The Infra tab isolates issues and causes of infrastructure problems. Further are described the frames you can see on the tab.
The Service Alerts – Infrastructure Alerts by Application name graph shows the service alerts collected by the New Relic infrastructure agent. This will show service restarts, many associated with deployments.
The Inode usage by mount frame shows inode usage by mount across the selected timeframe. Even though there may be plenty of storage free, if a node runs out of inodes, it will show a lack of available storage. Removing files (especially small ones) will free up both space and make inodes available.
The vCPU tier view over timeline GREATER 2 weeks frame shows vCPU tier view across the selected timeframe of more than two weeks. This frame looks at the number of vCPUs assigned to the New Relic application name shown.
The vCPU tier view over timeline frame shows vCPU tier view across the selected timeframe of more than 24 hours. This frame looks at the number of vCPUs assigned to the New Relic application name shown. It will show both cluster upsizes and downsizes.
The vCPU tier view over timeline BY NODE frame shows vCPU tier views across the selected timeframe by node. This frame is helpful in detecting loss of node(s) or when nodes are upsized or downsized. vCPU tier view over timeline BY NODE, should look at timeline LESS than 24 hours.
The Instance details table shows instance details of each New Relic application.
The Logging, if there is a broken line for a node, it indicates non-responsive node during that time period frame shows non-responsive nodes across a time period.