Verifying two different clusters by using the CCV utility
The cross-cluster verification (CCV) utility compares specific attributes of two different cluster configurations. The CCV utility compares data that is collected from different clusters, cluster snapshots, active configuration directory, or the default configuration directory of a local cluster. Learn about the requirements and data inputs for using the CCV utility, and how to use the CCV utility to compare two different clusters. Review the scenarios that illustrate how the CCV utility can be used.
Requirements for using the CCV utility
Because the CCV utility compares the configuration of different clusters, you must collect data from multiple clusters. The data must be gathered on a node where PowerHA® SystemMirror® is installed so that the data can be used by the CCV utility.
To gather data from different clusters, use any of the following methods:
- You can generate data on one node from each cluster, and then use File Transfer Protocol (FTP), Secure Copy Protocol (SCP), smart card or remote execution facilities to manually gather that data on a single node from where the CCV utility can use that data.
- If you configure the remote execution facilities such as Secure Shell (SSH) and SCP between different clusters, you can use the clmgr command or the clccv command to automate the generation, collection, and analysis of data from different clusters.
When you generate the cluster configuration data and collect it on a node, you must have adequate disk space to store data. The CCV utility can receive multiple data formats as an input, and the size of data can vary for each cluster configuration. You cannot determine specific disk space requirements in advance. Hence, you must ensure that adequate disk space is available for data input for the CCV utility and the resulting output.
Data inputs to the CCV utility
The CCV utility receives specific configuration information about two different clusters as an input for comparison. You can get the cluster configuration information from the following sources:
- You can use data that is generated by the CCV cl_ccv_collector utility.
- You can use the cluster snapshot utility to capture cluster configuration information about a cluster into a file, and then specify that file as an input to the CCV utility.
- You can run the CCV utility from a node of a PowerHA SystemMirror cluster. The local cluster configuration information is used as one of the input sources for comparison. The Active configuration directory (ACD) is available after you start cluster services.
You can specify any combination of inputs to the CCV utility for comparison. For example, you can:
- Compare two clusters by creating a snapshot from each cluster and then specifying those snapshots as inputs to the CCV utility.
- Compare the local cluster configuration with a remote cluster snapshot.
- Compare the active cluster configuration with data collected by the CCV utility on a remote node.
Comparing configuration attributes by using the CCV utility
The CCV utility compares a subset of configuration attributes from clusters. Each cluster is uniquely identified by the cluster name and ID. For more information about the list of entities that can be compared by using the CCV utility, see the cl_ccv_comparator.
The CCV utility uses the name of the entity as a unique identifier for comparison. Any two entities with the same name are compared and the difference in values are listed. All configuration attributes of all entities are not compared because of expected difference in values. For example, if a resource group with the same name exists on two different clusters, the service labels of those resources groups are expected to be different because the service labels must be unique on a network.
Therefore, when two entities with the same name are compared, only the difference in values are reported. For example, if a resource group with the same name exists on two different clusters, and if one resource group on a cluster contains three service labels and the resource group on another cluster contains only two service labels, this difference in values is reported by the CCV utility.
Output from the CCV utility
The CCV utility generates output files in the JavaScript Object Notation (JSON) format, which can be viewed by using any text editor or any graphic tool. You can also convert the output files to other data formats and view or store the output files by using other tools.
Usage scenarios
- Using the CCV utility for the first time
-
Review the scenarios to become familiar with the technical and configuration details.
When you are using the CCV utility for the first time, familiarize yourself with the features of the CCV utility. You can run the cl_ccv_collector command on the target node in each cluster and collect the resulting data on a cluster node where you can run the cl_ccv_comparator command. The output that is generated from the cl_ccv_collector command is written to the /var/hacmp/log/ccv_collector_data.json file.
You can use file transfer method such as file transfer protocol (FTP) or secure copy (SCP) to retrieve the output that is generated from the cl_ccv_collector command to a cluster node where you can run the cl_ccv_comparator command. The node that is used to run the cl_ccv_collector command need not be a part of either clusters that are being compared. Use the cl_ccv_comparator command to compare clusters and to generate the output file, /var/hacmp/log/ccv_comparison_data.json.
You can view the output file by using any text editor or you can use a JSON display tool to view the comparison results graphically. You can convert the output file to other data formats such as comma-separated values (CSV) format, which can be viewed by using a spreadsheet tool.
- Setting up the CCV utility for recurring use
-
If your enterprise allows the configuration of the SSH and SCP on your cluster nodes, you can automate the data collection and the cluster comparison process by using the clccv command. The clccv command takes the names of the cluster nodes as an input. The clccv command uses SSH to run the cl_ccv_collector command, and uses SCP to retrieve the output generated. If you want to use CCV utility regularly, but your enterprise does not allow you to enable of SSH or SCP continuously, you might enable them for a short duration. For example, you can enable them during a maintenance window. You can disable SSH and SCP after you use the CCV utility. You can retain the SSH or SCP configuration after you use the CCV utility to avoid configuring SSH or the SCP when you want to use the CCV utility the next time.
If you want to use the CCV utility regularly, and your enterprise allows password-less SSH authentication and SCP to be enabled continuously, you can set up the clccv command to start the data collection and the cluster comparison process on a recurring basis by using the at command. If your enterprise does not allow password-less SSH authentication and SCP to be enabled, you must use the clccv command interactively and provide a password when you are prompted. The results of the cluster comparison are written to the/var/hacmp/clccv/clccv.<first cluster name>.<second cluster name>.<date>.json file.