2014/03/27

Live Migration failed with error 0x00002745

I just got a strange error message during a failed live migration.

Type: Error
Event-ID: 21502
Source: Hyper-V-High-Availability

Message:   

”Virtual Machine {server}' live migration did not succeed at the source
 ....
Could not send data to the remote node: An Established connection was aborted by the software in your host machine (0x00002745)


or in german stating

"Es konnten keine Daten vom Remoteknoten empfangen werden: Eine bestehende Verbindung wurde softwaregesteuert durch den Hostcomputer abgebrochen. (0x00002745)"

Long story short:  Live Migration failed, we tried a quick migration which worked "well" but couldn't start the VM. After opening the console of the vm and trieing to start the saved quick migrated state we saw the real error "not enough memory available"

2012/10/29

SnapManager Exchange or SQL Service failed to start

or why .NET-Applications need long to start: "A common cause that I have seen often recently is Certificate Revocation List (CRL) checking when generating Publisher Evidence for Code Access Security (CAS)."[1]

In my case the Exchange Sever didn't had anymore direct access to the internet. The CRL checking during the SM Serverice Startup failed. This took longer than 30 seconds and causes the service start to time out. Adding some lines to the SnapMgrService.exe.config prevent the service to do the online CRL check. No further timeouts, services starts.

<?xml version="1.0" encoding="utf-8"?>
<configuration>
    <runtime>
        <generatePublisherEvidence enabled="false"/>
    </runtime>
    ...
 
Same issue some month's later with a SnapManager SQL installation. Good to have a cheat sheet for the most wired thing online.
 
 
[1] Best Practice - <GeneratePublisherEvidence> in ASPNET.CONFIG

2012/03/19

Granting Permissions to Domain Admin group on SQL Server 2008 Instance

I just wondered about why i couldn't login as member of the domain administrators group to the SQL-Server after i added this group to the SQL as sysadmin. If i add me as user itself the login workes fine.  Here is the right "procedure":


- Add Domain Admins group in the SQL Server Security Login
- Give Sysadmin rights to the Domain Admins group

AND (if you login as member of the group)


- Disable UAC or start the SQL Server Management Studio as an Administrator

2011/09/22

Backup Exec 2010 - Some Hints

LU1805 - LiveUpdate was unable to find any products to Update

  • Open the command prompt and browse to the following directory {DEFAULT.EN_US} C:\Program Files\Symantec\Backup Exec\
  • Run BeUpdateOps.exe -AddBE -OptOut 
  • Run Live Update
Found on the symantec enterprise website[1] (TECH66265)


NDMP Backup to disk to tape

Can cause troubles... In case of troubles use 2 jobs: one ndmp to tape, one ndmp to disk



... continued


[1] http://www.symantec.com/business/support/index?page=content&id=TECH66265

2011/01/19

How to backup SQL Server 2008 Express

I’m facing a lot of situations where customers prefer to use a bunch of free SQL Server 2008 Express or SQL Server 2005 Express Editions - mostly installed by every Client-Server-App they use - instead of using a paid but more powerful SQL Server.  The last time I had to use an Express Edition and implemented the backup for the databases I took the time to review my SQL backup script.
 
How to backup SQL Server 2008 (Express Edition) using the command line - the whole script is downloadable below:
  • Adjust the parameters within the sql file:
    • @BACKUP_PATH = <Backup target path> - the path where the sql server creates the media set for the backup
    • @MEDIASET_PREFIX = <Backup filename prefix> - Used to give backups meaningful names. The files get automatically a timestamp attached.
    • @RETENTION = <number of backups to keep> - Specifies the of backup sets to keep.

  • Call sqlcmd from a command line or trigger a batch job calling sqlcmd and pass through the sql script doing the backup for you:
sqlcmd -i <sqlfile> -o <logfile>

-i <sqlfile>: this sql-file to be executed
-o <logfile>: a log file for stdout redirection
Whats done within the sql script (You don't need to care about this - everything is put into the script):
  • Create a media set
EXEC sp_addumpdevice @devtype, @logicalname,@physname

@devtype: In our case set to 'disk' as we do a backup to disk not to tape.
@logicalname: the name of the mediaset created using the defined @MEDIASET_PREFIX and the current timestamp
@physname: the path and filename of the mediaset, simply the backup file
  • Iterarte through all online databases - offline databases cannot be backed up and do a backup to the created media set
DECLARE databaseNamesCursor CURSOR FOR
    SELECT 
        name
    FROM
        master.dbo.sysdatabases
    WHERE
        name != 'tempdb'
        AND DATABASEPROPERTYEX(name, 'Status') = 'ONLINE'
OPEN databaseNamesCursor;

FETCH NEXT FROM databaseNamesCursor INTO @databaseName;

WHILE @@FETCH_STATUS = 0
BEGIN
    /* create a backup of the current database*/
    BACKUP DATABASE @databaseName TO @mediaset;
    FETCH NEXT FROM databaseNamesCursor INTO @databaseName;
END

CLOSE databaseNamesCursor;
DEALLOCATE databaseNamesCursor;
  • Delete not needed backups 
sp_dropdevice @logicalname, @delfile='DELFILE'

@logicalname: the name of the mediaset to delete
@delfile: if set to 'delfile' the physical file behind the mediaset will be deleted


Download the files: 
- link broken, suggest  you send me an email if you need the scripts -

2011/01/10

The owner () of job does not have server access.

New week new troubles: SnapManager SQL created some backup jobs within the SQL Server 2005 - the job itself could not be run/executed.

The owner (...) of job <Job Name> does not have server access.

This strange behaviour i got only with SQL Server 2005. With SQL Server 2008 i could not reproduce it.  The user who created the job and was the owner of the job had SQL sysadmin role assigned. I found some hints to this behaviour: a restart of the SQL Server Agent should work. In my case i didn't work that easy. 

If i change the owner of the job to the builtin user sa the job is able to run...

It would be nice to understand the issue....

2011/01/06

Can’t create a new Receive connector after installing Update Rollup 1 for Exchange Server 2007 Service Pack 3

Recently i had to migrate Exchange Server 2007 to another host. I installed the latest release with SP3 included and ran the install of the Update Rollup 1 for Exchange Server 2007 Service Pack 3. Pretty easy straight forward until i tried to create a new Receive Connector
  

Error:
Active Directory operation failed on xxxx.local. This error is not retriable. Additional information: The parameter is incorrect.

Active directory response: 00000057: LdapErr: DSID-0C090C26, comment: Error in attribute conversion operation, data 0, v1772
The requested attribute does not exist.


I longer research took me to the hint - i didn't expect that comming with a Service Pack - that you have to extend the AD schema prior installing Exchange Server 2007 SP3[1]:

If you want to run the Setup program in upgrade mode to perform an in-place upgrade from a beta version of Exchange 2007 SP3, you must first prepare Active Directory and the domains. 


Download the Service Pack 3  again or like in my case use the install media including the SP3 and run Setup.com /PrepareSchema agains the schema master. This should fix the LDAP errors.