Versions Compared

Version Old Version 31 New Version 32
Changes made by

JuneL (Unlicensed)

헌규임

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

With synchronous mirroring, each write is captured and transmitted to the target system to be written on the target volume at the same time that the write is committed to the underlying storage device on the source system.  Once both the local and target writes are complete, the write request is acknowledged as complete and control is returned to the application that initiated the write.  With synchronous mirroring, each write is intercepted and transmitted to the target system to be written on the target volume at the same time that the write is committed to the underlying storage device on the source system.  Once both the local and target writes are complete, the write request is acknowledged as complete and control is returned to the application that initiated the write.  



Adding

Add the mirror disk resource to a group.

  1. When add a resource from a group name, select a group → right click → 'Add Resource'. Or select Edit(E) from the main menu bar → select 'Add Resource'. Or select 'Add Resource' icon from the tool bar.
     
  2. Select 'MirrorDisk' from Resource Type lists and click 'Next' button. When add a resource from the mirror disk resource type name, this step will be skipped.

  3.  Select  Select the mirror volume and MCCS will renew the information.

    If you skip enter resource name, this will given 'Mirror_value' which value is selected driver letter.(Ex: Mirror_G)

     

    Warning

    The drive characters to be used as the mirror disk should be same on the both servers.
    [Drive characters in the SAN environment (shared disk) are also displayed in the list. So, make sure you select the accurate drive characters.]
    Data will be copied to the target node from the source node.
    Be careful when you select the drive character where data exists.
    If you try to add the existing mirror volume, <mirrored> will be added to the volume letter.


    [Figure] Mirror Disk Resource Added

     

  4. As 'Additional Settings', Synchronized Mode and Check Disk option is available.

    Info

    <Synchronized Mode>

    Once selected, the synch mode cannot be reversed. So, be careful.

    Sync mode : Response is slow but there is no data loss involved.
    Async mode : Relatively faster than the synch mode but it can suffer from data loss.


    <Check Disk Option>

    Set the check disk command option. For more details on the function, refer to MSDN.

    [Figure] Defining mirror type and check disk option

     

  5. When click OK button, following popup message appear.
    Confirm selected server as source.
    If you select wrong sever, clock cancel button.
     
    [Figure] Alert popup message after click OK button

  6. Click the 'Finish' button to add the mirror disk resource.  
    You can immediately check the result in the management web console.

 

Deleting

Select resource type → right click → delete resource.

  1. Click "Delete resource" and a confirming message about deleting resource will appear. 

    [Figure] check resource view

  2. Click "OK" and a confirming message about deleting redundancy program of mirror configuration will. 

    [Figure] Delete physical mirror disk view

  3. Click "OK" and redundancy program is deleted. Click "Cancle" and only mirror disk is deleted. 
    The deleted resource will immediately disappear from the management web console.


Status

The following table explains the status switching of the MCCS resource caused by a user's command and the status.
The command assumes that it is generated by a user.

...

StatusAgent commandDescriptionNote

Online

In this status, you can access the source volume and perform a writing test properly.

Offline

Regardless of the mirror role, the LOCKVOLUME command is used to lock the volume.
If the mirror volume is not defined, the LOCKVOLUME command is ignored and regarded as a failure.

 
Monitoring

The monitoring process is also determined by the mirror volume status and roles.
The kernel driver of copying program records an event in the system event log when the mirror volume status is changed.
The MCCS event monitor is registered as an event receiver when starting the MCCS service. When a system event occurs, it will be automatically informed and it will detect the mirror volume status and role changes.
Once the status and the role are determined, a disk writing test is performed to finalize the resource status.

<Source volume>
If the writing test is successful, it is regarded as the online state.

<Target volume>
1. If the writing test failed on the disk, it is regarded as the offline state.
2. If a user unlocks the target volume while the mirror disk is online, an agent will automatically lock it again.
 Thus, the target volume of the node with the online mirror disk cannot be unlocked.
3. If the mirror volume is not defined,  it is regarded as the offline state.

 

Offline

Except for the online and trouble states, it is always offline.

Online

Type of operation is determined by the role of the mirror volume at the start node.

<Source volume>
1. Allow them to unlock and write to the volume.
2. Unlocking is done by 'UNLOCKVOLUME' of the copying program. The write function is executed after changing the MountReadOnly in registry value of the copying program to 0. After writing is available, the value is changed back to 1.

<Target volume>
1. An attempt to go online at the node where the mirror role is the target failed if the mirror status is not "Mirroring".
2. If the mirror volume is fully synchronized, you can change the mirror role. Such a process is determined by the node status (A).
3. If you cannot connect to mirror driver of node A, use the DELETEMIRROR command to delete the existing mirror volume. After that, switch the target and the source, and use the CREATEMIRROR command to create the reverse mirror volume.
When creating the mirror, if you set the option to 0x1, they are not all synchronized.
4. If you cannot connect the mirror driver of node A, use the DELETELOCALMIRRORONLY command to delete the mirror configuration of node B only. After that, use the CREATEMIRROR command to configure the reverse mirror volume.
Now, node A does not respond. The 0x2 option is added so that it does not wait until the target mirror volume is created.
If the final volume is locked, unlock it and change the MountReadOnly option to 0. After writing, change it back to 1.
If the mirror volume is not defined, it is processed as a failure without any operation.

 
Monitoring

Refer to the description of monitoring as above.

 

Trouble

If a writing test failed online, or an attempt to go online is failed, the trouble state is displayed.

*Failover deactivation state
When a mirror network communication failure, target node or target disk failure occurs, the mirror state is changed from 'MIRRORING' to other, such as 'MIRROR_PAUSED'.
If the mirror status is not 'MIRRORING', a failover will take place and it can cause data losses or damages.
To prevent this problem, an agent deactivates the failover mode if the mirror status is not "MIRRORING".
If the failover mode is deactivated, no failover will take place even if manual failover or troubles occur.
If mirror network communication failures are recovered, and communication resumes, it will automatically perform RESYNC. When the mirror status is changed to MIRRORING, the failover mode will be also automatically activated.

 

Online

Refer to the above online command.

 
Offline

Regardless of the mirror role, the LOCKVOLUME command is used to lock the volume.
If the mirror volume is not defined, the LOCKVOLUME command is ignored and regarded as a failure.

 

 

 

Attribute

Show the unique attribute of resource.

Attribute

Desctiption

TypeDefaultEditNote
AutoResolveSplitBrain

If it is set to 'true', it will automatically resolve the split brain issues.

BOOLEANtrueO 
ChkDsk

When the volume is mounted, according to the 'ChkdskArg', disk is tested for errors.
If it is set to true, and if dirty bits are set to the online volume, the ChkdskArg option will be used to run checkdisk.
Before running checkdisk, the mirror copying will be automatically stopped and resynchronization will be performed after finished..

BOOLEANfalseO 
ChkdskArg

When the Chkdsk option is set to true, the check disk command option is activated.

STRING/FO 
DirtyBit

It means that dirty bits are set to the target mirror volume.
In general, if you need to check the integrity of the target volume, the system will set this.

BOOLEANfalseX 
DomainMemberServer

If the node to configure the mirror resource is a member server of the Windows domain, you must set it to "true".
Now, the Windows service account to start the DK service should be set as Administrator for the local node.

BOOLEANfalseO 

DriveLetter

The mirror volume is assigned to the drive character.
It should be configured to obtain the same value at the source and the target systems.

STRING X 
MirrorIP

The IP address used to copy the volume.

STRING X 
MirrorRoleThe mirror role of the mirror volume.
It is renewed from an event log.		STRING X 
MirrorState

The mirror status of the mirror volume.
It is renewed from an event log.

STRING X 
MirrorTimeout

This is the time limit for mirror role or status updates in the process of volume failovers or switching.
If the status or role is not renewed within this time period, a time out error will occur.

INTEGER120 secO 
MirrorTypeDetermine the mirror synchronization mode. Set it to A or S.
( A - async mode, S - sync mode)
* If it is set to asynch mode, data loss can occur.		STRINGAX 
Password

It is the password for the designated user account.
Unless you set it otherwise before, now, enter the password for the 'Administrator' account.

PASSWORD O 
ResyncNeedIt represents the total sum of the blocks requiring resynchronization.INTEGER0X 
ResyncTotalIt is the total sum of the resynchronization blocks.INTEGER0X 
TimeAquiredSourceRoleIt is the time when the node's mirror roles are changed to the source.STRING X 
TotalSpaceIt represents the total size of the target mirror volume.INTEGER X 
UsedSpaceIt represents the available size of the target mirror volume.INTEGER X 
User

The user account to run the mirror related commands.
Unless you set it otherwise before, now, the 'Administrator' account will be used.

STRINGAdministratorO 

...