Choosing between Platform Manager, REdge, and RLSI
The SC//Reliant™ platform gives you a central interface and a local one, and they are not alternatives to each other. The SC//Reliant Platform Manager is where you define how your estate should be configured. A local node console, either REdge or the SC//Reliant Local Status Indicator (RLSI), is where you see what one node is doing right now, and act on that node immediately.
REdge is the newer local console and is not yet generally available. Once it is released, it is installed automatically with the SC//Reliant Virtual Controller (RVC) and is the recommended local console. The tables on this page describe RLSI. For the actions REdge adds, see Where REdge fits.
This page describes which interface to use for a given task.
The short version
Configuration goes in Platform Manager. Immediate local action goes in RLSI.
| SC//Reliant Platform Manager | SC//Reliant Local Status Indicator | |
|---|---|---|
| Scope | Your whole estate — sites, node groups, and nodes. | One node, the one you are connected to. |
| What it holds | The configuration your nodes are supposed to have. | What that node is actually doing, right now. |
| When changes apply | On the node's next Puppet run. | Immediately. |
| Does it persist? | Yes. It is the source of truth. | No. RLSI performs actions; it does not store configuration. |
| Reached from | Anywhere, at platform.reliant.io. | A network that can reach the node. |
There is a third path. Actions that have to run on a node — rebooting a VM, running Puppet, turning maintenance mode on — reach it through Remote Job Execution, which you call with the SC//Reliant API, and the same actions are available on the node itself through the SC//Reliant Virtual Controller command line. The matrix below names the API where that is the documented remote route. See Remote Job Execution.
Action matrix
Configuration
Everything in this group is Platform Manager only. RLSI has no configuration screens at all — even the properties that control RLSI itself live in Platform Manager.
| Action | Where | Notes |
|---|---|---|
| Change a node or node group property | Platform Manager | Applied by Puppet on the next run. |
| Configure which RLSI tabs and checks appear | Platform Manager | See Configuring RLSI. |
| Set the package version a node runs | Platform Manager | Puppet holds the node at the version you set. |
| Define the store's network configuration | Platform Manager | This is the intended configuration. RLSI shows what the node actually has. |
| Configure monitoring checks and handlers | Platform Manager | See Monitoring your environment. |
| Schedule automatic VM backups | Platform Manager | RLSI takes backups on demand but does not schedule them. |
| Add or manage users, permissions, and shell access | Platform Manager | |
| Add a managed file or script to a node | Platform Manager | See Using the Managed Files Utility. |
IMPORTANT: Configuration files on a node are managed by Puppet, so a change made directly on the node is overwritten on the next run. If a change needs to last, make it in Platform Manager.
Virtual machines
| Action | Where | Notes |
|---|---|---|
| See which VMs are on a node and whether they are running | RLSI, or the API | RLSI is quicker if you only care about one node. |
| Reboot a VM | RLSI, or the API | RLSI for one VM on one node. The API to reboot the same VM across many nodes at once. See Rebooting a virtual machine. |
| Start or shut down a VM | RLSI | |
| Force a VM off or reset it | RLSI | Data loss is possible. See Managing virtual machines in RLSI. |
| Take a backup of a VM now | RLSI | A full backup shuts the VM down for the duration, and puts the cluster into maintenance mode first. |
| Restore a VM from a backup | RLSI | |
| Define a VM's specification, disks, or network | Platform Manager | RLSI can act on a VM that exists, but cannot create or reconfigure one. |
| Build a new VM on a node | Platform Manager | Driven by node properties. See Controlling whether a VM is built on a node. |
Node status and operations
| Action | Where | Notes |
|---|---|---|
| Check whether a node is up, and its cluster role | Either | Platform Manager shows the whole estate. RLSI answers for one node without a login to the central tools. |
| Read a node's local checks — connectivity, disk usage, addresses | RLSI | See Reading the Status tab in RLSI. |
| Check whether the cluster is in maintenance mode or unclean, and whether the node is in standby | RLSI | Shown when you expand the role indicator. Maintenance mode and unclean describe the cluster; standby describes this node. RLSI reports all three but changes none of them. |
| Reboot the node itself | RLSI | Planned maintenance only. Prepare the cluster first — see Performing maintenance on a node. |
| Reboot the payment container | RLSI | |
| Run Puppet on a node now | RLSI, or the API | RLSI has a Run Puppet button for the node in front of you. The API targets many nodes at once. See Deploying changes to your environment. |
| Turn maintenance mode on or off | The API, or on the node | Not available in RLSI, which only reports the state. See Operating the cluster. |
| Put a node into standby | On the node | Set with the cluster tools rather than from either interface. See Operating the cluster. |
| Register a new node | Platform Manager | See Adding a node. Provisioning itself is zero-touch. |
| Track an install in progress at a site | Platform Manager | See Using the Edge Install Monitor. |
Diagnostics
| Action | Where | Notes |
|---|---|---|
| Ping or traceroute from the node | RLSI | Tells you what the node can reach, which is often different from what your own computer can reach. |
| See the node's actual network interfaces and addresses | RLSI | Compare against the intended configuration in Platform Manager. |
| Read a node's logs | RLSI, or your log destination | RLSI has a LOGS tab for the node in front of you. Logs are also forwarded to the destination you configure. See Managing logs. |
| Review who changed a setting, and when | Platform Manager | Change history is a Platform Manager feature; RLSI has no equivalent. See Using the audit trail. |
| See monitoring alerts across the estate | Platform Manager | RLSI reports only the node in front of you. |
Where REdge fits
Everything this page says about the role of RLSI also applies to REdge. REdge acts only on the node that serves it, and it holds no configuration of its own, so configuration still belongs in Platform Manager. What changes is how much you can do locally. REdge is a front end for RVC, and it adds the following actions that RLSI does not offer.
| Action | Where in REdge |
|---|---|
| Start or stop a container, and pull container images | See Managing containers in REdge. |
| Preview or apply the Puppet configuration, stop a run, and enable or disable Puppet | See Running and controlling Puppet in REdge. |
| Pause cluster resource management for planned work, and move cluster resources off the node | See Performing cluster maintenance in REdge. |
| Run the scripts that Puppet maintains on the node | See Running managed scripts in REdge. |
| Check whether a URL is reachable from the node | See Running network diagnostics in REdge. |
| Give each person their own local account, with roles and two-factor authentication | See Managing REdge users and roles. |
For acting on more than one node, the guidance below is the same for REdge as for RLSI.
Why configuration is never done in RLSI
RLSI is a window onto one node and a set of buttons that act on it. It deliberately holds no configuration of its own: the tabs it shows, the checks it reports, and the version it runs are all properties in Platform Manager, applied by Puppet.
This is what makes an estate predictable. A node that is rebuilt or replaced comes back with the configuration Platform Manager holds for it, and nothing that was only ever set locally survives.
Acting on more than one node
RLSI acts on one node at a time, because it is served by that node. When the same action is needed across a node group or a whole site, use the SC//Reliant API, which runs the action as a job against every target you select and reports the result per node.
This is the practical dividing line for support work: one store, use RLSI; a pattern across many stores, use the API. See Remote Job Execution.
Related documentation
- Using the local node consoles: REdge and RLSI.
- Using REdge: the newer local console and what each page does.
- Using the SC//Reliant Local Status Indicator (RLSI): reaching the console and what each tab does.
- SC//Reliant Local Status Indicator (RLSI): where RLSI fits in the platform.
- Understanding configuration data: how properties resolve across sites, node groups, and nodes.
- Deploying changes to your environment: applying a Platform Manager change to your nodes.