How To
Summary
This document describes the steps to take when a vSAN disk is replaced in IBM Cloud for VMware Solutions. This procedure includes how to check the status of the new disk, set the permissions of the new disk, and re-create the disk group in vCenter.
Environment
IBM Cloud for VMware Solutions
VMware
vSAN
Steps
If you experience a vSAN disk failure, then you perform this procedure after the IBM Cloud team replaces the failed vSAN disk in your infrastructure. Use these steps to configure the new disk, mark it as Flash disk, and create a new disk group so you can bring the server online.
1. Start the host, and connect to the host by using SSH. Run the following command:
/opt/lsi/storcli/storcli /c0 show
The command output varies depending on the steps taken by the data center team when they replaced the disk.
2. To configure JBOD on the new disk, review the Physical Drives (PD) list output, determine which scenario applies, and follow the steps for that scenario.
Scenario 1: Newly replaced disk shows in the PD List in UGood state
Example: In the screen capture that follows, disk 8:6 was replaced. It appears in the list and its state is UGood.
Now you can configure the JBOD on the new disk by running the following command, adjusting the drives= parameter to your output (in the example, 'drives=8:6' in the command corresponds to EID:Slt column in the PD list):
/opt/lsi/storcli/storcli /c0 add vd type=raid0 name=VSAN-SSD drives=8:6
Scenario 2: Newly replaced disk does not appear in the PD list. It shows in a separate JBOD list.
In this example, the newly replaced disk at slot 65:0 appears with state JBOD.
In this example, the newly replaced disk at slot 65:0 appears with state JBOD.
JBOD LIST :
=========
-----------------------------------------------------------------------
EID:Slt DID State DG Size Intf Med SED PI SeSz Model Sp Type
-----------------------------------------------------------------------
65:0 13 JBOD - 1.746 TB SATA SSD Y N 512B Micron_5200_MTFDDAK1T9TDN U -
Configure the disk by running the following command:
/opt/lsi/storcli/storcli /c0 set jbod=off
After you run the command, the disk appears in the PD List with UGood state.
3. Next, you update the name of the JBOD and set no readahead and writethrough.
(Note: The v6 parameter in the following examples corresponds to the Slt column in the example PD list output. Replace as needed with the values from your PD list.)
- To rename the JOB, run /opt/lsi/storcli/storcli /c0/v6 set name=VSAN-SSD
- To set no readahead, run /opt/lsi/storcli/storcli /c0/v6 set rdcache=nora
- To set writethrough, run /opt/lsi/storcli/storcli /c0/v6 set wrcache=wt
4. Run this command again: /opt/lsi/storcli/storcli /c0.
You see output similar to what's shown in this example screen capture:
5. Log in to vCenter. Go to the host in question and select Configure -> Storage Devices.
Find the newly replaced disk, which usually lists "type" as “disk” and “Drive Type” as “HDD”.
Select the option Mark as Flash Disk.
6. Go to the cluster where the host exists. Select Configure -> Disk Management.
Scroll until you find the host for which you are trying to re-create the disk group.
7. Select the host in question, and then select Create Disk Group.
8. The Disk Group creation wizard appears. Select the small disk for the “cache” tier, and then select the remaining larger disk for the “capacity” tier.
9. Allow some time for the disk group to be created and for VSAN to rebalance onto the new disk group.
Document Location
Worldwide
[{"Type":"MASTER","Line of Business":{"code":"","label":""},"Business Unit":{"code":"BU053","label":"Cloud \u0026 Data Platform"},"Product":{"code":"SSCLB3","label":"VMware Solutions"},"ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All Versions"}]
Product Synonym
VMware Solutions; VMware; vSAN
Was this topic helpful?
Document Information
Modified date:
25 January 2022
UID
ibm16538026