|
| |
"Resource" is the shared components inside a node, such as storage, IP address. An application may interact with several resources, and before a node can run the application, it should first get all the resources own by the application.
When the application owned the related resources, a resource group is created. Resource group provides certain kind of service to user. Resource group is treat the resources as one entity, exclusive resources will only be owned by one cluster member at one time, and share resources will allow to own by multiple members in the failover domain.
First time clicks on "Resource Group" which is located at the Component Area does not contains any resource group information. User needs to create the resource group by either right click on Resource group and select "Create New Resource Group" or click the button located at the shortcut bar.
A dialog box will shows and user needs to fill in the information about the
Table 6-12. Create Resource Group
| Name of Functions | Description |
|---|---|
| Resource Group ID | Give the resource group an unique name |
| Description | Provide description for this resource group. |
| Auto Start | When cluster manager starts, the resource group will be enabled |
| Manual Start | When cluster manager starts, the resource group will Not be enabled |
| Auto Failover | When resource failed, the resource will be switched to another node automatically. (Default is enable) |
| Auto Failback | When previous failed resource back online, the resource will be switched back to the node automatically. (Default is enable) |
| Load Balancing | When resource failed, the resource will be switched to the node with lesser loading, instead of high loading node. This feature is used with more then 2 nodes cluster environment. (Default is disable) |
| Maximum Owner | If a resource needs to run more then one node at the same time, user needs to change this value. For example: 2 out of 3 nodes have to run the same resources. (Default Value is 1, not accept value less then 1 ) |
| Minimum Owner | For a resource needs at least a number of node running. (Default Value is 1, and not accept value larger then Maximum Owner ) |
| Maximum Retry | The number of retry before resource switches to another node. (Default Value is 1 ) |
| Tie Breaker | Select a node in order to choose and form the cluster from two equal-sized node groups. This option is only available after resource group created. (Default is none) |
After user created the resource group, the resource group will be added to the component area, and icon for this resource group will be created in working area. User can move the mouse pointer over the icon and read the description, or switch to detail view by clicking shortcut or right click to select.
A new resource group has been created and user can double click on the resource group icon from the "Resource Group List" or, expand "Resource Groups" and select the group.
The "Resource Group" provide the information about this resource group setting with status. A new created resource group does not contain any member, a warning message "Failover Domain is empty! Please add nodes to Failover domain." in red will be appeared to remind user to add members to this group. User can change the resource group setting by clicking shortcut icon or right click the resource group and select "Edit Resource Group", excepts the resource group id.
Click the "Add" button and a dialog box will shows as follow;
Member is the available members within the cluster environment, and priority is the order and importance of member.
After the member add to the resource group, the warning message will disappear. If a node is added incorrectly, user can remove it by clicking Delete button. Or changing the node priority order, click the "Change Node Priority" or right click resource group and select it.
If user needs to add additional group, simply run through the "Create Resource Group" again. Remember, the new created group id must be with an unique group id.
After a resource group created, user can add the resource to this group.
Select the group that would like to add resources, then click on the shortcut bar or select "Add Resource" by right click. Following window will show up;
Table 6-13. Resource Window
| Name of Functions | Description |
|---|---|
| Resource ID | Give the resource an unique name |
| Resource Type | Depends on resource, user needs to select a suitable resource type to represent the resource. |
| Resource Config Type (Global / Local) | Global configuration is a setting that apply to all the member node. Local configuration is specific for a member. For example: an application with IP address is running on multiple members, a global setting will cause IP conflict. Therefore, IP resource setting must be using local. (Default is Global) |
| Priority | Some resources will depend on the other resource, such as web server and external data. The priority will be external data is higher then the web server. |
Adding resource is simple, user only need to pay attention of the resource type. Different resource types provide different parameters for user to input. Following table provide the information about each resource type.
Resource is a shared component inside the application failover domain. Different resource handles different types of components such as IP resource handles IP address, application resource handles different types of applications. SIC supports different resources internally, and functions as follow;
Applications resource is used to control the application status. User can define the application with specific application command with parameter. SIC will use this user define command to start, stop, and check status within a time that setup by user. When the application resource failed to response, SIC will restart the application before failover to another member.
Filesystem resource is used to define the storage path for specific usage. When a member failed, the Filesystem will switch the mount point from the failed member to another member.
IP Address resource is used to define IP address. If a member failed, the IP address resource will be switched to another member for IP address connection.
Logical Volume Manager resource is used to connect and activate the volume group which is setup on a shared disk storage. When member is failed, it will deactivate the volume group and switch to another member and activate the volume group again.
Multiple Disk resource is used to connect and activate the shared storage disk array with RAID setup. When member is failed, it will deactivate the RAID disk and switch to another member and activate the disk connection and RAID support.
Network Interface resource is used to monitor specific network interface controller. If the NIC failed, resource group will be switch to another member in order to continue operation.
System health monitoring is used to monitoring the CPU usage status. If the CPU usage is keep on a high demand in a period of time, it will be switched the resource group to another member.
Table 6-14. Resources
| Resource Type | Description | Parameter (Example) |
|---|---|---|
| Application | Control the applications |
Start Command: (Example: /etc/init.d/httpd start) Stop Command: (Example: /etc/init.d/httpd stop) Restart Command: (Example: /etc/init.d/httpd restart) Status Command: (Example: /etc/init.d/httpd status) Timeout (second): (Example: 10) |
| Filesystem | Mount filesystem / device to the specific location |
Mount Point:value (Example: /mnt/share) Mount Option:value (Example: rw,noauto) Device:value (Example: /dev/hda1 )Filesystem type:value (Example: ext3) |
| IP Address | Configure an IP address |
IP Address: (Example: 192.168.3.131) Net Mask: (Example: 255.255.255.0) Broadcast: (Example: 192.168.3.255) Network Interface: (Example: eth0:1) |
| Logical Volume Manager | Activate a logical volume |
volume_group [Volume Group]: value (Example: volume_group_name) vgchange_path [vgchange path]: value (Example: /sbin/vgchange ) vgck_path [vgck path]: value (Example: /sbin/vgck ) vgexport_path [vgexport path]: value (Example: /sbin/vgexport ) vgimport_path [vgimport path]: value (Example: /sbin/vgimport ) devices [Devices]: value (Example: /dev/sda1 /dev/sdb1 ) timeout: value |
| Multiple Disk | Start/stop a Raid device |
raiddev [Raid Device]: value (Example: /dev/md0 ) raid_level [Raid Level]: value (Example: 1 ) mdadm_path [mdadm Command Path]: value (Example: /sbin/mdadm ) devices[Devices]: value (Example: /dev/sda1 /dev/sdb1) |
| Network Interface | Monitoring the network interface controller status |
interface [Network Interface]: value (Example: eth0) |
| System Health Monitoring | Monitor the member CPU running status |
CPU Usage(0-100): value (Example: 85) Period(in seconds): value (Example: 150) |
After adding the necessary resources to the resource group, then user can enable the resource group. .
The resource group will be enabled automatically. The member list on the resource group page shows the status of top priority member is "Running". If user configure the resource group has more then one member, then the second high priority member will be changed to running status.
