You can monitor the Mine cluster health and activity through Prism. To determine the AOS version of your Mine the cluster is running, go to About Nutanix option under user_name drop-down list in Prism main menu.
Mine with Veeam Dashboard
The Prism web console includes a custom Mine with Veeam dashboard specific to a Mine™ cluster, which appears by default when you first log on to Prism. To view this custom dashboard at any time, select Mine with Veeam from the pull-down list on the far left of the main menu.
The custom dashboard displays the following eight information tiles (widgets):
-
Cluster. The cluster tile provides quick visual indicators for the following components: cluster health (overall), Veeam implementation, Mine™ components, and storage capacity (available). Green indicates a healthy component, yellow indicates a warning condition, and red indicates a critical condition.
-
Click the Launch Console link to open a Veeam console.
-
Click the Mine Platform link to open the Mine™ console.
-
-
Physical Cluster Usage. The physical cluster usage tile displays a timeline graph of storage usage in the cluster.
-
Storage Throughput. The storage throughput tile displays a timeline graph of storage throughput in the cluster.
-
Protection. The protection tile displays a summary of the VMs and hosts protected currently.
-
Capacity Usage. The capacity usage tile summaries the current storage usage and available capacity.
-
Job Status. The job status tile displays a summary of the running, disabled, and idle jobs in the cluster currently.
-
Capacity Trends. The capacity trend tile displays a timeline graph of free space in the cluster, including a low space warning line. The widget requires some time to collect data before it starts to show capacity trends. The widget simply displays "N/A" until enough data is collected to show a trend line.
-
Veeam Backup & Replication Alerts and Events. The Veeam alerts and events tile displays a list of the current Veeam-specific alerts and events.
-
Nutanix Alerts. The Nutanix alerts tile displays a list of the current Nutanix-specific alerts.
Managing Storage
When a Mine™ cluster is full, the cluster may become unavailable, and you will not be able to continue backup operations. To prevent such a situation, Mine™ includes a special monitoring feature (sometimes referred to as a "watchdog") that dynamically monitors storage usage and takes action as necessary to avoid reaching a storage full condition. If available storage space in the cluster falls below the minimum amount, the monitor automatically stops and disables Veeam Backup & Replication jobs. The monitor is regulated by three thresholds:
-
"Low on space" threshold (restart_jobs_threshold_percent parameter). When available storage space on a cluster reaches the specified threshold, Veeam Backup & Replication starts to upload the VM backup files to Capacity Tier. For details, see the "Capacity Tier" section of the Veeam Backup & Replication User Guide.
-
"Job processing is impacted" threshold (cancel_jobs_threshold_percent parameter). When available storage space on a cluster reaches the specified threshold, Mine™ disables all Veeam Backup & Replication jobs. Already started jobs continue the session, but subsequently scheduled runs are suspended until more storage space becomes available.
-
"Immediate stop" threshold (stop_issuing_jobs_threshold_percent parameter). When available storage space on a cluster reaches the specified threshold, Mine™ immediately stops all running jobs.
The monitor automatically calculates and defines the threshold values according to your environment resources. (If you want to change default threshold values, contact technical support.) In addition, the monitor regulates the location of VMs. If the Veeam repository extends and backup proxies are deployed on different AHV nodes, the monitor transfers them to one node.
By default, Mine™ reserves enough storage space for node rebuild should one of the nodes fail. If you want to add new VMs on the cluster or change the default monitor threshold values, consider leaving enough space for the rebuild of the node.