Collecting 1C:Enterprise Linux cluster metrics

Follow this guide to configure collection of 1C:Enterprise Linux cluster metrics in Prometheus format and their visualization in Yandex Monitoring. In addition to visualization, Monitoring allows you to set up alerts, create hybrid dashboards with metrics from various Yandex Cloud services and offers extensive scaling capabilities.

Cluster metrics will be collected using the freely distributed Prometheus 1C Exporter utility, transferred to Yandex Unified Agent, and then to Monitoring.

To use this tutorial, you need to have a 1C:Enterprise cluster deployed on a Yandex Compute Cloud VM. For a deployment example, see Creating a cluster of 1C:Enterprise Linux servers with a Managed Service for PostgreSQL cluster.

To get started with metrics in Yandex Monitoring:

1. Install and configure Prometheus 1C Exporter.
2. Install and configure Unified Agent.
3. Create a service account and link it to your VM.
4. Make sure Monitoring receives the metrics.
5. Create a dashboard to monitor metrics.

If you no longer need the resources you created, delete them.

Required paid resourcesRequired paid resources

Fee for sending metrics to Monitoring (see Monitoring pricing).

Install and configure Prometheus 1C ExporterInstall and configure Prometheus 1C Exporter

1. Download Prometheus 1C Exporter to the 1C:Enterprise VM:

   wget https://github.com/LazarenkoA/prometheus_1C_exporter/releases/download/v1.5.0/prometheus_1C_exporter-linux-amd64
   

2. Rename the exporter and move it to the system folder:

   mv prometheus_1C_exporter-linux-amd64 exporter_1C
   sudo mv exporter_1C /usr/local/bin/exporter_1C
   

3. Make the file executable:

   sudo chmod +x /usr/local/bin/exporter_1C
   

4. Create a configuration file for the exporter:

   sudo mkdir /etc/exporter_1C/
   sudo vi /etc/exporter_1C/settings.yaml
   

5. Add the configuration below to the settings.yaml file and, if necessary, customize the list of metrics to collect.

   Exporter’s configuration in settings.yaml

   # Metric values you can collect using the exporter:
   # ClientLic: Client licenses.
   # AvailablePerformance: Available performance (via RAC).
   # ScheduleJob: Checking the `Block scheduled jobs` option.
   # Session: Sessions.
   # Connect: Connections.
   # SessionsData: Various indicators from the 1C console (via RAC).
   # ProcData: Process data (collects Linux OS metrics).
   # CPU: Processor utilization.
   # disk: Disk metrics (currently only Linux and WeightedIO are supported).
   # Property: Additional parameters specific to each metric.
   # Processes: Additional ProcData parameter which defines the list of processes to monitor.
   
   # You can enable or disable any set of metrics for monitoring.
   Exporters:
   - Name: client_lic
   - Name: available_performance
   # - Name: processes
   # - Name: cpu
   # - Name: disk
   # - Name: shedule_job
   # - Name: session
   - Name: connect
   # - Name: sessions_data
   
   
   # HTTP service that returns a JSON array with database access parameters.
   # Required for the ScheduleJob metric.
   # JSON example:
   #[
   #  {
   #    "Name": "hrmcorp-n17", # 1C database name
   #    "UserName": "",
   #    "UserPass": ""
   #  }
   #]
   #DBCredentials: # This is a required parameter for the shedule_job metric.
   #  Address used by 1C to return a list of databases in JSON format:
   #  URL: http://ca-fr-web-1/fresh/int/sm/hs/PTG_SysExchange/GetDatabase
   #  User: ""
   #  Password: ""
   #  TLSSkipVerify: true # true: Ignore certificate verification errors when accessing the service.
   
   RAC:
   Path: "/usr/local/bin/rac"
   Port: "1545"      # This is an optional parameter.
   Host: "localhost" # This is an optional parameter.
   Login: ""         # This is an optional parameter.
   Pass: ""          # This is an optional parameter.
   
   LogDir:        # If the folder is not specified, logs will be written to the folder with the exporter’s executable file.
   LogLevel:  5   # Logging level: 2 (error), 3 (warning), 4 (information), 5 (debugging).
   

   This configuration enables sending the following metrics:

   • available_performance: Host’s available performance (Histogram).
   • client_lic: 1C client licenses (Histogram).
   • connect: 1C connections (Histogram).

   You can add other metrics if required.

6. Create a symbolic link for the 1C RAC management utility:

   sudo ln -s /opt/<1C_installation_path>/rac /usr/local/bin/rac
   

7. Create a unit file for systemd to manage the exporter:

   sudo vi /etc/systemd/system/exporter_1C.service
   

   Contents of the exporter_1C.service file

   [Unit]
   Description=Prometheus 1C exporter
   After=local-fs.target network-online.target network.target
   Wants=local-fs.target network-online.target network.target
   
   [Service]
   Type=simple
   ExecStart=/usr/local/bin/exporter_1C -port=9095 --settings=/etc/exporter_1C/settings.yaml
   ExecReload=/bin/kill -s HUP $MAINPID
   
   [Install]
   WantedBy=multi-user.target
   

   This configuration starts the exporter after the OS and network boot up. The exporter starts listening to port 9095 according to the parameters in settings.yaml.

8. Start the exporter:

   sudo systemctl daemon-reload
   sudo systemctl start exporter_1C.service
   sudo systemctl enable exporter_1C.service
   

9. Check the exporter status and make sure the service is active (running).

   sudo systemctl status exporter_1C.service
   

After you start the exporter, metrics will be available at curl http://localhost:9095/metrics. To work with metrics in Yandex Monitoring, you need to install Unified Agent and configure it to receive metrics from this address and send them to Monitoring.

Create a service account and link it to your VMCreate a service account and link it to your VM

You need a service account to send metrics from the VM to Monitoring.

1. Create a service account named sa-1c-monitoring in the folder you want to write metrics to.
2. Assign the monitoring.editor role to the service account.
3. Link the service account to the VM where 1C:Enterprise is installed.

Install and configure Unified AgentInstall and configure Unified Agent

1. Download the latest deb package:

   ubuntu_name="<OS_version>" ua_version=$(curl --silent https://storage.yandexcloud.net/yc-unified-agent/latest-version) bash -c 'curl --silent --remote-name https://storage.yandexcloud.net/yc-unified-agent/releases/${ua_version}/deb/${ubuntu_name}/yandex-unified-agent_${ua_version}_amd64.deb'
   

2. Check the deb package version using the ls command.

3. Install Unified Agent from the deb package by specifying its version:

   sudo dpkg -i yandex-unified-agent_<agent_version>_amd64.deb
   

   You can find other installation methods in Installing and updating Yandex Unified Agent.

4. Check that Unified Agent is running:

   sudo systemctl status unified-agent.service
   

5. Open the Unified Agent configuration file:

   sudo vi /etc/yandex/unified_agent/config.yml
   

6. Replace the file contents with the configuration below. Add the ID of the folder you want to send metrics to.

   Unified Agent configuration in config.yml

   monitoring:
     port: 16300
   
   status:
   port: 16301
   
   channels:
   - name: yc_metrics_channel
      channel:
         output:
         plugin: yc_metrics
         config:
            folder_id: "<folder_ID>"
            iam:
               cloud_meta: {}
   - name: monitoring_1c
      channel:
         output:
         plugin: yc_metrics
         config:
            folder_id: "<folder_ID>"
            iam:
               cloud_meta: {}
   
   routes:
   - input:
         id: linux_metrics_input
         plugin: linux_metrics
         config:
         poll_period: 15s
      channel:
         channel_ref:
         name: yc_metrics_channel
   - input:
         plugin: metrics_pull
         config:
         url: http://localhost:9095/metrics
         format:
            prometheus: {}
         namespace: app1c
      channel:
         pipe:
         - filter:
               plugin: transform_metric_label
               config:
               label: cluster
               rename_to: cluster_1c
         - filter:
               plugin: transform_metric_labels
               config:
               labels:
                  - servicename: "-"
         - filter:
               plugin: transform_metric_labels
               config:
               labels:
                  - user: "-"
         - filter:
               plugin: transform_metric_labels
               config:
               labels:
                  - base: "-"
         - filter:
               plugin: transform_metric_labels
               config:
               labels:
                  - licSRV: "-"
         channel_ref:
         name: monitoring_1c
   
   import:
   - /etc/yandex/unified_agent/conf.d/*.yml
   

   This configuration sets the app1c_ prefix for metrics from the /metrics endpoint. You can use any prefix.

   Yandex Monitoring has restrictions on labels and their values:

   • Label name must not be empty.
   • Label name and metric value may only contain Latin letters.
   • You cannot use the cluster label.

   To comply with these restrictions, the Unified Agent configuration file instructs to:

   • Rename the cluster label to cluster_1c.
   • Filter (remove) labels containing non-Latin characters: servicename, user, base, and licSRV.

   To learn more about configuring Unified Agent, see Configuration.

7. Make sure the configuration file is correct. The command should output the contents of the file:

   unified_agent check-config -c /etc/yandex/unified_agent/config.yml
   

8. Restart Unified Agent to apply the updated configuration:

   sudo systemctl restart unified-agent.service
   

9. Check the Unified Agent status:

   sudo systemctl status unified-agent.service
   

Make sure Monitoring receives the metricsMake sure Monitoring receives the metrics

Create a dashboard to monitor metricsCreate a dashboard to monitor metrics

Below is an example of a dashboard for monitoring metrics provided in the exporter configuration.

For additional dashboard settings, see Adding a widget to a dashboard.

Delete the resources you createdDelete the resources you created

Stop reading and writing metrics by doing one of the following:

• Stop Unified Agent.
• In the Unified Agent configuration file, remove the settings for collecting and sending metrics.
• Delete Unified Agent.