Skip to main content

Replacing a memory DIMM

While a node might be able to self-correct for certain memory errors, failed memory can lead to system degradation. Failed memory should be replaced as soon as possible. The indications of a failed DIMM are:

  • A post error on boot

  • An error in Lenovo XClarity Administrator

  • An error in the Prism web console

  • Not all memory is detected. For example, the appliance should have 256 GB per node and the host only shows 240 GB.

Here are the steps for replacing a memory DIMM:

  1. Identify the failed DIMM by either checking in XClarity Administrator or using the IMM web console to browse the post event log. The event log will contain information about the location of the DIMM fault.

  2. Shut down the node by following the hypervisor specific Node Shutdown procedure described in the Nutanix Hardware Replacement Documentation.

  3. Power off the node and disconnect all the cables if the cable management arm is not in place. Pull the node chassis out of the rack.

  4. Remove the top cover as described in Removing the top cover.

  5. Remove the air baffle as described in Removing the air baffle.

  6. Replace the memory DIMM as described in Replacing a memory module.

  7. Replace the air baffle as described in Replacing the air baffle.

  8. Replace the top cover as described in Replacing the top cover.

  9. Push the node chassis back into the rack and reconnect the cables as necessary. Power on the node.

  10. Start the node by following the hypervisor specific Node Start procedure described in the Nutanix Hardware Replacement Documentation.

  11. Verify that the DIMM memory failure is resolved by checking the following:

    • No error in the post event log.

    • No error shown in Lenovo XClarity Administrator.

    • No error shown in Prism.