Updrading SL1 Extended Architecture

This section provides detailed steps for performing an upgrade on the SL1 Extended Architecture.

New installations of SL1 Extended Architecture are available only on SaaS deployments.

Use the following menu options to navigate the SL1 user interface:

To view a pop-out list of menu options, click the menu icon ().
To view a page containing all of the menu options, click the Advanced menu icon ().

Workflow

The following sections describe the steps to plan and deploy an SL1 update.

If would like assistance planning an upgrade path that minimizes downtime, contact your Customer Success Manager.

The workflow for upgrading SL1 is:

Plan the update.
Schedule maintenance windows.
Review pre-upgrade best practices for SL1.
Back up SSL certificates.
Set the timeout for PhoneHome Watchdog.
Adjust the timeout for slow connections.
Run the system status script on the Database Server or All-In-One before upgrading.
Upgrade the SL1 Distributed Architecture using the System Update tool (System > Tools > Updates).
Remove SL1 appliances from maintenance mode.
Upgrade the Extended Architecture.
Upgrade MariaDB, if needed.
Reboot SL1 appliances, if needed.
Restore SSL certificates.
Reset the timeout for PhoneHome Watchdog.
Update the default PowerPacks.
Configure Subscription Billing (one time only). For details, see .

For details on all steps in this list except step 10, see the section on Upgrading SL1.

Prerequisites

ScienceLogic recommends that for production systems, each Compute Cluster contains six (6) Compute Nodes. Lab systems can continue to use Compute Clusters that include only three (3) Compute Nodes.
The Storage Cluster requires a (possibly additional) node to act as the Storage Manager.
Perform the installation steps in the Installation manual to install these additional nodes (for the Computer Cluster and the Storage Cluster) before upgrading your existing nodes.
Ensure that all nodes in the SL1 Extended Architecture can access the internet.
You must use the same password for the em7admin account during ISO installation of the Database Server and ISO installation of the appliances in the SL1 Extended Architecture.

To perform the upgrade, you must have a ScienceLogic customer account that allows you access to the Harbor repository page on the ScienceLogic Support Site. To verify your access, go to https://registry.scilo.tools/harbor/. For more information about obtaining Harbor login credentials, contact your Customer Success Manager.

Resizing the Disks on the Compute Node

The Kafka Messaging service requires additional disk space on each Compute Node. Before upgrading, ensure that each disk on each existing Compute Node in the Compute Node cluster is at least 350 GB.

If each disk on each existing Compute Node is not at least 350 GB, perform the following steps on each Compute Node:

Resize the hard disk via your hypervisor to at least 350 GB.
Note the name of the disk that you expanded in your hypervisor.
Power on the virtual machine.
Either go to the console of the Compute Node or use SSH to access the Compute Node.
Open a shell session on the server.
Log in with the system password for the Compute Node.
At the shell prompt, enter:

sudo lsblk | grep <disk_size>

where:

disk_size is your hard disk size from step #1.

Note the name of the disk that you expanded in your hypervisor.
At the shell prompt, enter:

sudo fdisk /dev/<disk_name>

where:

disk_name is the name of the disk you want to expand.

Enter p to print the partition table.
Enter n to add a new partition.
Enter p to make the new partition the primary partition.
Select the default values for partition number, first sector, and last sector.
Enter w to save these changes
Restart the VM.
At the shell prompt, enter:

sudo fdisk -l

Notice that now another partition is present.
To initialize the new partition as a physical volume, enter the following at the shell prompt:

sudo pvcreate <partition_name>

To add the physical volume to the existing volume group, enter the following at the shell prompt:

sudo vgextend em7vg <partition_name>

To verify and confirm that the volume group has grown to the expected size, enter the following at the shell prompt:

sudo vgdisplay | grep "VG Size

Installing ORAS

If you have not already installed OCI Registry as Storage (ORAS), you will need to do so before you can upgrade the SL1 Extended Architecture.

To do so:

Use SSH to access the Management Node. Open a shell session on the server. Log in with the System Password you defined in the ISO menu.
In the Management Node, navigate to the sl1x-deploy directory. To do this, enter the following at the shell prompt:

cd sl1x-deploy

Run the following commands:

sudo su

curl -LO https://github.com/oras-project/oras/releases/download/v0.12.0/oras_0.12.0_linux_amd64.tar.gz

mkdir -p oras-install/

tar -zxf oras_0.12.0_*.tar.gz -C oras-install/

mv oras-install/oras /usr/bin/

rm -rf oras_0.12.0_*.tar.gz oras-install/

exit

Obtaining Your Harbor Credentials

You will need to know your Harbor username and CLI secret when you upgrade the SL1 Extended Architecture. To obtain these credentials:

Log in to Harbor at: https://registry.scilo.tools/harbor/sign-in?redirect_url=%2Fharbor%2Fprojects
Click Login via OIDC Provider.
Click Customer Login.
Log in with the username and credentials that you use to access the ScienceLogic Support site (support.sciencelogic.com).
Click the username in the upper right and select User Profile.
On the User Profile page:

Note the username.
Click the pages icon next to the CLI secret field to copy the CLI secret to cache.

Exit the browser session.

Upgrading to 12.3.x

Before upgrading to SL1 12.3.0 or later, you must already be running SL1 on Oracle Linux 8 (OL8). If you are on a version of SL1 prior to 12.2.0 and running on OL7, you must first upgrade to SL1 12.1.1 or 12.1.2 and then migrate to OL8 before you can upgrade to SL1 12.3.x. For an overview of potential upgrade paths and their required steps, see the appropriate 12.3.x SL1 release notes.

To upgrade the SL1 Extended Architecture to 12.3.x from 12.1.x or 12.2.x instances running on Oracle Linux 8 (OL8), follow these steps:

Complete preupgrade steps.
Disable Scylla.
Upgrade the SL1 Extended Architecture.
Upgrade the SL1 Distributed Architecture.

Step 1: Preupgrade

If you have not already done so, you must install ORAS and obtain your Harbor credentials, which you will need for step 4.
Use SSH to access the Management Node. Open a shell session on the server. Log in with the system password you defined in the ISO menu.
In the Management Node, navigate to the sl1x-deploy directory. To do this, enter the following at the shell prompt:

cd sl1x-deploy
Log in to Harbor repository:

oras login registry.scilo.tools/sciencelogic/
- Enter the username you used to log in to the browser-based session of Harbor.
- Enter the password (CLI Secret) that you saved from the browser-based session of Harbor.
Exit out of the sl1x-deploy directory and download the deployment files:

cd /home/em7admin/

oras pull registry.scilo.tools/sciencelogic/sl1x-deploy:12.3.x

cd sl1x-deploy
Copy the inventory template file to the sl1x-inv.yml file:

cp sl1x-inv-template.yml sl1x-inv.yml
Edit the sl1x-inv.yml file to match your SL1 Extended system:

vi sl1x-inv.yml

Do not remove colons when editing this file.
- Make sure that the sl1_version value is sl1_version: 12.3.x.
- Supply values in all the fields that are applicable. For details on the sl1x-inv.yml file, see the manual Installing SL1 Extended Architecture, which can be obtained by contacting ScienceLogic Support.
- Save your changes and exit the file (:wq).
Pull the Docker image that is referenced in the docker-compose file:

docker-compose -f docker-compose.external.yml pull

Step 2: Disable the Scylla Cluster

To disable the Scylla cluster:

Use SSH to access the Management Node. Open a shell session on the server. Log in with the System Password you defined in the ISO menu.
Open a text editor for the sl1x-inv.yml file:

vi /home/em7admin/sl1x-deploy/sl1x-inv.yml
Edit the file to add the following additional variables:

all:

vars:

install_aiml: false

enableNonScyllaPipeline: true

enableLegacyScyllaPipeline: false
Run the following command to remove services that used the previous configuration:

docker-compose -f docker-compose.external.yml run --rm deploy app-purge
In the sl1x-inv.yml file, remove the Storage Node and Storage Manager IP addresses from the list. For example, you would remove the following lines:

sn:

hosts:

vars:

scylla_admin_username: em7admin

scylla_admin_password: <Scylla password>

sm:

hosts:

vars:

scylla_manager_db_user: em7admin

scylla_manager_db_password: <Scylla password>
Save your changes and exit the file (:wq).

Step 3: Upgrade the SL1 Extended Architecture

To upgrade the SL1 extended architecture:

Use SSH to access the Management Node. Open a shell session on the server. Log in with the System Password you defined in the ISO menu.
To upgrade the Management Node services, run the following script:

sudo bash package-update.sh
To upgrade RKE and Kubernetes on the Compute Nodes, run the following command:

docker-compose -f docker-compose.external.yml run --rm deploy cn
To update the SL1 Extended Architecture system services, run the following command:

docker-compose -f docker-compose.external.yml run --rm deploy app

Step 4: Upgrade the SL1 Distributed Architecture

Update your classic SL1 appliances. For more information, see the section on Updating SL1.

Upgrading to 12.2.x

Before upgrading to SL1 12.2.0 or later, you must already be running SL1 on Oracle Linux 8 (OL8). If you are on a version of SL1 prior to 12.2.0 and running on OL7, you must first upgrade to SL1 12.1.1 or 12.1.2 and then migrate to OL8 before you can upgrade to SL1 12.2.x. For an overview of potential upgrade paths and their required steps, see the appropriate 12.2.x SL1 release notes.

To upgrade the SL1 Extended Architecture to 12.2.x from 12.1.x instances running on Oracle Linux 8 (OL8), follow these steps:

Complete preupgrade steps.
Disable Scylla.
Upgrade the SL1 Extended Architecture.
Upgrade the SL1 Distributed Architecture.