Appendix H- Mirror Management Console
- 환기장
- 정은진
- JuneL (Unlicensed)
The mirror management console is provided only for the individual mirror resources managed by MCCS.
The mirror management console operates independently from the engine logic of MCCS, so during operation, it can collide with MCCS.
The mirror management console informs users of the mirror volume status and the capacity of the mirror in use. You can change the bandwidth and the mirror network addresses in it.
Table of Contents
Screen configuration
Mirror management screen
The following mirror management screen is used.
[Figure] Mirror management screen
Mirror set configuration
The mirror resource configured in MCCS is expressed as a single mirror set.
| Figure | Description |
Volume |
| The volume condition is Source. |
 | The volume condition is Target. | |
Status
|
| The volume condition is Mirroring. |
| The volume condition is Resyncing. Displays the progress bar for resyncing. | |
 | The volume condition is Pause. | |
 | There is a dirtybit in the volume. | |
 | Split brain occurred. |
Mirror management command descriptions
Right click with your mouse button on the mirror set volume to see the descriptions of the mirror management commands that can be selected.
Menu | Decription |
Connect | Set up communication between two hosts that share the copied data sets. |
| Disconnect | Disconnect communication between two hosts that share the copied data sets. |
| Forceful sync | If you use the forceful sync command, the status will be changed to RESYNCING and the full synchronization will be done forcefully. |
Descriptions of properties setting
[Figure] Mirror Management Screen
Attribute name | Description |
| Virtual device path | Shows the mounting point. |
Meta device path | Shows whether a meta device is internal or external. |
Resynchronization ratio | Set the resynchronization ratio. (Setting range: 0 - 640MB) |
| Mirror network address | Copy the volume to the set IP address. |
Policy for recovering auto split brain
Node Priority | policy | Description |
| Secondary-Secondary | Toss the last primary node | Start synchronization at the node that was primary before a split brain occurred. |
| Toss the first primary node | Start synchronization at the node that was second primary before a split brain occurred. | |
| Toss the one with small changes | Start synchronization at the node with greater changes when a split brain occurred. | |
| Toss the one with no changes | If only one node has changes, start synchronization on it. If the both nodes have no changes, synchronize either one. If the both node have changes, disconnect communication. | |
| Disconnect | Cut off a connection without auto recovery. | |
Primary-Secondary | Consensus | Apply the same recovery policy that is applicable when all the nodes are secondary. If auto recovery fails, disconnect communication. |
| Toss secondary node | Make the secondary as a sacrifice node. | |
| Disconnect | Cut off a connection without auto recovery. | |
Primary-Primary | Disconnect | Cut off a connection without auto recovery. |