Hi Guys,

 I backup my Active Directory server with KAV installed (MSDE as database). When SP kickstart, the event log show:

2010-01-08 13:08:51.83 server    Microsoft SQL Server  2000 - 8.00.760 (Intel X86)
    Dec 17 2002 14:22:05
    Copyright (c) 1988-2003 Microsoft Corporation
    Desktop Engine on Windows NT 5.2 (Build 3790: )

2010-01-08 13:08:51.83 server    Copyright (C) 1988-2002 Microsoft Corporation.
2010-01-08 13:08:51.83 server    All rights reserved.
2010-01-08 13:08:51.83 server    Server Process ID is 2712.
2010-01-08 13:08:51.83 server    Logging SQL Server messages in file 'C:\Program Files\Microsoft SQL Server\MSSQL\LOG\ERRORLOG'.
2010-01-08 13:08:51.85 server    SQL Server is starting at priority class 'normal'(2 CPUs detected).
2010-01-08 13:08:52.65 server    SQL Server configured for thread mode processing.
2010-01-08 13:08:52.66 server    Using dynamic lock allocation. [500] Lock Blocks, [1000] Lock Owner Blocks.
2010-01-08 13:08:52.74 spid4     Starting up database 'master'.
2010-01-08 13:08:53.12 server    Using 'SSNETLIB.DLL' version '8.0.760'.
2010-01-08 13:08:53.12 spid5     Starting up database 'model'.
2010-01-08 13:08:53.15 spid4     Server name is 'KRMIS001-SERVER'.
2010-01-08 13:08:53.15 spid4     Skipping startup of clean database id 4
2010-01-08 13:08:53.16 server    SQL server listening on 128.30.0.1: 1433.
2010-01-08 13:08:53.16 server    SQL server listening on 127.0.0.1: 1433.
2010-01-08 13:08:53.35 server    SQL server listening on TCP, Shared Memory, Named Pipes.
2010-01-08 13:08:53.35 server    SQL Server is ready for client connections
2010-01-08 13:08:53.38 spid5     Clearing tempdb database.
2010-01-08 13:08:54.47 spid5     Starting up database 'tempdb'.
2010-01-08 13:08:54.65 spid4     Recovery complete.
2010-01-08 13:08:54.65 spid4     SQL global counter collection task is created.
2010-01-08 13:13:59.35 spid51    Starting up database 'KAV'.
2010-01-08 13:13:59.82 spid51    Starting up database 'KAV'.
2010-01-08 13:14:00.12 spid51    Starting up database 'KAV'.
2010-01-08 13:14:00.41 spid51    Starting up database 'KAV'.
2010-01-08 13:14:31.55 spid7     Starting up database 'KAV'.
2010-01-08 13:15:40.57 spid51    Starting up database 'KAV'.
2010-01-08 14:57:25.73 spid58    Starting up database 'msdb'.
2010-01-08 14:57:31.96 spid12    This SQL Server has been optimized for 8 concurrent queries. This limit has been exceeded by 4 queries and performance may be adversely affected.
2010-01-08 14:58:10.80 spid57    Database model: IO is frozen for snapshot
2010-01-08 14:58:10.80 spid52    Database KAV: IO is frozen for snapshot
2010-01-08 14:58:10.80 spid58    Database msdb: IO is frozen for snapshot
2010-01-08 14:58:11.13 spid53    Database master: IO is frozen for snapshot
2010-01-08 14:58:15.26 spid53    Database master: IO is thawed
2010-01-08 14:58:15.29 spid58    Database msdb: IO is thawed
2010-01-08 14:58:15.26 spid57    Database model: IO is thawed
2010-01-08 14:58:15.29 spid52    Database KAV: IO is thawed
2010-01-08 14:58:16.22 backup    Database backed up: Database: model, creation date(time): 2002/12/17(14:36:05), pages dumped: 1, first LSN: 4:62:1, last LSN: 4:66:1, number of dump devices: 1, device information: (FILE=1, TYPE=VIRTUAL_DEVICE: {'{BC1EB5DB-4B0A-42FB-BFD3-50956C9ED0A3}3'}).
2010-01-08 14:58:16.30 backup    Database backed up: Database: master, creation date(time): 2010/01/08(13:08:52), pages dumped: 1, first LSN: 77:326:1, last LSN: 77:331:1, number of dump devices: 1, device information: (FILE=1, TYPE=VIRTUAL_DEVICE: {'{BC1EB5DB-4B0A-42FB-BFD3-50956C9ED0A3}2'}).
2010-01-08 14:58:16.34 backup    Database backed up: Database: KAV, creation date(time): 2010/01/08(13:13:59), pages dumped: 1, first LSN: 156:204:1, last LSN: 156:234:1, number of dump devices: 1, device information: (FILE=1, TYPE=VIRTUAL_DEVICE: {'{BC1EB5DB-4B0A-42FB-BFD3-50956C9ED0A3}1'}).
2010-01-08 14:58:16.34 backup    Database backed up: Database: msdb, creation date(time): 2002/12/17(14:41:12), pages dumped: 1, first LSN: 14:44:1, last LSN: 14:48:1, number of dump devices: 1, device information: (FILE=1, TYPE=VIRTUAL_DEVICE: {'{BC1EB5DB-4B0A-42FB-BFD3-50956C9ED0A3}4'}).
2010-01-08 14:58:36.93 spid5     This SQL Server has been optimized for 8 concurrent queries. This limit has been exceeded by 2 queries and performance may be adversely affected.

SP log

08-Jan-2010 14:56:55    service    100    service (build 43) started job manually as full
08-Jan-2010 14:56:55    service    104    3.5.0.3570 E94D-E3CF 8842-D386-F901-841F
08-Jan-2010 14:56:55    service    100    backup volume C:\ (NTFS)
08-Jan-2010 14:56:55    service    100    VDIFF was disabled and then enabled on C:\
08-Jan-2010 14:56:55    service    102    creating snapshot for  \\?\Volume{93ad9186-ad03-11d8-8733-806e6f6e6963}
08-Jan-2010 14:56:55    service    199    try snapshot by  VSS API by STC provider
08-Jan-2010 14:58:17    service    150    retrieving snapshot name.
08-Jan-2010 14:58:17    service    103    snapshot was created by  VSS API by STC provider. It took 1 minutes and 22 seconds
08-Jan-2010 14:58:17    service    104    image will be created by VDIFF
08-Jan-2010 14:58:17    sptask    111    sbrun -mdn ( sbvol -fi \\?\STC_SnapShot_Volume_21_0 \\?\Volume{93ad9186-ad03-11d8-8733-806e6f6e6963} : sbcrypt -50 : sbfile -wd smb://\\Backup_repo\IMAGES_REPO\KRMIS001-SERVER\KRMIS001_SERVER_C_VOL-b004.spf )
08-Jan-2010 14:58:17    sptask    110    Worker thread has started
08-Jan-2010 14:58:17    (loader)    112    corelogic version: 3.5.225.3570
08-Jan-2010 14:58:17    sbvol    107    throttled to 90% of the maximum speed
08-Jan-2010 14:58:17    sbvol    109    free space exclusion on
08-Jan-2010 14:58:17    sbvol    107    incremental tracking is on, generation count: 0
08-Jan-2010 14:58:17    sbvol    117    incremental tracking is on, 40963160 of 40963160 (100%) updated sectors
08-Jan-2010 14:58:19    sbvol    101    successfully opened volume \\?\Volume{93ad9186-ad03-11d8-8733-806e6f6e6963}
08-Jan-2010 14:58:19    sbvol    107    FAT system area sectors: 0
08-Jan-2010 14:58:19    sbvol    112    file \\?\STC_SnapShot_Volume_21_0\pagefile.sys excluded OK
08-Jan-2010 14:58:19    sbcrypt    109    filter started
08-Jan-2010 14:58:20    sbvol    107    disk MBR sectors: 1
08-Jan-2010 14:58:20    sbvol    107    first track sectors: 63
08-Jan-2010 14:58:20    sbvol    109    disk CHS 9399/255/63 partition 1
08-Jan-2010 14:58:20    sbcrypt    107    compression mode: 5
08-Jan-2010 14:58:20    sbcrypt    107    encryption mode: 0
08-Jan-2010 14:58:20    sbfile    620    Enter the user name to access smb://\\Backup_repo\IMAGES_REPO\KRMIS001-SERVER\KRMIS001_SERVER_C_VOL-b004.spf
08-Jan-2010 14:58:20    sptask    101    Answer was sent as UNICODE string
08-Jan-2010 14:58:20    sbfile    621    Enter the password to access smb://\\Backup_repo\IMAGES_REPO\KRMIS001-SERVER\KRMIS001_SERVER_C_VOL-b004.spf
08-Jan-2010 14:58:20    sptask    101    Answer was sent as UNICODE string
08-Jan-2010 14:58:20    sbfile    101    successfully opened file smb://\\Backup_repo\IMAGES_REPO\KRMIS001-SERVER\KRMIS001_SERVER_C_VOL-b004.spf
08-Jan-2010 15:06:01    sbfile    109    probing the target media for > 4GB files support
08-Jan-2010 15:09:09    sbvol    109    last volume read done
08-Jan-2010 15:09:09    sbcrypt    109    tail started
08-Jan-2010 15:09:09    sbcrypt    109    fini done
08-Jan-2010 15:09:09    sbfile    193    MD5:58db4d69637d207434f700e8cb3e5c75 smb://\\Backup_repo\IMAGES_REPO\KRMIS001-SERVER\KRMIS001_SERVER_C_VOL-b004.spf
08-Jan-2010 15:09:09    sbfile    200    OK smb://\\Backup_repo\IMAGES_REPO\KRMIS001-SERVER\KRMIS001_SERVER_C_VOL-b004.spf
08-Jan-2010 15:09:09    sbfile    109    fini done
08-Jan-2010 15:09:09    sbvol    109    incremental tracker reset OK
08-Jan-2010 15:09:09    sbvol    109    fini done
08-Jan-2010 15:09:09    sptask    114    Process finished successfully
08-Jan-2010 15:09:09    service    104    enforce retention policy
08-Jan-2010 15:09:12    service    105    snapshots were destroyed

i know MSDE have limited concurrent queries. Will this freeze server for long time when snapshot ( as log above show it took 1min 22 seconds to snapshot my c: drive)?

what is the best idea for this?

My SP is 1 job for C: Drive which contain OS, KAV and MSDE. Thanks

Thanks for help.

ahsiangsiang:

Will this freeze server for long time when snapshot ( as log above show it took 1min 22 seconds to snapshot my c: drive)?

During the snapshot process there is a restriction on any writes to used sectors on that volume, though there is often no restriction on any reads. In other words, the application should not be unresponsive - there will just be a restriction on write operations for the snapshot duration. In many cases, a snapshot may take less than 30 seconds, but it can depend on the server load, storage subsystem and state of the VSS framework.

It looks as though at least 46 seconds of the snapshot time (14:57:25.73 --> 14:58:11.13) is related to quiescing the KAV MSDE database. That makes sense if a basic snapshot takes about 30 seconds, and explains the snapshot duration you experienced.

If you are trying to reduce the snapshot time, then here are a few things to look into: 

1. Make sure that the VSS framework is in a good state. On a command line, use the same command as above, namely:
   vssadmin list writers
   If any writers are in state other than "Stable", then you need to fix it. There are some articles in the KB section of this website.
2. Try a backup out of hours, and see if the snapshot is quicker.
• If it is, then you need to look at the load on the server at the time of the backups. You can change the backup schedule to miss peak hours.  
• If the snapshot is not quicker, then see if anything is slowing down ShadowProtect and/or the VSS Framework. A good thing to check is any overly agressive Antivirus/Security application. If you can take the machine off the network out-of-hours and temporarily shutdown the security product, it might help confirm if this is a problem or not. If it is, then you'll need to contact the maker of that product for more information on how to minimize this.
3. I know that newer versions of MSDE can use the newer SQLServer VSS Writer, rather than the original MSDE VSS Writer. I'm not sure if one is faster than the other, but it may be worth check which writer is used and/or if an update is available from Kaspersky. You can check which writer is used (and its state) using the command provided above (vssadmin list writers).
4. Check with Kaspersky support and forum to see if there are any updates/issues related to VSS and snapshots. 

Regards

FT