MetroCluster Manuals ( CA08871-401 )

Monitoring the MetroCluster configuration

MetroCluster Tiebreaker software automates the recovery process by enabling you to monitor the MetroCluster configuration status, evaluate SNMP events and traps that are sent to Fujitsu support personnel, and view the status of monitoring operations.

SNMP events and traps

MetroCluster Tiebreaker software uses SNMP traps to notify you of significant events. These traps are part of the Fujitsu MIB file. Each trap contains the following information: trap name, severity, impact level, timestamp, and message.

Event name	Event detail	Trap number
MetroCluster Tie-Breaker is unable to reach the MetroCluster configuration	Warns the administrator that the software cannot detect a disaster. This event occurs when both clusters are not reachable.	25000
MetroCluster Tie-Breaker is unable to reach cluster	Warns the administrator that the software cannot reach one of the clusters.	25001
MetroCluster Tie-Breaker detected disaster at cluster	Notifies the administrator that the software detects a site failure. A notification will be delivered.	25002
All links between partner cluster are severed.	The software detects that both clusters are reachable, but all the network paths between the two clusters are down, and the clusters cannot communicate with each other.	25005
SNMP Test Trap	SNMP configuration can now be tested by running the snmp config test command.	25006

Event name

Event detail

Trap number

MetroCluster Tie-Breaker is unable to reach the MetroCluster configuration

Warns the administrator that the software cannot detect a disaster. This event occurs when both clusters are not reachable.

25000

MetroCluster Tie-Breaker is unable to reach cluster

Warns the administrator that the software cannot reach one of the clusters.

25001

MetroCluster Tie-Breaker detected disaster at cluster

Notifies the administrator that the software detects a site failure. A notification will be delivered.

25002

All links between partner cluster are severed.

The software detects that both clusters are reachable, but all the network paths between the two clusters are down, and the clusters cannot communicate with each other.

25005

SNMP Test Trap

SNMP configuration can now be tested by running the snmp config test command.

25006

Displaying the status of monitoring operations

You can display the overall status of monitoring operations for a MetroCluster configuration.

Step

Use the Tiebreaker CLI monitor show command to display the status of a MetroCluster operation with any of the following options:

Option	Description
-monitor-name	Displays the status for the specified monitor name
-operation-history	Displays up to 10 monitoring operations that were last performed on a cluster
-stats	Displays the statistics related to the specified cluster
-status	Displays the status of the specified cluster Note: The MetroCluster Tiebreaker software might take up to 10 minutes to reflect the completion status of operations such as heal aggregates, heal roots, or switchback.

Option

Description

-monitor-name

Displays the status for the specified monitor name

-operation-history

Displays up to 10 monitoring operations that were last performed on a cluster

-stats

Displays the statistics related to the specified cluster

-status

Displays the status of the specified cluster Note: The MetroCluster Tiebreaker software might take up to 10 minutes to reflect the completion status of operations such as heal aggregates, heal roots, or switchback.

The following example shows that the clusters cluster_A and cluster_B are connected and healthy:

MetroCluster Tiebreaker:> monitor show -status
MetroCluster: cluster_A
    Disaster: false
    Monitor State: Normal
    Observer Mode: true
    Silent Period: 15
    Override Vetoes: false
    Cluster: cluster_Ba(UUID:4d9ccf24-080f-11e4-9df2-00a098168e7c)
        Reachable: true
        All-Links-Severed: FALSE
            Node: mcc5-a1(UUID:78b44707-0809-11e4-9be1-e50dab9e83e1)
                Reachable: true
                All-Links-Severed: FALSE
                State: normal
            Node: mcc5-a2(UUID:9a8b1059-0809-11e4-9f5e-8d97cdec7102)
                Reachable: true
                All-Links-Severed: FALSE
                State: normal
    Cluster: cluster_B(UUID:70dacd3b-0823-11e4-a7b9-00a0981693c4)
        Reachable: true
        All-Links-Severed: FALSE
            Node: mcc5-b1(UUID:961fce7d-081d-11e4-9ebf-2f295df8fcb3)
                Reachable: true
                All-Links-Severed: FALSE
                State: normal
            Node: mcc5-b2(UUID:9393262d-081d-11e4-80d5-6b30884058dc)
                Reachable: true
                All-Links-Severed: FALSE
                State: normal

In the following example, the last seven operations that were run on cluster_B are displayed:

MetroCluster Tiebreaker:> monitor show -operation-history
MetroCluster: cluster_B
 [ 2014-09-15 04:48:32.274 ] MetroCluster Monitor is initialized
 [ 2014-09-15 04:48:32.278 ] Started Discovery and validation of MetroCluster Setup
 [ 2014-09-15 04:48:35.078 ] Discovery and validation of MetroCluster Setup succeeded. Started monitoring.
 [ 2014-09-15 04:48:35.246 ] MetroCluster Tiebreaker software is able to reach cluster "mcc5a"
 [ 2014-09-15 04:48:35.256 ] MetroCluster Tiebreaker software is able to reach cluster "mcc5b"
 [ 2014-09-15 04:48:35.298 ] Link to remote DR cluster is up for cluster "mcc5a"
 [ 2014-09-15 04:48:35.308 ] Link to remote DR cluster is up for cluster "mcc5b"

Displaying MetroCluster configuration information

You can display the monitor name and IP address of all instances of MetroCluster configurations in the Tiebreaker software.

Step

Use the Tiebreaker CLI configuration show command to display the MetroCluster configuration information.

The following example shows the information for clusters cluster_A and cluster_B:

MetroCluster: North America
    Monitor Enabled: true
    ClusterA name: cluster_A
    ClusterA IpAddress: 10.222.196.130
    ClusterB name: cluster_B
    ClusterB IpAddress: 10.222.196.140

Creating dump files

You save the overall status the Tiebreaker software to a dump file for debugging purposes.

Step

Use the Tiebreaker CLI monitor dump -status command to create a dump file of the overall status of all MetroCluster configurations.

The following example shows the successful creation of the /var/log/fujitsu/mcctb/metrocluster-tiebreaker-status.xml dump file:
```
MetroCluster Tiebreaker :> monitor dump -status
MetroCluster Tiebreaker status successfully dumped in file /var/log/fujitsu/mcctb/metrocluster-tiebreaker-status.xml
```