Reloading the OS

The OS reload process is used to rebuild a IBM Cloud® Juniper vSRX server.

The process typically performs the following actions:

Reload the server host's operating system (OS). This action generates a new host OS password.
Install KVM in the operating system
Create a vSRX VM in the KVM
Reconfigure the vSRX with the default configuration for IBM® Cloud

The process can take between 2 and 5 hours to complete. Stand-alone gateways are out of service during this period. For Juniper High Availability (HA) gateways, when you reload the OS on one of your servers, the vSRX fails over to another server in the cluster and continues to process data traffic. After the reload completes, the server rejoins the cluster.

If the OS reload is changing your vSRX version, then refer to Upgrading using OS Reload to understand the downgrade and upgrade process.

For a successful OS Reload on a vSRX, ensure the following:

The vSRX configuration should not be modified during the OS reload operation. For example, automated software agents that attempt to modify one or both vSRX nodes. Configurations changes like these can corrupt the OS reload.
The root password for the provisioned vSRX gateway must match the root password that is defined in the vSRX portal. The password in the portal was defined when you first provisioned the gateway, and it might not match the current gateway password. If so, then use SSH to connect to the vSRX gateway and change the root password to match. You can then proceed with the OS reload operation.
The vSRX configuration must allow root SSH access to the vSRX private IP before the OS reload request. This is required to rejoin the cluster. After the OS reload completes, if wanted, you can disable SSH access again.
"Do NOT" perform an OS reload on both servers of the Highly Available gateway at the same time.

Performing an OS reload on both servers of the HA gateway at the same time destroys the vSRX cluster and causes the gateway to be out of service. If the vSRX cluster is destroyed, you must use the Rebuild cluster option to reprovision the vSRX and re-create the HA cluster.
The IPMI interface for the bare-metal server must be enabled or the following error is issued:
```
You cannot toggle the IPMI interface while transactions are running.
```

Performing an OS reload

To reload your OS, follow these steps:

From your browser, open IBM Cloud console and log in to your account.
Run a Readiness check for “OS Reload” and address any errors that are found.
Select the Menu icon from the upper left, then click Classic Infrastructure.
Choose Network > Gateway Appliances.
Click the server that you want to reload.
Click the server name in the Hardware section.
Select OS Reload from the Actions menu on the upper right of the page.
In the OS Reload screen, click Edit for the Category that requires an update. Select Juniper as the Vendor, and the OS version you want to reload.
Click Reload Above Configuration to proceed to the Review window. Click Cancel to cancel the changes to the device and exit the screen.
Verify that all details in the New Configuration section are correct. Click Next to advance to the Confirm window.
Click Confirm OS Reload to confirm and initiate the OS Reload. Or click Cancel to cancel the action.

vSRX version mismatches

A IBM Cloud® Juniper vSRX cluster must have the same vSRX version on each node to fully support the High Availability feature. If a cluster has mismatched vSRX versions, then you must reload the OS of the node with the older version (by using the procedure in this topic) so that both nodes are at the same version level.

A version mismatch applies to both minor release mismatches and major release mismatches.