 Hello there, this is Leo Choa from the CNF inbound interaction team at the continuing in Red Hat and today we are going to demonstrate how to install upstream run operators for 5G telco deployments To do so, we will start by installing depth of requisites then we will create the index catalog for the run operator from scratch Later on, we are going to configure the disconnected openchip clusters and lastly, we are going to deploy the run operators into that environment As commented during the intro, let's start by installing depth of requisites First, we installed the container tools model which includes Putman, Scopeo, among other tools The Putman utility will be used to create the container image index Meanwhile, Scopeo will be used to inspect the container image bundles looking for tags and other informations We need to install also Yellipz, which is a dependency of the OPM tool Finally, let's install the OPM CLI, which is a tool that helps us to initialize a catalog in the new file-based format also renders the run operator records into it and validates the catalog Let's now proceed to create the run operator's catalog index To do so, let's first clone the skeleton index structure Let's now take a look to the tree As we can see, there are going to be three folders and many files within There is going to be one folder for the cluster group of grades operator one for the hardware event processor operator and the run operator index folder is where we are going to store the index file Let's now populate the run catalog index with the package definition of the cluster group of grade operator and the hardware event processor operator We need to add channel entries for each one of these bundles as well Now, in order to render the operator's bundle into the new index we need to first look up for the available task for that bundle for that we are going to use a scope view Once that you know the available tags in the bundles let's now render the operator bundles using the 410.0 tag A very important step before building the container image index is to first validate the created index Once this is done, let's build the container image index and push it to our organization registry Optionally, we are now safe to remove the generated artifacts Before the mirroring task and installation of the new catalog source into our disconnected environment let's first take a look to the current cluster As you can see, we got only three sources now The certified operated index, the community operated index and the rehab operated index, which were mirrored before Let's now take a deeper look also to our disconnected registry which in our case was implemented using Quay As you can see, we got different namespaces one dedicated for the OCP release images and one dedicated for the OLM contents images After we finish this part of the demo we are going to see here also the run operators Let's start this part by installing the OC mirror utility This Open Chief CLI plugin is available as a technical preview feature since Open Chief 4.10 and it enables us to manage the mirroring task in a declarative way Now that we got OC mirror we can optionally inspect the created run catalog index To do so, let's first list all the available operator packages in the created catalog Now we are going to list all the operator versions in that index for both operators, for the cluster group operator as well as for the hardware event processor and we are going to corroborate that those are the same that we have defined before during the index creation Let's now proceed to create the configuration file of also a mirror plugin which is called the image set config Here we are going to define in a declarative way not only the disconnected registry under the image URL key but also we are going to be mirroring the operators from the catalog previously created the two packages defined as well as the corresponding channels Once that we got the image set configuration in place it's very easy to store the mirroring task A nice recommendation here is we can have this task run periodically as a cron job and that way we can have our disconnected local registry always updated As an outcome of this operation the OC mirror plugin will write for us the catalog source and the image content source policy YAML file In this scenario, the catalog source is a collection of the run operators metadata The OLM will use this catalog source to build the list available of run operators in the console Let's take a look to the generated catalog source YAML file and apply that object to the disconnected cluster Let's check in the bottom left part of the terminal the available catalogs in our disconnected cluster Let's also inspect the image content source policy Once this resource is updated into the disconnected cluster all nodes from master to workers are going to be updated and rebooted This process is automatically handled by the machine config operator and it could take some minutes Let me speed that for you Once the image content source policy object is updated across all the cluster nodes Let's now take a look to the new catalog sources and run operators from the OpenChiff console Before installing the run operators into the disconnected cluster let's first take a look to the current state of installed operators Once everything is set installing the run operators is as simple as creating the corresponding OLM objects In order to deploy the cluster crop upgrades operator we just need to create the operator's subscription For the case of the hardware event ProxyOperator since it is a namespaced operator we need to create a namespace, the operator group and its corresponding subscription as well Once these operator resources are created let's just watch the run operators deployment from console and this is how upstream run operators can be deployed in a disconnected environment for 5G telco deployments Thanks for watching