The vSAN cluster previously experienced disk failure on one or more hosts; following disk replacement, the following symptoms were observed.
There are no VMs or any workload impacted.
VMware vSAN 8.x
The inaccessible objects are stale and belong to FSVM(File Server VM) which were residing on the host with a faulty disk.
The objects are stale as the components which make up these objects resided on failed disk.
Example of an inaccessible FSVM object impacted by this issue:
Object UUID: ########-####-####-####-############ Version: 20 Health: inaccessible - Lost data availability.(APD) Owner: 00000000-0000-0000-0000-000000000000 Size: 0.00 GB Used: 0.00 GB Used 4K Blocks: 0.00 GB Policy: stripeWidth: 1 proportionalCapacity: [0, 100] hostFailuresToTolerate: 0 affinity: ['########-####-####-####-############'] forceProvisioning: 1 affinityMandatory: 1 spbmProfileId: ########-####-####-####-############ spbmProfileGenerationNumber: 0 objectVersion: 20 migrateForDecom: 1 CSN: 1237 spbmProfileName: FSVM_Profile_DO_NOT_MODIFY locality: HostLocal
Configuration:
Component: ########-####-####-####-############ Component State: ABSENT, Address Space(B): 8589934592 (8.00GB), Disk UUID: ########-####-####-####-############, Disk Name: N/A Votes: 1, Host UUID: None
Type: N/A Path: N/A Group UUID: ########-####-####-####-############ Directory Name: N/A
It could be seen that :
They belong to FSVMs from the "sbpmprofilename".
This object belongs to FSVM pinned to host with UUID mentioned in the affinity parameter. To find this host the output of command esxcli vsan cluster get can be validated.
The object is inaccessible as the component is absent from "Component State"
The inaccessible object is stale as the "Disk Name" and "Host UUID" is empty.
If the above symptoms and cause match, kindly open a case with Broadcom Technical Support to assist with clearing up the stale inaccessible vSAN objects.