Overview of call home

IBM Storage Scale users can use the call home feature to share the basic configuration data of a cluster with IBM support and development team. The shared details ensure that the cluster is configured to function according to its maximum potential.

With the call home feature enabled, the IBM Storage Scale users can upload the daily or weekly cluster configuration or debugging data from the customer site to a specific IBM® server. The IBM server then sends the data to the IBM backend, Enhanced Customer Data Repository (IBM ECuRep). For more information, see ECuRep.

The uploaded data is further analyzed by the IBM support team to determine whether there is a common misconfiguration, best practice violation, or any other issues. This enables the IBM support team to provide solutions to any problems encountered on the customer site or leverage the data to provide more benefits to the customer. For more information on understanding the call home benefits and supported use cases, see Benefits of enabling call home.

The call home component in IBM Storage Scale must be configured before it can be used for manual or automated data uploads to the IBM server. The call home configuration process consists of the following steps:

  1. Configuring the call home settings.
  2. Creating call home groups.

For more information, see Configuring call home.

Call home architecture

The following figure shows the basic call home group structure:

Figure 1. Call home architecture
One call home node is defined here by using two call home child nodes and one call home server. The collected data is transferred to the IBM server.

Call Home Group
The Call Home Group is a group of call home nodes that are configured by using the mmcallhome group command. A call home group consists of at least one child node, which also acts as its call home node. A call home group can have more than one child node but has only one call home node. Multiple call home groups can be configured within an IBM Storage Scale cluster. You can automate the call home group creation by using the mmcallhome group auto command.
Call home groups help to distribute the data-gather and data-upload workload to prevent bottlenecks. You can create groups of any size between one and the number of nodes in your cluster. The larger the group, the higher the workload on the call home node. Therefore, you are recommended to limit the group size to 1024 nodes. Larger groups are possible, but it might lead to performance issues on the call home node.
Call Home Node
The Call Home Node is the node that performs the data upload. If a scheduled data gathering is enabled, then the call home node initiates the data collection within its call home group and uploads the data package to the IBM support center. A gather-send task runs on the call home node, which collects the data from the call home child nodes and uploads that data to a specific IBM server. The IBM server then sends the data to the IBM backend, ECuRep. For more information, see ECuRep. The gather-send configuration file includes information about the data that is collected from the child nodes.
Note: The call home node is also a child node of the group. If the call home node becomes unavailable, then the whole call home group is unable to perform any data uploads until the call home node is online again.
Important: The call home node needs to have access to the external network through 443 port number. For more information on network-related requirements, see Installing call home.
Call Home Child Node
The Call Home Child Node is a member of a call home group. The call home node can collect data from all the call home child nodes in a call home group.
Note: If a call home child node, which is not a call home node, becomes unavailable, then the rest of the call home group continues to work. However, the scheduled uploads do not include the details of the unavailable child node.

mmcallhome commands impact

The call home data upload and share are managed by using the mmcallhome command.

The mmcallhome command provides options to configure, enable, run, schedule, and monitor call home related tasks in the IBM Storage Scale cluster. Information from each node within a call home group is collected and securely uploaded to the IBM ECuRep server.

The uploaded packages that contain the daily or weekly scheduled uploads and non-ticket-related sent files are saved on the corresponding call home node for an inspection by the customer.

You can use the mmcallhome status list --verbose command to find the exact location of the uploaded packages. A maximum of 1 GB of space is used by the packages of each of these categories:
  • Sent packages
  • Failed uploads
A maximum of 48 such packages can be kept.

The uploaded data is stored for at least two weeks on IBM ECuRep, and can be identified by using your customer information. If you need to access this data, contact IBM support. For more information, see ECuRep. The PMR or Salesforce case-related data is saved and the PMR or Salesforce case is still open.

The following mmcallhome command options respond in the same way when run on nodes, which belong or do not belong to a call home group:

  • mmcallhome group
  • mmcallhome capability
  • mmcallhome info
  • mmcallhome proxy
Note: For compatibility reasons, in a mixed cluster configuration, the mmcallhome capability, mmcallhome info, and mmcallhome proxy commands are available only to the global settings if the corresponding nodes are not part of a call home group. If the call home node of this group also has IBM Storage Scale 4.2.3 PTF 6 or older nodes, then they also manage a separate settings configuration for their group.
All other mmcallhome command options
All other mmcallhome commands can be run only from a node, which is a member of a call home group.

For more information, see mmcallhome command.