It's also highly recommended that you capture the current status of the operating system running the Ceph software and also the status of the underlying hardware. Capturing things such as CPU and RAM usage will alert you to possible resource starvation before it potentially becomes critical. Also, long-term trending on this data can help to plan hardware choices for Ceph. Monitoring of the hardware to capture hardware failures, such as disks, PSUs, and fans, is also highly recommended. Most server hardware is redundant and may not be obvious unless monitored that it is running in a degraded state. In addition, monitoring network connections so that you can be sure that both NICs are available in bonded configuration are working is also a good idea.