Manage the Platform#
The Unryo platform requires minimum administration. As an administrator, you have a web-based console for managing your Unryo platform, from where you see the platform health and deploy new components.
The console is multi-tenant: you can view the deployment status for all your Unryo stacks if you have several deployments.
Managing the Unryo platform is mainly for On-Prem customers. For SaaS deployments, the Unryo platform is hosted and self-maintained, and administrators focus on collector(s) deployment and customization.
Unryo Connect Console#
To access the Unryo Connect Console, go to https://portal.unryo.com
This console provides a central place for administrators to monitor platform health, check status, manage tokens and deploy new components.
With the Unryo console, administrators can:
Other typical configuration tasks include:
Check Platform Health#
The Overview Page#
Overview page to display all your platform components with their status (online, offline), their last seen time, as well as availability and performance information.
Statuscan be Online or Offline. Online means that the component is up and running. Offline means that it is not connected to the rest of the Unryo platform.
Availabilityis the most important availability issue message of a node.
Performanceis the most important performance issue message of a node.
For more details you can reach the Component Health section.
From this page, administrators can:
- Use the
Searchto filter the list or find a component.
- Access the
Quick Start / Start Heretutorial to get started with Unryo.
- Launch the
Wizardto deploy collectors or other Unryo components.
- Access the
Integrationspage to configure data collection.
- Access the
Unryo Portal(s)by clicking on the URL.
- Manage the
Accountby clicking on your login name top-right corner.
The Topology Map#
Use the topology map to display all your Unryo platform with component interconnections.
From this page, administrators can:
Highlighta component to display its properties.
- Use the
namespacedropbox to show components per configured environment.
Note: If the node is
offline, the icon will be displayed with a grey color.
|Name||visual name of that component. It doesn’t have to be unique, but it is a name which will help users identify this component visually, for example, when presented in a component inventory list.|
|Version||version of the image the component is running|
|Namespace||indicates the environment where the components belong. It is used to create a network of services, and can be used to implement multitenancy.|
|Status||indicates if the node is online or offline. If the node is offline, the icon will be greyed out.|
|Availability||current availability status for the component|
|Performance||current performance status for the component|
|Role||indicates the role of the component. That information will be used to build the relationships and interconnections between components. The protocol supports four roles: producer, supplier, receiver, consumer. A producer produces data, and pushes this data to an endpoint, typically a receiver. A supplier makes data available to other components, typically consumers.|
|Type||type of the node. For example: Collector, Report, Database, Analytics, ...|
|External IP Address||external IP address of the component|
|Provider||the service provider hosting the component|
|Protocol||protocol(s) used within the connections|
|Last Ping||shows when the node last contacted the Unryo Registry. After a period of inactivity, the component is unregistered from the Unryo system.|
The Geo-Map View#
Use the geo-map to display your components with their geo-location, interconnections and health information.
Deploy Platform and Collectors#
Unryo supports Docker, vApp and Kubernetes deployment methods.
From the Unryo Cloud Management Console, click
+DEPLOY from the top banner. From there's an easy wizard that will guide you to pick a deployment method.
Unryo comes with an All-In-One easy setup that install the complete platform on a single host machine in your data-center.
Three possible options are:
vApp: Follow the procedure here
Docker: Follow the procedure here
Kubernetes: Follow the procedure here
Add a Collector#
Collectors gather data from your infrastructure and ship data to your Unryo platform. You can install as many as collectors you want.
To deploy a new collector, follow the procedure here:
- Collector deployed as a Docker Container
- Collector deployed on Kubernetes
- Collector deployed as a vApp
Tokens are unique identifier associated with your organization. For your protection, treat them as you would a password. Never share tokens with other users or applications.
From this panel, administrators can:
Listtokens with their properties.
Revokean active token.
Enablea revoked token.
Create your own monitoring checks#
From the console, you see the status and heath for all deployed components. By default, Unryo monitors components status (online/offline), availability and performance.
You can extend this monitoring, and add your own checks.
Learn more here.
Operate the Platform#
Start/Stop/Deregister a Component#
This guide assumes that you are comfortable working Linux shell, Docker and Kubernetes.
Unryo components are all containers.
if you are using
Kubernetes, you can use the K8S web management portal.
If you are using docker, you can use the following commands:
To stop a component:
docker stop <container-id|container-name>
To start a component:
docker start <container-id|container-name>
To remove a component:
docker rm <container-id|container-name>
- Removing a component, especially a core component (database, portal, analytic) can impact your production environment. You might also want to backup custom configurations before removing a component.
- If you do not have access anymore to a component you want to remove, you could revoke the token used by this component from the "TOKEN" menu, but WARNING, every component installed using this token will then be unconfigured and disconnected.
- If a collector is removed, data already collected will not be affected and will always remain in the Unryo databases.
To move a component:
- Moving a database component requires planning and storage container expertise. We recommend avoiding moving databases if you do not have proper knowledge of your persistent container storage configuration.
- For the reporting portal component, you can start a new instance in the new site/cloud in parallel to the original portal. You can then validate that network connectivity is working and put in place your custom reports if any. When everything is validated, you can then delete the previous portal container.
- For a collector component, you can proceed the same way. Install on your new site/cloud a new collector in parallel, copy your personalized collect configuration files. At this point you can validate the logs of that new container and make sure the data you intended to collect is being collected. When everything is validated, you can then delete the previous collector.
Change portal logo#
Go in the portal container, and copy your logo in
Unryo is capable of using LDAP servers for authentication. Find out more here.
Note: upgrading the Unryo platform is mainly for On-Prem customers. For SaaS deployments, the Unryo platform is hosted and self-maintained, and administrators only have to upgrade their Unryo collector(s) and, if used, their Unryo agents.
All Unryo components are containers, and all your configuration files and DBs are stored in persistent container
volumes or centrally in the
Unryo Vault. They therefore persist container upgrades.
To update manually Unryo:
Log on your host
Pull the update images:
bash docker-compose pull
- Then rerun your components with the new images
bash docker-compose up -d
You must be in the same directory as the
docker-compose.ymlfile you used to deploy your nodes. If your file is not called
docker-compose.yml, you must modify the above commands slightly; for example
docker-compose -f your-yml-file.yml pull
To update automatically Unryo:
- Use crontab:
bash 0 3 1 */1 * docker-compose pull && docker-compose up -d
If you are using
Kubernetes, you can do this from the K8S management portal.
Back up & restore#
All Unryo-Monitor components are containers, and all your configurations and DBs are stored in persistent container volumes. They therefore persist container upgrades.
On top of this, all the configurations are centrally stored in the Unryo Vault, which prevents loss in case of a component's failure or error.
Information about important configuration files:
|Metric Collector||All configurations are stored encrypted in the Unryo vault. They are copied locally in the collector at
|Analytics Engine||All configurations are stored encrypted in the Unryo vault. They are copied locally in the analytics engine at
You can backup the configuration files stored in the vault in two steps. First,
list all your configuration files using the
endpoint. Then, iterate through your files, calling the
/vault/api/v1/configurations/$configId endpoint for each file to retrieve said
Here are the best practices to quickly identify potential problems with the Unryo platform.
Component connectivity issue#
Login to the Unryo Admin UI and go to the topology view to validate that all components are correctly connected and in good health.
Validate component logs using the Unryo
You can also use your container management tools. For Docker, use:
docker logs <container-id|container-name>
To log in the container:
bash docker exec -ti <container_id> bash
To copy file from/to the container:
bash docker cp <container_id>:/<file> . docker cp <file> <container_id>:<target path>
To troubleshoot a metric data collection issue, you can use the --test command:
docker exec <container_id> /usr/bin/telegraf --test --config /etc/telegraf/telegraf.d/<config_to_troubleshoot>.conf
- If your module is NOT connected, validate your cloud infrastructure: is your host healthy? are the docker services running? If you see the container running, consult the container logs to identify internal problems.
Database High Cardinality#
If you have high cardinality issues, you can fine-tune metric data retention and configure automatic data deletion rules.
- The data retention is set on a global basis, meaning it applies to all metrics.
- The automatic data deletion can be customized per resource type. For example, high producers of metrics, such as Kubernetes Pods or vSphere VMs, can be configured separately so their metric data and event data is deleted after a certain period of inactivity.
To configure it, go on the Data Retention & Cleanup section.