Symptoms

TimeFinder SnapVX Snapshots RDP space is reaching capacity thresholds. Replication Data Pointer (RDP) threshold alert Replication Data Pointer has reached 50% capacity. Replication Data Pointer has reached 60% capacity. Replication Data Pointer has reached 70% capacity. Replication Data Pointer has reached 80% capacity. Replication Data Pointer has reached 85% capacity. Replication Data Pointer has reached 90% capacity. Replication Data Pointer has reached 95% capacity. Replication Data Pointer Meta Data Usage reached 90-99%.

Cause

The definition of this alert is that the array's Replication Data Pointer (RDP) reached or surpassed a set threshold value for percent utilization, while using TimeFinder SnapVX for local replication operations on the array. An alert is generated by the array and dials home to Dell to inform the RDP space reached a certain utilization percentage. A subsequent email is sent from Dell to the user contact for the array to notify of the issue. RDP is an internal File System on the array which keeps Metadata pointers for SnapVX Snapshots, and allows access to the version or original data that is protected by a SnapVX Snapshot. The number of pointers that are allowed in RDP is finite; by default, roughly 24% of the Metadata space is devoted for Back-End volumes (TDATs) on the array. Once this RDP space reaches 100%, SnapVX Snapshots may fail because the array is no longer able to track new versionings of data on Source devices that is not previously versioned within the active Snapshot.

Resolution

Instructions: Review the current utilization of RDP space in the array.Review your current SnapVX Snapshots and their device utilization.Take necessary actions to reduce the RDP usage. Recommended Action Options: Terminate unneeded SnapVX Snapshots. Review your Snapshots. One or more Snapshots may not be needed any longer. Note: If you have one or more Linked Targets that are associated with the Snapshot, the Linked Targets must be unlinked before the Snapshot can be terminated. Consider using self-expiring timers on SnapVX Snapshots.If the refresh rate of the Source volumes on the Snapshots is too high, consider using TimeFinder Clone. TimeFinder Clone does not use RDP for version tracking. Expand the amount of RDP space. At HYPERMAX OS 5977.810.784 and later, there is a capability to expand the RDP consumption space in the cache from default values. Contact your local Dell Technologies representative or the Dell Technologies VMAX Support team. How to View RDP usage: The following are available on Dell Software and HYPERMAX OS levels or later: Solutions Enabler: 8.2.xUnisphere for VMAX: 8.2.xHYPERMAX OS 5977.810.784 Solutions Enabler: Issue a symcfg list -v, where you can use 'Replication Usage' with grep (Linux) or FIND (Windows). Unisphere for VMAX: Go to > Data Protection > Protection DashboardA 'Replication Cache Used %' or 'Replication Utilization' bar is present. Unisphere for PowerMax: Select the array serial number > Select capacity > Enable Show detailed > System Usage > Replication Meta Data OR Select the array serial number > Performance > Dashboard > Select Array > Array Efficiency. Note: The term for RDP has been rephased as "Metadata Replication Used" as per online help.% Metadata Replication Used - The used percentage of the allocation for local replication (RDP) in the metadata page-file. How to Monitor for Alerts for RDP usage: The following are available on Dell Software and HYPERMAX OS levels or higher: Solutions Enabler: 8.2.xUnisphere for VMAX: 8.2.xHYPERMAX OS 5977.810.784 Solutions Enabler:Use symevent monitor to monitor for real-time events.Use symevent list to list past events. When the alert occurs, the following message appears: The device pool or SRP is almost full. Note: See the Solutions Enabler 8.x Array Controls and Management CLI Guide that is on https://support.emc.com for further details about symevent. Unisphere for VMAX or Unisphere for PowerMax:Go to Home > Administration > Alert Settings > Symmetrix Threshold Alerts. Ensure that Local Replication Utilization is enabled. The default notification threshold cannot be changed, but a new alert policy can replace it. Click the Create button and follow the wizard prompts to create your thresholds. To be notified by Unisphere once the alert is set up, go to Home > Administration > Alert Settings > Notifications. You can enable Syslog, Email, or SNMP alerting.

Support Cases