150).aspx
DJ Grijalva | MCITP: EMA 2007/2010 SPA 2010 | www.persistentcerebro.com

Urgent Help / suggestion Needed - DAG in Differnt Site - Implementaion

Hi All
Current setup of my lab is below -
MAIN SITE - Named Head Office
Root Domain Controller : IP 172.29.154.1 / 255.255.0.0
Two node DAG : - Mailbox servers - DAG NAME - DAG1
Casarray (casaarray1 , and casarray2) servers
One hub transport server
Total 6 VM servers are running in the head office site
Created a new AD site for name Branch Office and installed a addition domain controller with IP of 10.10.10.1/255.0.0.0
Now i would like to implement the DAG between the sites .
how can i implement, i mean what are the best ways to implement the DAG between two sites
1. shall i go for typical installation of exchange (CAS, HUB , MAILBOX ) 2010 in main office and Branch office . will it support DAG between two sites for typical installation?
2. shall i install one more Mailbox server alone in MAIN SITE and install CAS and hub in one server and mailbox alone in the branch office site? i mean - two Mailbox servers in each site
any other best ways other than above two points , the main intention is to implementing, testing and learning the site resilience between the sites and want to know the stuffs in and around DAC
Thanks in advance

so, i was assuming that i will add one mbx server in main site and another mbx server in branch site and make it HA via DAG --- is it possible
Next option
i can go for typical installation of ( CAS , HUB MBX server (all three in in one server) in main site and another three server roles (aal in one server) of Branch office and make the mailbox for HA -- will it possible
how can i achieve the HA with my current setup between the sites
If your bandwidth and latency between sites  you
can have  HA for mailbox databases by adding a mailbox server as a member in the current DAG. DAG will
give you only HA for mailbox not HUB/CAS.
After you installed the Mailbox role on the branch server, You have to set autodiscover site affinity.
http://technet.microsoft.com/en-us/library/aa998575(v=exchg.141).aspx
It is little bit complicated if you want HA over 2 sites, It depends on the bandwidth and latency between sites. If
you install all the 3 roles and configure the branch site as part of the main branch ADsite you can have highly availability for CAS by a load balancer and this is not a small change. i.e. add the branch server to the Main office CAS
Array and load balance.
http://channel9.msdn.com/Events/TechEd/NorthAmerica/2011/EXL401
If the branch site is not Internet facing, Please check this.
http://technet.microsoft.com/en-us/library/bb310763(v=exchg.141).aspx
adding an MBX/HUB/CAS in your branch office and buy a hardware load balancer of your budget. List
of MS supported HLBs here.  Configure the load balancer to according to your bandwidth and latency
between sites.
Thanks, MAS
Please mark as helpful if you find my comment helpful or as an answer if it does answer your question. That will encourage me - and others - to take time out to help you.
Couple of comments please!
Autodiscover sitescope is a function of CAS not mailbox. This is written into AD when the CAS role is installed.
Why are you splitting all the roles out like this? The recommendation in 2010 and 2013 is to multirole whenever possible. There are valid reasons for splitting, I just want to know what's driving you to this.
The preferred DC switch is just for the current PowerShell session IIRC. I'll check up on that though.
I'd ask you to look at the design examples on TechNet
http://technet.microsoft.com/en-us/library/dd979781(v=exchg.141).aspx
In all of this, you have to consider what the impact of a failed WAN link(s) will do. If you have a single DAG spanning both locations, and the WAN goes down then in the case of a single DAG spanned between both of the sites some users will not have
access to a local mailbox server. The cluster service terminates and so does the Exchange store. That is by design.
Cheers,
Rhoderick
Microsoft Senior Exchange PFE
Blog:
http://blogs.technet.com/rmilne
Twitter:   LinkedIn:
  Facebook:
  XING:
Note: Posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

2013 HA Design Questions

Hi everyone,
We're preparing for a GroupWise to Exchange 2013 migration. I'm currently coming up with the Exchange environment design and have a few questions about how best to set up the DAG(s) and CAS HA for our environment. I've been using the sizing calculator as
a guide but have been a bit thrown off by some of the suggestions.
We have two primary data centers, each with three ESXi hosts in HA clusters. I would like the overall Exchange design to be active/active. We have about 2,000 email users, so the split can be about 50/50 for each DC. In the calculator, it states that there
are two approaches to the active/active design - one is to have one DAG stretch across the DC's, the other is to have two separate DAG's.
Which approach will allow us to remain available if one DC goes down? I can't think of a situation where two DAG's will be needed. We do have a third DC where a witness server can be placed. When I went with the dual DAG approach in the calculator, the mailbox
server recommendation jumped to 32 which seems like I've done something incorrectly.
Any input would be greatly appreciated!
Thanks!

Hi everyone,
We're preparing for a GroupWise to Exchange 2013 migration. I'm currently coming up with the Exchange environment design and have a few questions about how best to set up the DAG(s) and CAS HA for our environment. I've been using the sizing calculator as
a guide but have been a bit thrown off by some of the suggestions.
We have two primary data centers, each with three ESXi hosts in HA clusters. I would like the overall Exchange design to be active/active. We have about 2,000 email users, so the split can be about 50/50 for each DC. In the calculator, it states that there
are two approaches to the active/active design - one is to have one DAG stretch across the DC's, the other is to have two separate DAG's.
Which approach will allow us to remain available if one DC goes down? I can't think of a situation where two DAG's will be needed. We do have a third DC where a witness server can be placed. When I went with the dual DAG approach in the calculator, the mailbox
server recommendation jumped to 32 which seems like I've done something incorrectly.
Any input would be greatly appreciated!
Thanks!
If you have a 3rd datacenter, then place the File Share Witness there and have one DAG stretched across the 2 peer datacenters with an equal amount of DAG members in each DC. That will give you automatic failover in the event one DC does down.
http://blogs.technet.com/b/exchange/archive/2014/04/21/the-preferred-architecture.aspx
Twitter!:
Please Note: My Posts are provided “AS IS” without warranty of any kind, either expressed or implied.

Exchange logs DAG best practice

In real world scenarios do may of you enable circular logging? I have never had to replay log files and usually enable circular logging.
Also in a DAG do the transaction logs get created on each member if circular logging is disabled?
finally what is best practice for backing up databases in a dag if i have 2 active with split mounted copies of each database so db 1,2,3 are on server one with passive 4,5 and 123, are passive on server 2 with 4,5 active then I have server 3 with 1,2,3,4,5
all passive where do I back up from?
***Don't forget to mark helpful or answer***

I wouldn't enable the circular logging and let my Daily Backup take care of log truncation. However you can enable the circular logging, it all depends on how you want to design your infra, backup and recovery strategy...
Yes in DAG, transaction logs are copied over to other server where database copy is located, inspected and checked that it is not needed by other database before it got truncated due to circular logging.
I would take backup from the server 3 where all the DBs are not mounted...
Some references:
Circular Logging and Mailbox Database Copies - http://blogs.technet.com/b/scottschnoll/archive/2011/06/27/circular-logging-and-mailbox-database-copies.aspx
Managing mailbox database copies - http://technet.microsoft.com/en-us/library/dd335158(v=exchg.150).aspx
Amit Tank | Exchange - MVP | Blog:
exchangeshare.wordpress.com

DAG Database Storage Location in addition to Configuring a Proper Backup/Restore Policy

Hello everyone! I have two questions that have LOTS of detail so I'll jump right into describing our Exchange 2010 layout and then the questions:
EXCHANGE SERVERS:
- cvnit-ex01.***.***.*** - Hub,CAS,MB
Database: MBDB1
- cvnit-ex02.***.***.*** - Hub,CAS,MB
Database: MBDB2
- Both exchange servers have one DB and they are both part of the same DAG.
- Another server is the DAG witness server.
- Only two servers in forest.
- Members of CASArray, using 3rd part DNS for round-robin
- DAG failover has been actively tested and works - users with mailboxes on MBDB2 experience less than 5 seconds of an outage if I shut off cvnit-ex02
OKAY, so here is my question/statement. I spent a lot of time on the storage side creating LUNs for each exchange server to hold its edb and logs. For example, on cvnit-ex01:
MBDB1.edb = G:\
MBDB1 logs = H:\
I then also created a separate LUN, assuming I could tell DAG where to store the passive copies of a database on the target server.
"passivedb.edb" = I:\
"passive logs" = J:\
The problem is, I see DAG copied the DB from cvnit-ex01 over to cvnit-ex02's G:\ drive. Is there any way I can configure it to store the copied databse on that separate, designated LUN (I:\ for edb and J:\ for logs)?
The end goal would be for ex01 and ex02 to have four drives EACH: 2 for its primary db and logs, and two for the passive db and logs. Both edb files are relatively the same size and the capacity of the LUNS holding the passive copies is not an issue.
On to the other one...
Is there any step-by-step instruction on how to properly enable DAG backups of databases so that I can use lagged copies and then in turn enable circular logging? Keep in mind that this is a fresh exchange 2010 installation and implementation, and I still
have time before it goes into production. I don't quite understand the whole "need three copies of databases and a lagged copy" ordeal.
Thank you in advance!

for the 1st question: No, that letters must match. If the Database is on the G: drive and you replicate to another server, G: must exist.
http://technet.microsoft.com/en-us/library/dd979802(v=exchg.141).aspx
All copies of a database use the same path on each server containing a copy. The database and log file paths for a database copy on each Mailbox server must not conflict with any other database paths.
I was not a big fan of lagged copies in Exchange 2010. Some instructions:
http://www.msexchange.org/articles-tutorials/exchange-server-2010/high-availability-recovery/eliminating-traditional-backups-using-native-exchange-2010-functionality-part4.html
Twitter!: Please Note: My Posts are provided “AS IS” without warranty of any kind, either expressed or implied.

Number of DB's in a DAG

I'm using the requirements calculator and having trouble understanding the recommendations it's coming up with in a certain scenario. At first I have put in that there will be a total of 4 Mailbox servers (2 in each DC). For "Total number of HA DB Copy
instances within DAG" I have put 4 (thinking there would be 1 active DB on each server). I then put 2 for the "Number of HA DB Copy Instances in SDC" so there would be 2 active in PDC and 2 active in SDC. We have 2,000 users, so there should
be 500 in each active DB.
This results in the Role Requirements showing what I expect - 2 active DB's in each DC, a total of 4 DB's with 500 users each.
What I don't understand is what happens when I switch the number of mailbox servers to 8 total (4 in each DC). I change the "Total number of HA DB Copy Instances" to 8 (one active DB per server), with the number in the SDC to 4. I would expect this
to result in a recommendation of 8 active DB's with 250 users each. Instead the calculator is suggesting there be 24 databases with only 83 users each. Each server is stated to have 3 active DB's instead of the 1 I was expecting.
Can anyone shed some light onto why this might be happening?

Hi MRCUR,
According to your description, I understand that you want to add more DAG members and active DB to reduce loss when mailbox server failed.
I agree with Ed, we have to balance cost, availability, and performance when we design a DAG. We can enable more active DB under each member to balance the amount of member and DB.
More details about Exchange server performance, please refer to Example of Sizing for an Exchange 2010 Multiple-Role Scenario section under
Understanding Multiple Server Role Configurations in Capacity Planning:
http://technet.microsoft.com/en-us/library/dd298121(v=exchg.141).aspx
more details about Database availability groups, for your reference:
http://technet.microsoft.com/en-us/library/dd979799(v=exchg.150).aspx#DAGforSiteResilience
Best Regards,
Allen Wang

Exchange server 2013 - Setting up DAG

Hi Guys,
I have an Exchange 2013 server(a single dedicated box with all the roles(MailBox and Client Access) configured in it. Now I want to enable data redundancy and Failover. So, I want to setup a DAG and my idea is given below.
I Want to setup a 2 node DAG. The first node is my current working box and the second node is another box at a different location. I would then Install Exchange on the 2nd node with mailbox role on it. As it is a 2 node DAG, I would require another box as
a witness server.
Please let me know if there is any flaws with the above setup. Also, let me know if I require a different Exchange server license for the second installation.
Thanks and Regards,
Nash Burns

Hi,
From the article blow, we know a license must be assigned for each instance of the server software that is being run. The second Exchange server need a license.
Exchange Server 2013 licensing
http://office.microsoft.com/en-us/exchange/microsoft-exchange-server-licensing-licensing-overview-FX103746915.aspx
And the type of license is based on the number of mailbox databases.
Standard: designed for the mailbox needs of small to midsize organizations. Also appropriate for non-mailbox roles in a larger Exchange deployment. This edition supports 1 to 5 mailbox databases.
Enterprise: designed for larger organizations that may require a greater number of mailbox databases. This edition supports 1 to 100 mailbox databases.
Best regards,
Belinda
Belinda Ma
TechNet Community Support

DAG in exchange server 2010 across different sites

hello ;
I want to know the best way to implement DAG for exchange server 2010 across 2 sites in the same domain ,
I have 2 exchanger server in one site in DAG ,with 300 gb database size
but I want to implement exchange server 2010 in a new site within the same domain and this will be my 3rd exchange server and will be a member on my existing DAG in the head office
the new site is linked to my head office using ipvpn connection which is 6 mb , and ADC is there in the new site,
what is the best way to implement the new exchange server in the new site to join the existing DAG .
is there is any problem related to DAG in this scenario ?
what is the best way to implement this scenario ?
Thanks

Hi,
For each AD site, it is recommended to have at least there Exchange roles: CAS, HUB, and MBX roles.
Do you plan to have active users in this new site?
If there are active users in this new site, you can deploy a Three-Member DAG in two AD sites. However, it is recommended to deploy two DAGs in two AD sites and you should have at least two DAG members each AD site. This scenario, two witness servers in
separate site is necessary.
If the new site is only for DR purpose, you can install this new Exchange server and add it to DAG. After that, add database copy to this new server.
Besides that, here are two articles which may help you for your reference.
Database Availability Group Design Examples
http://technet.microsoft.com/en-us/library/dd979781(v=exchg.141).aspx
Understanding Database Availability Groups
http://technet.microsoft.com/en-us/library/dd979799(v=exchg.141).aspx
Hope this helps.
Best regards,
Belinda
Belinda Ma
TechNet Community Support

DAG/cluster network troubleshooting

We have a 2 DAG cluster EX01 and EX02. EX02 holds all active copies and EX01 holds all passive copies.
When trying to move active copies from 02 to 01 it fails everytime. Also trying to move Primary active manager from 02 to 01 fails. There is an error in Cluster manager: Cluster resource 'IPv4 Static Address 1 (Cluster Group)' in clustered service
or application 'Cluster Group' failed. Event ID 1069. All Replication test and health test come back good. I think the error is related to network design and need some advice.
EX01 (HUB,CAS, MB)                 EX02 (MB)
Mapi 10.11.1.4/24                   Mapi 10.11.1.20/24
Replication 10.0.5.2/30           Replication 10.0.5.1/30
We have a single DAG IP 10.11.1.21 does it need to be on the same network as replication ?? Do we need to have 2 IPs for DAG??
We want EX01 be the primary and hold active copies and EX02 be secondary
Please advice

Yes that have never been an issue - the DAG is online all passive copies are healthy. The issue is that EX01 can hold only passive copies - when I run move-activedatabasecopy cmdlet it fails to mount it every time. Currently EX01 holds all
passive copies and EX02 holds all active copies and it should be the other way around. It looks like the versions of updates/rollups are slightly different between 2 servers so I'm exploring options how to sync updates
alex serdyukov
Hi,
Can you create a test database and then try to move it ? Paste the error if you get any.
Amit Rawat | MCITP - Exchange 2007/2010| CCNA |MCSE- 2003| Lync 2010|http://blog.amitrawat.net

Exchange 2013 HA Design

All,
We are going to design a new Exchange 2013 enviroment and have 800 mailboxes totally. 600Mailboxes will have 10GB mail quota, 200 Mailboxes will have 25GB quota.
I want to know that if normally design of this case, how many Mailbox server is required ? And we will have 10% mailboxes enabled archiving function. DR Site also required.
Can anyone able to provide some suggestion for consider ?
thanks

Thanks all reply.
I tried to use Exchange sizing calculator but cannot get the result as I want. If I want to use 2 mailbox
servers for DAG and want to have 4 DAGs. How can I fill the value ?
I filled below value for calculate
"Number of mailbox servers Hosting active mailboxes / DAG" is 2
"Number of Database availability Groups" is 4
"Total Number of HA Database Copy Instances (includes Active Copy) within DAG" is 4 (yellow shown)
Is it I cannot create 4 DAGs on 2 node cluster ?
thanks

Synchronization between multiple DAGs

Hi,
Is there any possibility that we can have synchronization between 2 DAGs. We want to have 2 DAGs in 2 sites. Replication between 2 DAGs should be realtime.
Is there any such possibility?
Thanks
Anees

Right and this is by design. Exchange will not mount the databases in Site B, because Site B doesn't had quorum. Site A still has node majority.
In that case I would look into Datacenter Activation Coordination (DAC).
Section towards the bottom titled "Enabling Database Activation Coordination Mode (DAC)"
http://www.msexchange.org/articles-tutorials/exchange-server-2010/management-administration/planning-deploying-testing-exchange-2010-site-resilient-solution-sized-medium-organization-part7.html
A later part in that series walks you through the Site Failover experience and all the pitfalls to watch out for.
http://www.msexchange.org/articles-tutorials/exchange-server-2010/management-administration/planning-deploying-testing-exchange-2010-site-resilient-solution-sized-medium-organization-part12.html
Check out my latest blog posts on www.supertekboy.com A tech blog for I.T. Superheroes.

DAG Design

Similar Messages

Maybe you are looking for