none
Exchange 2013 backups - logs not being truncated RRS feed

  • Question

  • I have an issue with a fairly new Exchange 2013 (SP1) setup. Currently 5 Exchange Mailbox servers all part of a DAG. After backups run successfully, logs are not being truncated. I use HP Data Protector 8.0 (fully patched) with VSS integration. The backups run successfully from the App point of view, no errors. However the logs do not truncate. I get 2 confusing entries in the App event log on the Exchange MB servers:

    ESE 225: Information Store - DB1 (16632) DB1: No log files can be truncated. 

    ollowd by: MSExchangeRepl 2046: The Microsoft Exchange Replication service VSS Writer instance ac70fe72-4463-41a9-b47e-5c5e5d0c82a4 has successfully completed the backup of database 'DB1'.  

    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.

    No log files are truncated. This is backing up the ACTIVE copy of the DB.

    When I manually run a backup via Windows Backup, I get the same results. No errors, ESE 225 and ESE 2046. Both ways I get a MSMQ 2230 event "A backup (components: registry, sysq, mapping, queue, storage) was completed successfully." and an ESE 2006 "Information Store - DB1 (16632) Shadow copy instance 15 completed successfully. " and MSExchangeRepl 2033 "The Microsoft Exchange Replication service VSS Writer (Instance 80e81897-81d9-489d-ab11-b9f08126d737) has successfully processed the backup completion event." and MSExchangeRepl 2037 "The Microsoft Exchange Replication service VSS Writer (Instance 80e81897-81d9-489d-ab11-b9f08126d737) backup has been successfully shut down."

    This seems to be happening across all of my 2013 mailbox servers.


    Wednesday, June 4, 2014 3:00 PM

Answers

  • Hi Troy12n,

    Thanks for your update.

    Since all the database copies don’t have any issue but no logs being truncated after successful full backup of the active database, could you please let me know how many logs within a database’s folder now?

    Also, as mentioned before, we can enable circular logging for one database, it will truncate the logs by automatically, then disable the circular logging. After some time when the database generate some logs, please try to take a full backup of the database use Windows Server Backup to see if the logs will be truncated after full backup.

    Or if needed, to know if it’s a database related issue, we can schedule a down time to dismount one affected database, use “eseutil /mh” to make sure the database is in “Clean Shutdown” status, then remove all the existing logs manually, mount the database, it will start a new sequence for the logs, then we can try a full backup to see if the logs can be truncated normally.


    Regards, Eric Zou

    Friday, June 6, 2014 9:06 AM
    Moderator

All replies

  • Something else, I noticed this in the logs...

    Event WSBEXchange id 0

    The description for Event ID 0 from source wsbexchange cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

    If the event originated on another computer, the display information had to be saved with the event.

    The following information was included with the event:

    Service started/resumed

    Wednesday, June 4, 2014 4:00 PM
  • Have you verified if all copies of Databases is in healthy state? If there are unhealthy passive copied then these logs are required for log shipping.

    Have you ever deleted logs manually from the folder? I would suggest that you create a new Database with required number of passive copied, move couple of mailboxes, run the backup and verify if that helps.

    Else, you may have to enable circular logging for a while, to truncate all logs and disable it again.


    - Sarvesh Goel - Enterprise Messaging Administrator

    Wednesday, June 4, 2014 4:29 PM
  • Circular logging is not enabled

    All DB's are in a healthy state, including content index. I have never deleted any files manually. I am not going to move mailboxes and re-create databases... that is silly. Anyone here have any real suggestions?

    [PS] C:\>Get-MailboxDatabase | Get-MailboxDatabaseCopyStatus

    Name                                          Status          CopyQueue ReplayQueue LastInspectedLogTime
                                                                  Length    Length
    ----                                          ------          --------- ----------- --------------------
    MB1-DB0\EX2013-MB1                            Mounted         0         0
    MB3-DB0\EX2013-MB3                            Mounted         0         0
    MB2-DB0\EX2013-MB2                            Mounted         0         0
    DB1\EX2013-MB1                                Mounted         0         0
    DB1\EX2013-MB2                                Healthy         0         0           6/4/2014 1:38:20 PM
    DB1\EX2013-MB3                                Healthy         0         0           6/4/2014 1:38:20 PM
    DB1\EX2013-MB1-DR                             Healthy         0         0           6/4/2014 1:38:20 PM
    DB1\EX2013-MB2-DR                             Healthy         0         0           6/4/2014 1:38:20 PM
    DB2\EX2013-MB1                                Mounted         0         0
    DB2\EX2013-MB3                                Healthy         0         0           6/4/2014 1:36:22 PM
    DB2\EX2013-MB2                                Healthy         0         0           6/4/2014 1:36:22 PM
    DB2\EX2013-MB2-DR                             Healthy         0         0           6/4/2014 1:36:22 PM
    DB2\EX2013-MB1-DR                             Healthy         0         0           6/4/2014 1:36:22 PM
    DB3\EX2013-MB2                                Mounted         0         0
    DB3\EX2013-MB1                                Healthy         0         0           6/4/2014 1:48:30 PM
    DB3\EX2013-MB3                                Healthy         0         0           6/4/2014 1:48:30 PM
    DB3\EX2013-MB1-DR                             Healthy         0         0           6/4/2014 1:48:30 PM
    DB3\EX2013-MB2-DR                             Healthy         0         0           6/4/2014 1:48:30 PM
    DB4\EX2013-MB2                                Mounted         0         0
    DB4\EX2013-MB3                                Healthy         0         1           6/4/2014 1:51:59 PM
    DB4\EX2013-MB1                                Healthy         0         1           6/4/2014 1:51:59 PM
    DB4\EX2013-MB2-DR                             Healthy         0         1           6/4/2014 1:51:59 PM
    DB4\EX2013-MB1-DR                             Healthy         0         1           6/4/2014 1:51:59 PM
    DB5\EX2013-MB3                                Mounted         0         0
    DB5\EX2013-MB1                                Healthy         0         0           6/4/2014 1:45:06 PM
    DB5\EX2013-MB2                                Healthy         0         0           6/4/2014 1:45:06 PM
    DB5\EX2013-MB1-DR                             Healthy         0         0           6/4/2014 1:45:06 PM
    DB5\EX2013-MB2-DR                             Healthy         0         0           6/4/2014 1:45:06 PM
    DB6\EX2013-MB3                                Mounted         0         0
    DB6\EX2013-MB2                                Healthy         0         0           6/4/2014 1:44:52 PM
    DB6\EX2013-MB1                                Healthy         0         0           6/4/2014 1:44:52 PM
    DB6\EX2013-MB2-DR                             Healthy         0         0           6/4/2014 1:44:52 PM
    DB6\EX2013-MB1-DR                             Healthy         0         0           6/4/2014 1:44:52 PM
    MB1-DR-DB0\EX2013-MB1-DR                      Mounted         0         0
    MB2-DR-DB0\EX2013-MB2-DR                      Mounted         0         0

    Wednesday, June 4, 2014 6:08 PM
  • Hi Troy12n,

    Please help to confirm the following:

    1. All the database copies content index status healthy without any issue?
    2. All the logs not truncated after backup or only some logs truncated but others remain in the folder?
    3. Logs not truncated on the active or passive copy?
    4. How did you perform the backup? Did we take a full backup of the databases?
    5. Do we have lagged database copies?

    Since we can find the log truncation event in event log, please wait for the log truncation to take effect. Also, for log truncation within DAG about highly available (non-lagged) mailbox database copies, the following must be true:

    • The log file has been backed up, or CRCL is enabled.
    • The log file is below the checkpoint.
    • The other non-lagged copies of the database agree with deletion.
    • The log file has been inspected by all lagged copies of the database.

    For truncation to occur on lagged database copies, the following must be true:

    • The log file is below the checkpoint.
    • The log file is older than ReplayLagTime + TruncationLagTime.
    • The log file is deleted on the active copy of the database.

    Please refer to this article:

    Title: Backup, Restore, and Disaster Recovery

    Link: http://technet.microsoft.com/en-us/library/dd876874(v=exchg.150).aspx#VSS

    Meanwhile, as a test, we can enable circular logging for one database, it will truncate the logs by automatically, then disable the circular logging. After some time when the database generate some logs, please try to backup the database to see if the logs will be truncated after a successful full backup.


    Regards, Eric Zou

    Thursday, June 5, 2014 6:57 AM
    Moderator
  • Here are answers to your questions:

    All the database copies content index status healthy without any issue?

    - yes, see the previous post where I posted the results of get-mailboxdatabasecopystatus

    All the logs not truncated after backup or only some logs truncated but others remain in the folder?

    - no logs truncated.

    Logs not truncated on the active or passive copy?

    - no logs truncated, I am backing up the active copy only

    How did you perform the backup? Did we take a full backup of the databases?

    - full backup via both HP data protector, which uses VSS, and also same result when I run Windows Server Backup

    Do we have lagged database copies?

    - No, no lagged copies


    • Edited by Troy12n Thursday, June 5, 2014 11:15 AM
    Thursday, June 5, 2014 11:14 AM
  • Hi Troy12n,

    Thanks for your update.

    Since all the database copies don’t have any issue but no logs being truncated after successful full backup of the active database, could you please let me know how many logs within a database’s folder now?

    Also, as mentioned before, we can enable circular logging for one database, it will truncate the logs by automatically, then disable the circular logging. After some time when the database generate some logs, please try to take a full backup of the database use Windows Server Backup to see if the logs will be truncated after full backup.

    Or if needed, to know if it’s a database related issue, we can schedule a down time to dismount one affected database, use “eseutil /mh” to make sure the database is in “Clean Shutdown” status, then remove all the existing logs manually, mount the database, it will start a new sequence for the logs, then we can try a full backup to see if the logs can be truncated normally.


    Regards, Eric Zou

    Friday, June 6, 2014 9:06 AM
    Moderator
  • Hi Troy12n,

    If we have any update on this issue, please feel free to let me know.


    Regards, Eric Zou

    Thursday, June 12, 2014 3:36 AM
    Moderator
  • This could be the issue with exchange search indexer service not allowing exchange replication to truncate logs.

    dismount databases, stop the Exchange Information Store Service then delete catalogs from all database copies (active and passive) and start Exchange Information Store Service and mount the databases

    • Proposed as answer by TontonLud Friday, March 16, 2018 11:16 AM
    Thursday, June 12, 2014 4:33 AM
  • You made my day Mallik, thank you.

    I had the issue with a single Exchange 2016 server.

    Stop Exchange Search services

    Stop-Service MsExchangeFastSearch
    Stop-Service HostControllerService

    Then go to you EDB folders, to find out

    Get-MailboxDatabase | fl Name, EdbFilePath

    Delete rename or move to another location the Catalog or Guid folder.

    Then restart your search services

    Start-Service HostControllerService
    Start-Service MsExchangeFastSearch

    Then try an incremental backup.




    • Edited by TontonLud Friday, March 16, 2018 11:25 AM
    Friday, March 16, 2018 11:21 AM