Data Collection

Topics | How To | Troubleshoot | Related Topics

Overview

After installing the SRM Server, the SRM Agent software must be installed on your selected client computers before you can collect data from them. When data collection is performed, this information populates the SQL database on the SRM Server's repository with current statistics, for the purpose of generating summaries and reports.

When you configure data collection for a client, from the SRM node you have the options of choosing which Agent(s) to use (if more than one is available on the client), the type of data collection to be performed (determined by selecting either an Analysis or Discovery Job Type), a Schedule Policy, and a start date. You can exclude volumes and directories from the data collection process for Subclients filters. You can also balance the collection load across different subclients, each with a different schedule policy.

Data collection jobs run automatically according to the parameters you have selected, and will appear in the Job Controller as a scheduled job. The system can also be configured to generate advisories, a brief text message describing an actual or potential problem and the suggested corrective action.

Reports and SRM Summaries viewed through the console are generated from the data collected and stored on the SRM Server. Schedule your first reports to run after your first data collection jobs have completed. Note that data collection and Report Generation jobs are scheduled independently, at different times and intervals. Thus, you can, for example, collect data nightly, hourly, or on-demand, but generate reports only once per week if desired. For more information about summaries, see SRM Summaries. For more information about running and viewing reports, see Reports.

Scheduling Data Collection jobs

Data Collection jobs can be scheduled to regularly collect data at specified intervals. Data Collection jobs can be scheduled from the SRM agent level in the CommCell Browser. In addition, Analysis Data Collection jobs can be scheduled for the individual subclients at the subclient level. When the Data Collection job is scheduled the job will be displayed in the Job Controller. After the successful completion of the Data Collection job, the client information is gathered at the SRM Server and the job status is shown as completed in the Job Controller. For more information on scheduling, see Scheduling. For step-by-step instructions on scheduling a Data Collection job see Run/Schedule Data Collection.

Considerations

Data Collection is not performed for mounted network drives, volumes, or directories. If you want to collect data for network file systems, install the appropriate SRM Agent on the local computer for those file systems.
SRM Exchange Agent based Analysis Data Collection job will report storage groups only if the storage groups are mounted.
For NAS Data Collection, all CIFS and NFS content will be included as content in the default subclient, unless some or all of it is specifically listed as content in a user-defined subclient.
For NAS Data Collection, if subclient SubB is child of subclient SubA meaning that SubB's paths are contained in paths of SubA, then when collecting SubA none of the contained paths of SubB will be collected. Instead they will be collected by Data Collection of SubB.
For UNIX file systems, all mount points (comprised of a logical volume or partition) will be included as content in the default subclient, unless they are specifically listed as content in a user-defined subclient.
Consider this example:
- You have created a subclient named sub1.
- You have defined a directory named /mount_points as contents for sub1.
- Within /mount_points there are three mount points, mnt1, mnt2, and mnt3.
- Data will not be collected on those three mount points as part of sub1; instead, they are part of the content of the default subclient.
- To collect data on those mount points as part of sub1, you must specify each mount point, /mount_points/mnt1, /mount_points/mnt2, and /mount_points/mnt3, as content for sub1.
If you would like to configure the SRM Exchange Agent for cross-domain mailbox access for data collection purposes, this can be accomplished by creating the AdServerList key in the registry.
After adding, editing or removing a subclient, you must run an Analysis data collection job on all subclients associated with an agent for reports to reflect the changes.
For the SRM Agent for SharePoint only - ensure that all necessary SharePoint services are running and that the configuration and content databases are online; otherwise, installation may appear successful, but a subsequent Data Collection will fail.

Best Practices

Oracle
Consider the following configuration items before running data collection from the Oracle Agent:
- Verify that Oracle databases are in OPEN status before running a data collection job.
- Verify that the Oracle databases are running properly, and that there are no database-related memory issues.
- Verify that the shared_pool_size parameter in the init<SID>.ora file is set to the maximum valid value so that SQL queries can run successfully for data collection.
- Data collection by the appropriate file system SRM Agent must be scheduled to occur before data collection by the SRM Oracle Agent.
SharePoint
"Usage Analysis" must be enabled in SharePoint for SRM to collect data about hits, for those reports that include this information. Refer to Microsoft SharePoint documentation for information about enabling "Usage Analysis".
Solaris
Ensure that the NFS client specifies the soft option when mounting NFS shares, or else data collection for NFS data will hang whenever the NFS server is down or not accessible. Also, increase the value for the retrans and timeo options; this will compensate for the soft option limiting the number of retries that the client can attempt to access the NFS server.
Windows File System
Activating Data Classification will decrease the amount of time required to perform an Analysis level data collection job for the SRM Windows File System Agent.
NAS
Configure NAS filer (including NetApp vFiler, EMC Celerra CIFS Servers and Data Movers) with DNS server in a way that reverse DNS lookup (from NAS filer to SRM NAS proxy) is resolved. If that is not the case then CIFS connection or NFS mount from SRM NAS proxy to NAS filer will fail.