none
SCDPM 2012 R2 Backup with Exchange 2013 SP1 CU6 DAG RRS feed

  • Question

  • Hi,

    Correct me if I'm wrong.

    DAG setup as follow,

    1. Server1\MBX01 (Active)

    2. Server2\MBX01 (Passive)

    DPM

    1. Select Full for Active

    2. Select Copy for Passive

    We always select active copy so that logs can be purge.  On the other hand, if we select backup passive copy and purge logs. My guess is logs will be replicated from Active copy.  

    How about when active copy change passive copy?  Do I need to change the backup option?

    Please advice. Thanks.


    Kelvin Teang


    Monday, December 1, 2014 8:20 AM

All replies

  • Hi Kevin,

    The Full Backup will backup the databases and the log files and then truncate the log files. Copy backup will just not truncate the log files.

    I'm a little unclear about "active copy change passive copy". You need to keep at least 1 full backup if you need to backup more than 1 Exchange database. 


    If you have any feedback on our support, please send to tnfsl@microsoft.com.

    Tuesday, December 2, 2014 5:28 AM
    Moderator
  • Greetings.

    Typically DAG consists of Active and passive.  In our environment, we have Active, passive, passive.

    Scope is to backup active copy as Full without backing up both passive copies.  If not, I will be having X size multiple by 3 eg. backup size is 20GB * 3 copies = 60GB.

    If we select Active as Full backup now, what happen when DAG change to passive, Active, passive.  In this case, I think DPM has to change accordingly else DPM is backing up passive copy.


    Kelvin Teang

    Tuesday, December 2, 2014 7:42 AM
  • Hi,

    You just need to perform a full backup from one copy, be that passive or active, it does not matter.  Should the dag member hosting the copy you are doing the full backup on goes down, you have two choices.

    1) Wait for that dag member to come back online and full backups will continue.

    2) Change full backups to be on a dag member that is online and change the full to read only for the dag member that is down.   You can change them back after the node comes back up.


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.

    Tuesday, December 2, 2014 11:12 PM
    Moderator
  • Hi,

    I think it does matter for 1 copy of full backup. For example,

    Case 1

    Full Backup - Active copy

    Results : Purged exchange logs and replicate to passive copy

    Case 2

    Full Backup - Passive copy

    Results : Purged exchange logs from Passive copy, DAG (Database copies) queue legth will increase follow by take the replication from active copy. In this case, exchange logs will never be purged.


    Kelvin Teang

    Wednesday, December 3, 2014 2:18 AM
  • Hi,

    I believe log truncation is negotiated amongst all dag nodes. I believe this blog covers the subject.

    http://blogs.technet.com/b/timmcmic/archive/2012/03/12/exchange-2010-log-truncation-and-checkpoint-at-log-creation-in-a-database-availability-group.aspx


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.

    Wednesday, December 3, 2014 4:09 AM
    Moderator
  • Hi,

    I have tested both cases; backup Full for active node - logs have truncation and replicated to passive node.

    However, backup Full for passive node does not truncate any logs. Hence, no errors when I executed vssadmin list writers.  Any advice?

    Many thanks.


    Kelvin Teang


    Monday, December 8, 2014 2:47 AM
  • Hi,

    It sounds like an exchange problem - this was the application event log entry on the passive node after taking an incremental backup from DPM.  After the below event was posted, I checked both active and passive node and the logs were successfully truncated.  It took a few minutes before the logs were deleted.

    Log Name:      Application
    Source:        MSExchangeRepl
    Date:          12/9/2014 3:40:46 PM
    Event ID:      2046
    Task Category: Exchange VSS Writer
    Level:         Information
    Keywords:      Classic
    User:          N/A
    Computer:      MJLC-E14-2.mjacquet.com
    Description:
    The Microsoft Exchange Replication service VSS Writer instance 5cc17efd-1acc-41c3-b68a-f667027032b6 has successfully completed the backup of database 'N1-MB2'. 

    Database log truncation has been requested for this database. Log truncation will occur on the active copy after the next log generation is created. Log truncation will occur automatically on the passive copies after that log file is copied.
    Event Xml:
    <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
      <System>
        <Provider Name="MSExchangeRepl" />
        <EventID Qualifiers="16388">2046</EventID>
        <Level>4</Level>
        <Task>2</Task>
        <Keywords>0x80000000000000</Keywords>
        <TimeCreated SystemTime="2014-12-09T23:40:46.000Z" />
        <EventRecordID>399855</EventRecordID>
        <Channel>Application</Channel>
        <Computer>MJLC-E14-2.mjacquet.com</Computer>
        <Security />
      </System>
      <EventData>
        <Data>5cc17efd-1acc-41c3-b68a-f667027032b6</Data>
        <Data>N1-MB2</Data>
      </EventData>
    </Event>


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT] This posting is provided "AS IS" with no warranties, and confers no rights.

    Tuesday, December 9, 2014 11:51 PM
    Moderator