Running corrective actions during verification
You can run automatic corrective actions during cluster verification on an inactive cluster. By default, automatic corrective action is enabled for the standard configuration paths and disabled for custom configuration path.
Automatic corrective actions can be disabled for the advanced verification dialog (from the menu) but it cannot be disabled for the standard paths. You can run verification with corrective actions in one of two modes:
- Interactively. If you select Interactively, when verification detects a correctable condition related to importing a volume group or to re-importing mount points and file systems, you are prompted to authorize a corrective action before verification continues.
- Automatically. If you select Yes, when verification detects that any of the error conditions exists, it takes the corrective action automatically without a prompt.
If an error discovered during verification has a corrective action, the item is corrected and the run continues. For situations when the correction involves importing a shared volume group, re-importing a shared volume group, or updating the /etc/hosts file, the utility runs all verification checks again after it corrects one of the previous conditions. If the same error condition is triggered again, the associated corrective action is not executed. The error is logged and verification fails. If the original condition is a warning, verification succeeds.
PowerHA® SystemMirror® detects active service IP labels and active volume groups on nodes regardless of whether or not nodes are running cluster services. PowerHA SystemMirror looks at the resource group state instead of cluster services. When verification detects active resources on a node that does not have the resource group in ONLINE state or does not have the resource group in an UNMANAGED state, verification gives you the option to bring these resources OFFLINE according the following table.
Verification cannot tell which node will actually acquire the resource group that has active resources. Thus, the warning messages mentioned in the following table are printed every time active resources are found on a node that is or is not stopped and the state of the resource group to which active resources belong is UNMANAGED, OFFLINE, or ERROR.
| Resource Group Attribute: Manage Resource Group Automatically | Resource Group Attribute: Manage Resource Group Manually | |
|---|---|---|
| Interactively correct errors | Display message with option to bring resources offline. | Display message with option to bring resources offline. |
| Automatically correct errors | Reset the startup attribute Managed Resource group to: Manually. Display warning message. | Print reminder/warning and steps to take. |
| No corrective actions | Print reminder/warning and steps to take | Print reminder/warning and steps to take |
| Cluster Services are running | N/A | N/A |