How to reduce wear on the boot drive of a Netapp HCI compute node
Applies to
- HCI Compute Hypervisor andPlug-in
- HCI Deployment Engine/MonitoringAgent
- SolidFire Host Plugins
- Third Party Application
- Third Party Operating System
Description
Issue:
When using boot media such as USD, SD, or NVDIMM, keeping the system logs on that media results in a high occurrence of update and writes on that media. This will, over time, degrade the boot media. Also, if a core dump occurs, this configuration can be problematic since the boot media may not have the necessary space to hold the core file.
Finally, if the boot media holding logs or core dumps fails, troubleshooting the cause of the failure becomes more difficult since the logging and/or core dumps become inaccessible.
Alternative:
Host logging can be moved to a shared storage location, preventing degradation of the boot media over time due to busy log writes. Moving the core dump location to shared storage ensures that, when a condition occurs which generates a core dump, it will be written safely.
For more details on how to configure this alternative, either log into your NetApp support account when viewing this KB or contact NetApp support.
Log Insight:
While not directly part of the solution, it is also a strong suggestion to forward all vCenter and ESXi hosts logs via Knowledge BaseSyslog to Log Insight or a comparable syslog solution. This is very powerful in that all logs can be viewed on the same time scale, which is quite useful for troubleshooting.