Newer documentation is now live.You are currently reading an older version.

Switch to latest✕

Replace a failed Ceph OSD with a metadata device

Warning

This procedure is valid for MOSK clusters that use the unsupported KaaSCephCluster custom resource (CR) instead of the MiraCeph CR that is available since MOSK 25.2 as a new Ceph configuration entrypoint.

For the equivalent procedure with the MiraCeph CR, refer to the following section:

Replace a failed Ceph OSD with a metadata device

The document describes various scenarios of a Ceph OSD outage and recovery or replacement. More specifically, this section describes how to replace a failed Ceph OSD with a metadata device:

• If the metadata device is specified as a logical volume in the BareMetalHostProfile object and defined in the KaaSCephCluster object as a logical volume path

• If the metadata device is specified in the KaaSCephCluster object as a device name

Note

Ceph OSD replacement implies the usage of the KaaSCephOperationRequest custom resource (CR). For workflow overview, spec and phases description, see High-level workflow of Ceph OSD or node removal.

• Replace a failed Ceph OSD with a metadata device as a logical volume path
  • Remove a failed Ceph OSD by ID with a defined metadata device
  • Re-create a Ceph OSD with the same metadata partition
• Replace a failed Ceph OSD disk with a metadata device as a device name
  • Remove a Ceph OSD with a metadata device as a device name
  • Re-create the metadata partition on the existing metadata disk
  • Re-create the Ceph OSD with the re-created metadata partition