Monitoring
The monitoring solution consists of the following components:
- Kibana with system performance dashboards
- Elasticsearch for storing metrics data
- Logstash for aggregation and forwarding metrics data
Access monitoring dashboards
To browse monitoring dashboards, follow the steps below:
Go to Kibana, for example, at https://app_hostname/kibana/.
Log in to Kibana as the Admin role user assigned proper access permissions in the WorkFusion's User Management (Keycloak). For more information on the role and permissions, refer to the Manage and assign realm and client roles.
In Kibana, on the Welcome page, you can see available preconfigured dashboards. To open the full list of preconfigured dashboards, click Dashboard.
Browse dashboards
Disk usage
The dashboard contains metrics on disk usage percentage and disk size in megabytes.
You can see disk usage details for each component, such as S3, Nexus, and others.
Environment status
The dashboard displays the statuses of the services in the Work.AI platform environment:
The TCP monitors status and HTTP monitors status diagrams display the overall status:
For details, see the Heartbeat Services, HTTP up status, and HTTP ping times diagrams:
Healthcheck
The compound dashboard provides information on the status of components and components' versions.
The dashboard also includes information on the status of the OCR component, warnings about expiring licenses, and the number of licenses used to process pages.
Host overview
The dashboard contains detailed time-series information of the host:
System overview
The dashboard contains a summary of all configured hosts:
Click a host to open the Host overview dashboard with detailed information on the respective instance.
RPA overview
The dashboard contains data on the monitoring of CPU and memory usage for each RPA server.
At the top of the page, select a needed RPA server hostname:
The available dashboards are as follows:
The RPA servers uptime (Server/master-bot-agent, Workers) dashboard displays the uptime percentage per bot and the service state during a specific period based on the data received from Heartbeat.
The RPA resources usage (Memory/CPU usage) dashboard displays memory and CPU usage per each Java process on the RPA server.
Performance dashboards
You can use Kibana dashboards to localize issues and monitor performance across the Work.AI components. For details, refer to Performance dashboards | Overview.
Create custom dashboards, visualizations, and alerts
If you need to create customized dashboards, visualizations, or alerts based on default ones, do not edit existing objects directly. Instead, make a copy of an existing object and customize it. Otherwise, you may lose all your changes in case of the product upgrade.
Reporting
You can generate a report that contains a Kibana dashboard, visualization, saved search, or Canvas workpad. Depending on the object type, you can export the data as a PDF, PNG, or CSV document, which you can keep for yourself or share with others.
Reporting is available from the Share menu in Discover, Dashboard, Visualize Library, and Canvas.
Manually generate and download reports
You can manually generate and download PDF, PNG, and CSV files of dashboards, visualizations, Canvas workpads, and saved searches. For that, follow the steps below:
Open a dashboard, visualization, Canvas workpad, or saved search.
On the Kibana toolbar, click Share and select one of the following options:
PDF Reports: generates a PDF file of the dashboard, visualization, or Canvas workpad
PNG Reports: generates a PNG file of the dashboard or visualization
CSV Reports: generates a CSV report of the saved search
Generate the report. When the generation is completed, a notification appears.
Click Download report.
In some cases, you can see a similar error:
To fix it, ensure that the following packages are installed:
- ipa-gothic-fonts
- xorg-x11-fonts-100dpi
- xorg-x11-fonts-75dpi
- xorg-x11-utils
- xorg-x11-fonts-cyrillic
- xorg-x11-fonts-Type1
- xorg-x11-fonts-misc
- fontconfig
- freetype
Change monitored directories
Logs can be of several logging level: INFO > WARNING > DEBUG. Most services emit logs of the INFO type to save space and increase the performance of the ELK stack. Increasing the logging level may consume too much RAM and affect the final cost of the system. Thus, tracking additional directories or increasing the current logging level of services is not recommended.
ELK monitors log files of processes in the /logs directories, specific for each service. To edit or add a new directory, change the Ansible roles. If such a need arises, create a Service Desk ticket for changing the directories list.
For more information on using Kibana dashboards for platform monitoring, read the Performance dashboards guides.
For instructions to enable read-only access to Kibana, read the guide.