Exchange 2010 DAG backup & Transaction logs

Hi, 
What is Microsoft recommended best practise for Exchange DAG group backup in an environment where there are Active & multiple (2-3) passive copies of the databases?
Is it a good practice to backup transaction logs as frequently as possible in addition to daily full backup ? This I belive will allow to restore the DB to the latest possible state
using last good full backup & transaction logs (Like restoring SQL databases)
Thanks

Hi,
Windows Server Backup can't backup passive copy. If you want to backup both active and passive copies, you need to use DPM or other third party.
Here is a similar thread for your reference.
Exchange 2010 DAG Backup Best Practices
http://social.technet.microsoft.com/Forums/exchange/en-US/269c195f-f7d7-488c-bb2e-98b98c7e8325/exchange-2010-dag-backup-best-practices
Besides, here is a related blog below which may help you.
Backup issues and limitations with Exchange 2010 and DAG
http://blogs.technet.com/b/ehlro/archive/2010/02/13/backup-issues-and-limitations-with-exchange-2010-and-dag.aspx
Hope this helps.
Best regards,
Belinda
Belinda Ma
TechNet Community Support

Similar Messages

  • Exchange 2010 - Load balancing transaction logs in a DAG

    I have a single Exchange 2010 DAG.  Within the DAG, I have two mailbox servers.  I have 2 DB's mounted on server A, and 3 DB's mounted on server B.
    The size and quantity of transaction logs on server A always seem to be larger.  It might be the nature of the mailboxes on server A, but I was wondering if there is a tool or script to verify best practices for how the transaction logs should be configured.
    Thanks
    Ron

    I have a single Exchange 2010 DAG.  Within the DAG, I have two mailbox servers.  I have 2 DB's mounted on server A, and 3 DB's mounted on server B.
    The size and quantity of transaction logs on server A always seem to be larger.  It might be the nature of the mailboxes on server A, but I was wondering if there is a tool or script to verify best practices for how the transaction logs should be configured.
    Thanks
    Ron
    So one of the DBs is not replicated? Im not sure what you mean by "how the transaction logs should be configured".
    You only real choices are disk location, whether you want to lag a copy or if you want to enable circular logging. Things that with 2 mailbox servers you wouldn't want do other than define the disk location. Otherwise, there isn't much to configure as far
    as the logging.
    Twitter!: Please Note: My Posts are provided “AS IS” without warranty of any kind, either expressed or implied.

  • Exchange 2010 Excessive Archive Transaction Logs

    I've got an Exchange 2010 environment with about 225 users over 6 mailbox DB's with corresponding Archive DB's.
    They're seeing log files for the archive DB's growing by about 25-30% of the total DB size daily. Those logs are generally about 8 x the logs generated by the corresponding primary mailbox database.
    Something like
    DB1 = 20GB  daily average logs 2GB
    DB1Archive = 90GB daily average logs ~28GB
    I don't see any Retention policies that have duplicated Tags and the Default policy has no tags assigned.
    The customer wants to continue with their existing backup process of weekly fulls and daily incremental/differentials, and missing two days is enough to max out their transaction log volumes. (which are as large as their combined database size).
    The archive db's have been in place for at least 12 weeks now so it's not just initial population, which is obvious by the fact that the archive db's are much larger than the original db's and the total logs in a week exceed the DB volumes by 2-3 times.

    Hi,
    First, please make sure there is no corruption in the retention policies.
    If there is no corruption, I recommend you create a new archive mailbox database and move archive mailbox to this new database to check result.
    Besides, here is a similar thread for your reference.
    http://social.technet.microsoft.com/Forums/en-US/77845ba9-96c4-4b94-8d55-b47c51ef8974/exchange-2010-archive-transaction-logs?forum=exchange2010
    Hope this is helpful to you.
    Best regards,
    Belinda Ma
    TechNet Community Support

  • Exchange 2010 DAG netowrk replications issues - Causing database copies to be disconnected and resynchronizing.

    My
    environment is as follows.<o:p></o:p>
    We have 3 exchange 2010 servers.<o:p></o:p>
    EX1, EX2, DREX1. All servers are exchange 2010 14.2 build 247.5<o:p></o:p>
    EX1 and EX2 have two nics, one LAN and one for BACKUP <o:p></o:p>
    The dag is configured to to communicate to all three.<o:p></o:p>
    The IP for EX1 is 192.168.100.51 - LAN, switch A; 160.100.3.51 - BACKUP, switch B<o:p></o:p>
    The IP for EX2 is 192.168.100.52 - LAN, switch A; 160.100.3.52 - BACKUP, switch B<o:p></o:p>
    DREX1 is located in another network. It is in another building connected via vpn. The ip
    for DREX1 is 192.170.100.51. There is no BACKUP network for this server<o:p></o:p>
    (These are not the actual address, but the concept is similar)<o:p></o:p>
    The dag network configuration is as follows.<o:p></o:p>
    DAGNetworkDR has a subnet of 192.170.100.0/24; Network Interface of 192.170.100.51<o:p></o:p>
    DAGNetwork01 (LAN) has a subnet of 192.168.100.0/24; Network Interfaces of 192.168.100.51
    and 192.168.100.52<o:p></o:p>
    DAGNetwork02 (BACKUP) has a subnet of 160.100.3.0/24; Network Interfaces of
    160.100.3.51 and 160.100.3.52<o:p></o:p>
    Last Friday, 3/7/14 something happened.<o:p></o:p>
    To lead up to that, I noticed the issue because we have DPM 2010 making 15minute express
    full backups of our exchange databases when I reviewed the backup jobs, the
    exchange database backups using dpm2010 failed all weekend long starting at
    11:00pm that Friday.<o:p></o:p>
    The issue that occurred was that the database copies went to a status of Disconnected and
    Resynchronizing, and after about 30 seconds, the status would show
    Resynchronizing.<o:p></o:p>
    All attempts to resolve this with the research methods failed.  Deleting the
    logs and reseeding failed, suspending, then rebuilding failed.<o:p></o:p>
    Creating a new database worked, creating a database copy worked, but seeding the copy
    failed. All with the same pattern, a status of Disconnected and
    Resynchronizing, and after about 30 seconds, and the status would show
    Resynchronizing.<o:p></o:p>
    The database was mounted on EX2, and seeding to EX1
    DREX1 has no issues.<o:p></o:p>
    All the DAG networks were in a good status, all Green and all up.  I could ping from
    EX1 to EX2 and from EX2 to EX1 on both networks fine.<o:p></o:p>
    In the middle of the week, about 3/12/14, we resolved the issue. We rebooted EX1, no
    fix, but after rebooting EX2, it all resolved.  It did a fail over to DREX1. I think it failed to do the failover to EX1. However, when it came back up, it was happy, it was reseeding slowly. After about an hour and a half, it
    was all healthy.<o:p></o:p>
    We thought it was all good EXCEPT, on Friday night at 11:00pmish on 3/14/14, the same
    issue occurred; a status of Disconnected and Resynchronizing, and after about
    30 seconds, the status would show Resynchronizing.<o:p></o:p>
    The database was mounted on EX2 and seeding to EX1 again. I did not notice the
    issue until I performed my review of the backup jobs again on Monday.<o:p></o:p>
    Late Monday morning, my boss and I were looking at more articles, but all the same results,
    delete the database copy and reseed. Which we knew would not work.<o:p></o:p>
    I mentioned that it was strange because the networks were all up. What made us try the next
    scenarios was just "luck" I guess. I made a statement about not remembering
    which networks DAG replication took place over.  Again, we have a LAN and
    a BACKUP network.  Bother were listed to use replication, so we decided to
    disable replication over the BACKUP network in the EMC. Immediately, the
    database copies switch to healthy and started seeding.<o:p></o:p>
    Later that evening I enabled replication over the BACKUP network and it failed again with
    a status of Disconnected and Resynchronizing, and after about 30 seconds, the
    status would show Resynchronizing.  I disabled the replication over the
    BACKUP network and the database copy went back to healthy<o:p></o:p>
    The last round of windows updates were pushed to these boxes the Friday before the
    issues began, 2/28/14.  I don't believe the updates to be the issue. <o:p></o:p>
    Can anyone make any suggestions as to where to look?  It worked with the BACKUP
    network enabled to so long that is eluding me as to why the issues would start
    all of a sudden. I am also confident that the switch configuration for the
    BACKUP network has not been changed at all in at least 2-3 months.  I am
    still open to look at the config. Please help if you can.<o:p></o:p>
     <o:p></o:p>
    Lastly, this is more of a monitoring thing, if anyone can suggest a good way to monitor
    the database copies so that I can be alerted when the status changes.  We
    use WhatsUpGold 16.2 to monitor the devices.  <o:p></o:p>
    Thanks for any future assistance.<o:p></o:p>
    Ian
    Ian

    You don't have to enable the Replication on Backup Network.
    You should collapse your DAG network.
    http://blogs.technet.com/b/timmcmic/archive/2011/09/26/exchange-2010-collapsing-dag-networks.aspx
    http://blogs.technet.com/b/samdrey/archive/2012/12/04/exchange-2010-dag-collapse-the-mapi-networks-in-a-dag.aspx
    Post the result of Below command:
    Get-DatabaseAvailabilityGroupNetwork 
    Cheers,
    Gulab Prasad
    Technology Consultant
    Blog:
    http://www.exchangeranger.com    Twitter:
      LinkedIn:
       Check out CodeTwo’s tools for Exchange admins
    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.

  • Exchange 2010 DAG and NIC Teaming | Team broke and now cluster doesn't see the box

    Hello everyone,
    Question for the group. We are running an Exchange 2010 DAG cluster with four boxes with active/passive DB's on each of the four boxes. We run HP boxes for Exchange and the NIC's are teamed. Now mind you, I arrived at this company well after this environment
    was set up so I have inherited everything here.
    What has happened is one of the four boxes had an issue after an otherwise normal reboot. We could no longer ping it. Logged into the ILO and was able to log into the server, only to discover its team NIC was messed up and our only option was to rebuild
    the team NIC. So I rebuilt the team and gave it the original IP/subnet/gateway it had. The server reboots and we can ping it.
    However, when I look in the EMC at Database Management, all the passive copies that were on that DB are showing Failed. Ok, so I suspend the copy and attempt to reseed. No luck. I get the error "The NetworkManager has not yet been initialized"
    and it references my machine which had the bad team NIC. Curious, I go look at my Failover Cluster Manager.
    Looking at the fail over cluster manager, I notice a X on the cluster and I am showing my server which had the bad team NIC being listed as down. I attempted to start the cluster service for the machine, but it did not work.
    Is there something I a missing? What are my exact steps to take to get my cluster configuration working correctly again and see my box? Do I need to remove the box from the DAG, evict it from the cluster, and rejoin it again? If so, can I do this without
    affecting the other three machines so users don't notice any outage?
    Thanks for any help you can give!

    Hi Kyle,
    I am Chetan Savade from Symantec Technical Support Team & sorry to know the problem you faced because of SEP.
    Symantec recommends that the Exchange server's OS always be protected by the latest available release of SEP.  The Exchange server's message flow and Information Store must be protected by a dedicated mail security product, such as Symantec
    Mail Security for Microsoft Exchange. 
    For cluster environment I would recommend to follow this article:
    Installing the Symantec Endpoint Protection client to a Windows cluster server
    http://www.symantec.com/docs/TECH91154
    Installation best practices for Endpoint Protection on Windows servers
    http://www.symantec.com/docs/TECH91154
    Best Regards,
    Chetan

  • Exchange 2010 dag with only 2 servers

    I have my exchange 2010 running on win server 2012 R2 and I have installed another exchange 2010 server with the same FQDN and I want to configure DAG with only these 2 servers. Can I implement DAG with only 2 servers?Can the mailbox,client access and hub
    transport roles still work with only 2 servers? If Mailbox and Hub transport could work it will be fine for me as CAS switchover can always be done manually.My main concern is securing data. Any article on how to implement that would be highly appreciated.
    Thanks

    Hi,
    Yes, we can create a DAG with 2Servers and Each server can have upto 100 database depends upon Standard or Enterprise.
    Here is a great link that explains about DAG
    Exchange 2010 DAG Installation
    Mark as Answer/Helpful Post if it helps
    Regards.
    Naren Neelam, Messaging Consultant, ITBigBang (P) Ltd Www.ITBigBang.Com | Hire Us for Messaging Consulting

  • Exchange 2010 dag migration to exchange 2013 help

    If I want to use the Same mailbox servers and upgrade them to exchange 2013, what is the recommended way? I have a 4 member exchange 2010 DAG.
    I was thinking (where A,B,C,D are mailbox servers, A,B are in one site and C,D another active directory site):
     A  B  -------------- C D
    Remove A and C from the 2010 DAG, Format and install Exchange 2013 on both, and create a 2013 dag :
    A1 ----------- C1
    B ------------- D
    Then move across all mailboxes from B to A1. Upgrade B and D to exchange 2013 and put them in the DAG
    A1 B1 ----------- C1 D1
    Does this make sense ? I am trying to do this while having as much high availability as possible during the transition. I have no extra servers so I have to use the existing server hardware.
    Anand_N

    Hi,
    When you want to migrate a DAG from Exchange 2010 to Exchange 2013, you should first remove the mailbox server from the DAG in Exchange 2010 and then move mailboxes from database in this mailbox server to Exchange 2013. After all mailboxes have been migrated
    to the Exchange 2013 mailbox server, when the new server is ready, it can become a member of the Exchange 2013 DAG.
    Here is an article about migration DAG from Exchange 2010 to Exchange 2013 for your reference.
    Migrating an Exchange 2010 DAG to Exchange 2013
    http://windowsitpro.com/blog/migrating-exchange-2010-dag-exchange-2013
    Note: Microsoft is providing this information as a convenience to you. The sites are not controlled by Microsoft. Microsoft cannot make any representations regarding the quality, safety, or suitability of any software or information found there. Please make
    sure that you completely understand the risk before retrieving any suggestions from the above link.
    Best regards,
    Belinda
    Belinda Ma
    TechNet Community Support

  • Exchange 2010 server backup.

    What is the best practise to take the Exchange server 2010 servers.
    We have 4 servers in DAG with 1 active and 3 passive copies.  We are currently taking Full backup in the weekend and differential backup daily. We use HP Data Protector for taking backup. Current DB size is 450GB.
    What is the prefered method of backup (Differential or Incremental).
    If we take incremental backup will it purge logs from all the exchange servers in DAG.

    Hi,
    Yes, Incremental backup will truncate the transaction log files. Here is an related blog below, please refer to the "DPM 2010 (beta) or every other 3rd party" section.
    Backup issues and limitations with Exchange 2010 and DAG
    http://blogs.technet.com/b/ehlro/archive/2010/02/13/backup-issues-and-limitations-with-exchange-2010-and-dag.aspx
    A similar thread which may help you for your reference.
    http://social.technet.microsoft.com/Forums/en-US/e76d57a5-5536-4c43-95b8-ea2a636ba054/dpm-2010-and-exchange-dag-recommended-way-to-backup
    Hope this helps.
    Best regards,
    Belinda
    Belinda Ma
    TechNet Community Support

  • Removing mailbox server from Exchange 2010 DAG - Node not fully cleaned up

    Hi,
    We are in the process of decommissioning some old Exchange 2010 servers. I have just attempted to remove one of our mailbox servers and have received a number of errors, related to its removal from the DAG.
    I have already removed the Public Folder replicas and all mailbox database copies. I then tried to evict the server from the DAG and received the following warning:
    I then ran the command that was recommended, from another server and in an administrative command prompt:
    I have looked in this log on both the server I ran the command on, a CAS server, and the mailbox server itself and can see no evidence of this error 2 that is listed here.
    If I run the Powershell command to look at the DAG I cannot see it there any more. I am assuming that this error message might only be an issue if I was planning on re-using this server somewhere else in Exchange or perhaps adding this server to another
    DAG. But I just want to decommission it.
    Can anyone put my mind at rest as to whether this is a genuine concern or not. If not my next step is to uninstall Exchange 2010 from this server and turn it off.

    Hi Belinda,
    I get the following:
    [PS] C:\Windows\system32>Get-DatabaseAvailabilityGroup
    Name             Member Servers                                                         
    Operational Servers
    DAG01            {WSR-EX-01, WSR-EX-101, WSR21}
    It doesn't look like that server is part of the DAG any more as far as I can see. I just can't explain that error message reported from the cluster command. As far as I can see the server is isolated. I already removed its database copies as a pre-requisite
    before removing it from the DAG as per this guide:
    http://www.experts-exchange.com/Software/Server_Software/Email_Servers/Exchange/Q_27039933.html
    I found a nice little script that gave reports on your Public Folder configuration and it highlighted that some of our folders were not being replicated properly across our mailbox servers. This helped me in removing these services from the server.
    It is currently doing nothing at this time as far as Exchange Services are concerned.

  • Exchange 2010 DAG DB Failed, low disk space

    We have a DB vol that's run out of space. We have 3 other copies, 2 others are healthy. Tlogs have filled the disk. The DB is in Failed status with the following errors:
    ErrorMessage                     : The Microsoft Exchange Replication service encountered an error while
                                       attempting to start a replication instance for jdcdb04\JDCXMETSP001.
    If the
                                       copy doesn't recover automatically, administrator intervention
    may be
                                       necessary. Error: No space left on disk (-1808)
    ErrorEventId                     : 4126
    What is the best way to recover from this?  Should we reseed?  It's large and will take hours to reseed.  Can we move off already committed Tlogs , and then roll in the logs by mounting (Get-MailboxDatabase DBName -status reports it's mounted) ,
    or using eseutil to roll the logs in?
    Any feedback is appreciated.

    Hi,
    Based on your description, you have 3 other copies, so I understand that you have one active database and three passive copy in your environment. About the passive copies, two are healthy and one is failed. Is it right?
    You need to perform full backup on active database to truncate transaction log files. After that, please reseed the failed database copy.
    Best regards,
    Belinda Ma
    TechNet Community Support

  • Exchange 2010 DAG Replication - too much data crossing the wire

    I’m replicating 3 Exchange databases from our production active Exchange 2010 server across the WAN to another passive Exchange 2010 server at our DR site. 
    The Exchange server at the DR site does not have any active databases, i.e. no users are hitting that server. 
    We are running Update Rollup 8 for Exchange Server 2010 SP1 on both these servers. 
    The two sites are connected via a 10Mb/s MPLS connection and all the databases are in sync and the replication is working fine. 
    I have setup a network sniffer at the primary site and see a sizable amount of data crossing the wire from the production Exchange server to the DR Exchange server. 
    When I query the production Exchange server using the Tracking Log explorer and only choose the EventID ‘RECEIVED’ that should show me the amount of data that has been committed to the database. 
    If I choose a date range that is the same exact range that I have used to capture the raw data with my sniffer the amount of data the sniffer shows crossing the wire is 10 fold compared to what the Tracking Log Explorer shows. 
    If I actually count up the data in the LOG files it is about 20% more than what is crossing the wire but that seems to be because the DAG is compressing the data. 
    If in a one hour timeframe there is 500MB of data crossing the wire to the DR Exchange server the tracking log explorer will show only 50MB. 
    I would like to know why the data crossing the wire far exceeds the amount of data that is truly being sent/received from the primary exchange server. 
    Perhaps I’m simply not getting a true view of the amount of data being committed to the exchange server using the Tracking Log Explorer. 
    Maybe there is a better way to report how much data is being committed to the exchange databases. 
    Any assistance would be appreciated…

    Two things.
    1 - you must update Exchange to SP3 and a recent RU.  Willard has already provided the links, which point back to my blog if you want to see the lifecycle map for Exchange 2010.  SP1 has been out of support since January 2013.  Time to move
    on pllease
    2 - Looking at the tracking log is not sufficient.  I would not expect that to show everything.
    I want to know what traffic you see as excessive.  What ports are you seeing used here?
    My money is on content indexing.  CI will use additional traffic over and above log repl traffic.  Expect CI traffic to be roughly the same again.  You can test this by disabling CI on the database or stopping the services on the DR server. 
    To disable the CI for the database:
    Set-MailboxDatabase DBName -indexEnabled $False
    Or stop the Exchange search services on the DR box to leave production unaffected.
    Again - you need to update Exchange.  You would be better to do that now rather than when something breaks and Microsoft support cannot fully assist you since you are not uptodate.    I'll leave discussion of the security issues resolved
    in recent Exchange RUs aside. 
    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.

  • Error vss writers exchange 2010 windows backup

    I have this error, when running windows backup on server with exchange 2010:
    Writer name: 'Microsoft Exchange Writer'
       Writer Id: {76fe1ac4-15f7-4bcd-987e-8e1acb462fb7}
       Writer Instance Id: {400de426-a419-459f-a953-91026051bf91}
       State: [8] Failed
       Last error: Retryable error
    I've rebooted the server several times, but without success.
    Does anyone have any idea?
    Thank you
    Cristiano Ianiri

    Hi,
    Do you have any running AV software? If there is, please set the AV Exclusion on database file or disable AV temporarily.
    From the description, you have restarted server several times. It didn't work. To narrow down the issue, could you please restart the Microsoft Exchange Information Store service again on the server where you did backup at your convenience?
    After that, please run the VSSadmin list writers command to determine whether the writer state returns to a stable state.
    Besides, please check the database that you want to backup to see if it is in healthy state. You can use the get-mailboxdatabasecopystatus cmdlet to check this.
    If you have checked steps above, please increase the diagnostic logging level for the Exchange Writer and check if there is any clues.
    Best regards,
    Belinda
    Belinda Ma
    TechNet Community Support

  • Migrating Exchange 2007 to Exchange 2010; initial backup slow...

    Just migrated about 300 Exchange 2007 mailboxes to Exchange 2010 database.  Temporarily using Windows Server Backup for Server 2008 R2 to backup the databases to a remote storage.  Initial full backup of the database is taking longer than 24 hours
    for a 235 GB database and what is now 334,000 log files...as a result of the migration.  Seems like a long time.
    I tested network connectivity to and from the SAN, and the Exchange server and there are no server or disk I/O issues.  Server is running 16 cores and 48 GB RAM.  Is this normal for it to take this long to run through the consistency checks and
    backups for this amount of data?
    Fuel

    Download Exchange server recovery tool that gives you best EDB recovery app for repairing EDB file data to PST Outlook format quickly. All users
    can recover their Exchange EDB file and fix EDB to PST file with best EDB file converter software. Software works on whole versions of Exchange EDB file.
    Read more:
    http://www.microsoft.edbtopsthelp.com/

  • Exchange 2010 DAG error adding members

    I am attempting to setup a DAG with two exchange 2010 server running sp2.  The DAG is there but when I try to add either server I get the following error:
    Summary: 1 item(s). 0 succeeded, 1 failed.
    Elapsed time: 00:00:11
    EXCHANGE
    Failed
    Error:
    A server-side database availability group administrative operation failed. Error: The operation failed. CreateCluster errors may result from incorrectly configured static addresses. Error: An error occurred while attempting a cluster operation. Error: Cluster
    API '"CreateCluster() failed with 0x35. Error: The network path was not found"' failed. [Server: EXCHANGE.domain.local]
    An Active Manager operation failed. Error An error occurred while attempting a cluster operation. Error: Cluster API '"CreateCluster() failed with 0x35. Error: The network path was not found"' failed..
    The network path was not found
    Click here for help...
    http://technet.microsoft.com/en-US/library/ms.exch.err.default(EXCHG.141).aspx?v=14.2.328.9&t=exchgf1&e=ms.exch.err.ExC9C315
    Warning:
    The operation wasn't successful because an error was encountered. You may find more details in log file "C:\ExchangeSetupLogs\DagTasks\dagtask_2013-01-16_01-21-06.060_add-databaseavailabiltygroupserver.log".
    Exchange Management Shell command attempted:
    Add-DatabaseAvailabilityGroupServer -MailboxServer 'EXCHANGE' -Identity 'dag1'
    Elapsed Time: 00:00:11
    The log file doesn't seem to have much more info but I can include it if anyone thinks it will help.  Please let me know what you think, thanks!
    Dan

    Update:  It was the remote registry service being turned off.  I enabled it and was able to add my primary exchange server to the dag but the other server now fails with:
    A server-side database availability group administrative operation failed. Error: The operation failed. CreateCluster errors may result from incorrectly configured static addresses. Error: An error occurred while attempting a cluster operation. Error: Cluster
    API '"AddClusterNode() (MaxPercentage=12) failed with 0x800706d3. Error: The authentication service is unknown"' failed. [Server: exchange.local]
    Dan

  • Exchange 2010 DAG's - 4 Mailbox Servers in 2 Datacenters with per site redundancy

    I recently came into a company with an Exchange 2010 sp1 environment where there are two data centers in two different subnets (10.70.62.0 and 10.80.56.0)
    Each datacenter has 2 mailbox servers (4 servers total) but their plan was to have only enough storage space to host 1/2 of the databases on each server. So instead of server 1, 2, and 3 being copies of each other, ServerA is a copy of ServerA and ServerB
    is a copy of ServerB
    Server     -            Databases
    Site1mbA - Primary Database copy 1-10
    Site1mbB - Primary Database copy 11-20
    Site2mbA - Secondary Database copy 1-10
    Site2mbB - Secondary Database copy 11-20
    Normally if there were enough storage space I would keep a copy on each server and replicate within 1 DAG.
    However since not all the databases are on each member server do I need to use 2 DAGS? 1st for Site1mbA & Site2mbA and 2nd for Site1mbB & Site2mbB
    I feel I would need the 2 DAG's because with 1 DAG if Site1mbA server went offline the DAG couldn't switch to the Failover Cluster IP for Site2 because there are still active mailboxes in Site1 on Site1mbB.
    Does that make any sense or am I over complicating it?
    Thanks!

    There is no rule that every server in a DAG must have a copy of every Database. You could do this with one DAG, but if the WAN goes down then the DAG will be active in only a single site not both.
    http://technet.microsoft.com/en-us/library/dd979781(v=exchg.141).aspx#FourTwo
    As illustrated in the previous example, using a single four-member DAG extended across two datacenters can provide high availability and site resilience for the mailbox services and data. However, if a WAN outage occurs, only the primary datacenter retains
    service because it contains the majority of the voters. The datacenter with the minority of voters loses majority, and the DAG members in that datacenter lose quorum and go offline.
    To deploy highly available Mailbox servers in a multiple datacenter environment, where each datacenter is actively serving a local user population, we recommend that you deploy multiple DAGs, where each DAG has a majority of voters in a different datacenter,
    as illustrated in the following figure.
    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.

Maybe you are looking for