none
System Attendant could not be online in Exchange Cluster Server RRS feed

  • 問題

  • Dear Sir,

     

    In my Exchange Site, there are three Exchange virtual servers as below.

    1. Windows Server 2000(domain controller) with Exchange 2000 Server Ent.

    2. Windows Server 2003(domain controller) with Exchange 2003 Server Ent

    3. Two Windows Cluster Server 2003(joined domain) with two Exchange 2003 Cluster Server.

     

    I found my Exchange 2003 Cluster Server(active node) could not start up after reboot.

    The Exchange System Attendant in the cluster resource could not be onlined although the Exchange System Attendant service started. I have to manually retry and retry to bring the Exchange System Attendant resource online in the Cluster Admin. It needed several hours to bring the resource online.

    So I have scheduled a task to run a cluster command to bring it online.

    After the Exchange System Attendand resource had onlined, other Exchange resources could be onlined.

     

    I found the errors in the application log shown below:-.

     

    Source: MSExchangeMU

    EventID: 1042

    Metabase Update failed to read the Configuration namespace property from the domain controller. Error code is 80040a01.

     

    Source: MSExchangeSA

    EventID: 1031

    One of the System Attendant's task is blocked.
    Function: ScStartVM

     

    Source: MSExchangeCluster

    EventID: 1019

    Exchange System Attendant (NODE1): Failed to bring this resource online because of timeout.

     

    Source: MSExchangeCluster

    EventID: 1003

    Exchange System Attendant (NODE1): Failed to bring the resource online.

    Source: MSExchangeCluster

    EventID: 1012

    Exchange Information Store Instance (NODE1): The RPC call to the service to take the resource offline failed.

     

    Source: MSExchangeCluster

    EventID: 1021

    Exchange System Attendant (NODE1): Failed to terminate this resource because of timeout.

     

    Source: MSExchangeCluster

    EventID: 1011

    Exchange System Attendant (NODE1): The RPC call to the service to bring the resource online failed.

    Source: MSExchangeCluster

    EvnetID: 1024

    Exchange System Attendant (NODE1): Failed to stop service 'MSExchangeSA'.

     

    I also have the error message prompt up in the System Manager when purged the mailbox,

    "unable to contact the global catalog server, facility win32 ID no:c0072019".

     

    Would you help me fix the problem or direct me to related kb ?

     

    Thanks in advance.

     

     

     


     

     

     





     

    2008年8月19日 上午 07:22

解答

  •  

    Dear customer:

     

    Form your picture1 and picture2 screenshot, I found the SERVERF-LAN node is running. From your previous cluster and event log, it seems that issue didn’t occur since July 21, 2008. Since the log file is too old, from the current information we are almost unable to determine the root cause.

     

    If the issue reoccurs in the further, please let know. I will follow up.

     

    From your post, I found that you posted two questions, however these two questions are unrelated. Please understand that our forum is an issue based service, meaning we usually respond to one question/issue per post. This will lessen the confusion for both of us, as well as ensure that our results are accurate and not a result of a test for a different question. I advice you to post the following question into a new post.

     

    “Also have the error message prompt up in the System Manager when purged the mailbox,

     

    "unable to contact the global catalog server, facility win32 ID no:c0072019”.

     

    Thanks for your cooperation and understanding.

     

    Rock Wang - MSFT

    2008年9月10日 上午 09:22
    版主
  • Dear Mr. Wang,

     

    Thank you for your help. Your job I always appreciate.

    I decided to re-install the Exchange Cluster Servers. I think those problems should be solved.

     

    Regards.

     

    2008年9月10日 下午 02:08

所有回覆

  •  

    Dear customer:

     

    In order to better troubleshoot the issue, please help collect the following information:

     

    1. Collect EXBPA log file:

     

    1. Download and install Microsoft .NET Framework 1.1 on a client computer from this

    link:

    <http://www.microsoft.com/downloads/details.aspx?FamilyID=262d25e3-f589-4842-8157-034d1e7cf3a3&DisplayLang=en>

    Note: If you have already installed Microsoft .NET Framework 1.1, skip this step.

     

    1. Download and install Exchange Server Best Practices Analyzer Tool from this link:

    <http://www.microsoft.com/downloads/details.aspx?familyid=DBAB201F-4BEE-4943-AC22-E2DDBD258DF3&displaylang=en>

    1. Log on the computer as Domain Administrator.
    2. Click Start, and then click Run.
    3.  In the Open box, type cmd, and then click OK.
    4. In the command window, locate the directory where Exchange Server Best Practices Analyzer Tool is installed. By default, the directory is C:\Program Files\ExBPA.
    5. Type exbpacmd.exe in the command line; and ExBPA begins to examine the server deployment and configuration.
    6.  Wait until ExBPA finishes its examination and shows you the directory where the output xml and log files are located.
    7. Copy the output xml and log files from the locations shown on the screen and send it to v-rocwan@microsoft.com.

     

    2. Download MPS Reports for Exchange from the following link and install it on each node, and then run it, send the .cab file to v-rocwan@microsoft.com.

     

    http://www.microsoft.com/downloads/details.aspx?FamilyId=CEBF3C7C-7CA5-408F-88B7-F9C79B7306C0&displaylang=en

     

    3. Collect Cluster Log and send it to me for analyze.

     

    Note: when you send e-mail to me, please add the subject of the post.

     

    Thanks for your cooperation.

     

    Rock Wang - MSFT

    2008年8月19日 上午 10:56
    版主
  •  

    Dear Rock Wang,

     

    I have sent the MPSReports to you on 21 Aug 2008.

    Have you received ?

    Thanks a lot.

    2008年8月26日 上午 07:22
  •  

    Dear customer:

     

    Sorry, I haven't received your e-mail. Please send it again.

     

    When you send e-mail to me, please attach the subject of the post.

     

    Thanks for your cooperation.

     

    Rock Wang - MSFT

     

    2008年8月28日 下午 12:41
    版主
  • Thanks for your reply.

    I have sent the email with MPSReport to v-rocwan@microsofto.com at 10:05 a.m. today.

    Thanks.

    2008年8月29日 上午 02:09
  •  

    Dear customer:

     

    My e-mail address is v-rocwan@microsoft.com, I found you misspelt it. Please send it again.

     

    Thanks for your cooperation.

     

    Rock Wang - MSFT

    2008年8月29日 上午 06:18
    版主
  •  

    Dear customer:

     

    Please help collect the following information:

     

    1.      On the active node, open Event Viewer, navigate to application, right click it and save it as .evt file, and send it to me.

    2.      On the passive node, repeats above steps, and send the .evt file to me.

     

    Note: when you send mail to me, please add the subject of the post.


    Thanks for your cooperation.

     

    Rock Wang - MSFT

    2008年9月1日 上午 07:32
    版主
  •  

    Dear customer:

     

    Based on my research, you can try to use the following resolution in the following knowledge base article, and then check the effect.

     

    The Microsoft Exchange System Attendant service may not come online if the RootVer registry value is invalid on a server that is running Exchange Server 2003

    http://support.microsoft.com/default.aspx?scid=kb;EN-US;906154

     

    Hope it helps. If anything is unclear, please feel free to let me know.

     

    Rock Wang - MSFT

    2008年9月1日 上午 07:41
    版主
  • Thanks for your reply.

    I have checked the value in Registry was same as that in kb.

    Should I follow the kb to run the domainprep ?

     

    2008年9月2日 上午 01:03
  • Dear customer:

     

    You can try the following command, and check the effect.

     

    drive:\setup\i386\setup /domainprep
     
    Rock Wang - MSFT
    2008年9月3日 上午 06:30
    版主
  • Is any impact on the Exchange Organization and the Exchange Site ?

    My Exchange Services account has full rights in the site only.

    BTW. Have you received the MPSReports ?

     

    Thanks.

    2008年9月3日 上午 07:17
  •  

    Dear customer:

     

    It is best practice that you backup Exchange before you run setup /domainprep command.

     

    In addition, the account you use to run DomainPrep must be a member of the Domain Administrators group in the local domain and a local computer administrator.

     

    Note:

     

    Running DomainPrep does not require any Exchange permissions. Only Domain Administrator permissions are required in the local domain.

     

    I have received your MPSreport. I am currently doing research and tests on this issue, and will get back to you as soon as possible. Thanks for your patient and understanding.

     

    Rock Wang - MSFT

    2008年9月3日 上午 07:32
    版主
  •  

    Dear customer:

     

    From your application log file, I found the following error information:

     

    vent Source:      MSExchangeCluster

    Event Category:                Services

    Event ID:              1021

    Date:                     7/19/2008

    Time:                    11:05:30 PM

    User:                     N/A

    Computer:          SERVERF-LAN

    Description:

    Exchange System Attendant (SERVERM-LAN): Failed to terminate this resource because of timeout.

     

    Event Type:        Error

    Event Source:    MSExchangeCluster

    Event Category:                Services

    Event ID:              1011

    Date:                     7/19/2008

    Time:                    11:05:30 PM

    User:                     N/A

    Computer:          SERVERF-LAN

    Description:

    Exchange System Attendant (SERVERM-LAN): The RPC call to the service to bring the resource online failed.

     

    Event Type:        Error

    Event Source:    MSExchangeCluster

    Event Category:                Services

    Event ID:              1024

    Date:                     7/19/2008

    Time:                    11:05:30 PM

    User:                     N/A

    Computer:          SERVERF-LAN

    Description:

    Exchange System Attendant (SERVERM-LAN): Failed to stop service 'MSExchangeSA'.

     

    The date of the error appears is 7/19/2008, however from the cluster log file,

     

    000011cc.00001d2c::2008/08/20-11:15:25.813 INFO Physical Disk <Disk Q: E:>: Query maintenance mode, InMaintenance 0 MaintainenceModeType 0 InternalState 2 status 0.

     

    The date of the last log appears is 2008/08/20, in order better troubleshoot the issue, and please help collect the following information:

     

    1. Does the Exchange System Attendant in the cluster resource start now?
    2. On active node, open cluster manager, navigate to the Exchange virtual server, send the screenshot of it to me,
    3. Right click system attendant and select properties, click general tab, send the screenshot of it to me.

     

    Note: when you send e-mail to me, please add the subject of the post.

     

    Thanks for your cooperation.

     

    Rock Wang - MSFT

    2008年9月5日 上午 03:12
    版主
  •  

    Dear customer:

     

    In order to better troubleshoot the issue, I want to confirm the following information with you:

     

    1.       Which node is the original active node, is it SERVERG-LAN or SERVERF-LAN?

    2.       Which node does the following “NODE1” refer to?

     

    Source: MSExchangeCluster

    EventID: 1019

    Exchange System Attendant (NODE1): Failed to bring this resource online because of timeout.

    3.       From the following description, what does “three Exchange virtual servers” means? Does the Exchange cluster have four nodes or not? Does the Windows Server 2000(domain controller) with Exchange 2000 Server Ent is a cluster? What about Windows Server 2003(domain controller) with Exchange 2003 Server Ent?

     

    In my Exchange Site, there are three Exchange virtual servers as below.

     

    Windows Server 2000(domain controller) with Exchange 2000 Server Ent.

     

    Windows Server 2003(domain controller) with Exchange 2003 Server Ent

     

    Two Windows Cluster Server 2003(joined domain) with two Exchange 2003 Cluster Server.

     

    Thanks for your cooperation. If anything is unclear, please feel free to let me know.

     

    Rock Wang - MSFT

    2008年9月8日 下午 01:20
    版主
  • Thanks for your reply. Here is the information you want.

    1. SERVERF-LAN is active node.

    2. NODE1 is SERVERF-LAN

    3. In the site, there are four Exchnage Servers. Two are Exchnage 2003 Ent Servers(one domain controller and one domain member server). One is Exchange 2000 Ent Server(domain controller). Another one is Exchnage 2003 Cluster server(SERVERF-LAN & SERVERG-LAN, both are domain member server).

     

    Thanks.

     

     

    2008年9月9日 上午 03:55
  •  

    Dear customer:

     

    Form your picture1 and picture2 screenshot, I found the SERVERF-LAN node is running. From your previous cluster and event log, it seems that issue didn’t occur since July 21, 2008. Since the log file is too old, from the current information we are almost unable to determine the root cause.

     

    If the issue reoccurs in the further, please let know. I will follow up.

     

    From your post, I found that you posted two questions, however these two questions are unrelated. Please understand that our forum is an issue based service, meaning we usually respond to one question/issue per post. This will lessen the confusion for both of us, as well as ensure that our results are accurate and not a result of a test for a different question. I advice you to post the following question into a new post.

     

    “Also have the error message prompt up in the System Manager when purged the mailbox,

     

    "unable to contact the global catalog server, facility win32 ID no:c0072019”.

     

    Thanks for your cooperation and understanding.

     

    Rock Wang - MSFT

    2008年9月10日 上午 09:22
    版主
  • Dear Mr. Wang,

     

    Thank you for your help. Your job I always appreciate.

    I decided to re-install the Exchange Cluster Servers. I think those problems should be solved.

     

    Regards.

     

    2008年9月10日 下午 02:08
  • Hi SM Chung

     

    On 8 Oct, after first day of TechEd, we are going to host a community gathering where you can have a drink and network with your peers in the industry, please RSVP with me at hkdevcom@microsoft.com.

     

    Event Details

    Date: 8 Oct, 2008 (Wednesday)

    Time: 7pm to 9pm

    Venue: Cenna Bar & Lounge (銅鑼灣軒尼詩道525號恒和鑽石大廈23)

     

    Thanks and looking forward to seeing you at the event!

     

    Cheers,

    Angela Ip (MSHK)

    2008年9月25日 上午 08:29