Pandas icon

Pandas

Pandas

Plugin: python.d.plugin Module: pandas

Overview

Pandas is a de-facto standard in reading and processing most types of structured data in Python. If you have metrics appearing in a CSV, JSON, XML, HTML, or other supported format, either locally or via some HTTP endpoint, you can easily ingest and present those metrics in Netdata, by leveraging the Pandas collector.

This collector can be used to collect pretty much anything that can be read by Pandas, and then processed by Pandas.

The collector uses pandas to pull data and do pandas-based preprocessing, before feeding to Netdata.

This collector is supported on all platforms.

This collector supports collecting metrics from multiple instances of this integration, including remote instances.

Default Behavior

Auto-Detection

This integration doesn’t support auto-detection.

Limits

The default configuration for this integration does not impose any limits on data collection.

Performance Impact

The default configuration for this integration is not expected to impose a significant performance impact on the system.

Setup

Prerequisites

Python Requirements

This collector depends on some Python (Python 3 only) packages that can usually be installed via pip or pip3.

sudo pip install pandas requests

Note: If you would like to use pandas.read_sql to query a database, you will need to install the below packages as well.

sudo pip install 'sqlalchemy<2.0' psycopg2-binary

Configuration

File

The configuration file name for this integration is python.d/pandas.conf.

You can edit the configuration file using the edit-config script from the Netdata config directory.

cd /etc/netdata 2>/dev/null || cd /opt/netdata/etc/netdata
sudo ./edit-config python.d/pandas.conf

Options

There are 2 sections:

  • Global variables
  • One or more JOBS that can define multiple different instances to monitor.

The following options can be defined globally: priority, penalty, autodetection_retry, update_every, but can also be defined per JOB to override the global values.

Additionally, the following collapsed table contains all the options that can be configured inside a JOB definition.

Every configuration JOB starts with a job_name value which will appear in the dashboard, unless a name parameter is specified.

Name Description Default Required
chart_configs an array of chart configuration dictionaries [] yes
chart_configs.name name of the chart to be displayed in the dashboard. None yes
chart_configs.title title of the chart to be displayed in the dashboard. None yes
chart_configs.family family of the chart to be displayed in the dashboard. None yes
chart_configs.context context of the chart to be displayed in the dashboard. None yes
chart_configs.type the type of the chart to be displayed in the dashboard. None yes
chart_configs.units the units of the chart to be displayed in the dashboard. None yes
chart_configs.df_steps a series of pandas operations (one per line) that each returns a dataframe. None yes
update_every Sets the default data collection frequency. 5 no
priority Controls the order of charts at the netdata dashboard. 60000 no
autodetection_retry Sets the job re-check interval in seconds. 0 no
penalty Indicates whether to apply penalty to update_every in case of failures. yes no
name Job name. This value will overwrite the job_name value. JOBS with the same name are mutually exclusive. Only one of them will be allowed running at any time. This allows autodetection to try several alternatives and pick the one that works. no

Examples

Temperature API Example

example pulling some hourly temperature data, a chart for today forecast (mean,min,max) and another chart for current.

temperature:
    name: "temperature"
    update_every: 5
    chart_configs:
      - name: "temperature_forecast_by_city"
        title: "Temperature By City - Today Forecast"
        family: "temperature.today"
        context: "pandas.temperature"
        type: "line"
        units: "Celsius"
        df_steps: >
          pd.DataFrame.from_dict(
            {city: requests.get(f'https://api.open-meteo.com/v1/forecast?latitude={lat}&longitude={lng}&hourly=temperature_2m').json()['hourly']['temperature_2m']
            for (city,lat,lng)
            in [
                ('dublin', 53.3441, -6.2675),
                ('athens', 37.9792, 23.7166),
                ('london', 51.5002, -0.1262),
                ('berlin', 52.5235, 13.4115),
                ('paris', 48.8567, 2.3510),
                ('madrid', 40.4167, -3.7033),
                ('new_york', 40.71, -74.01),
                ('los_angeles', 34.05, -118.24),
                ]
            }
            );
          df.describe();                                               # get aggregate stats for each city;
          df.transpose()[['mean', 'max', 'min']].reset_index();        # just take mean, min, max;
          df.rename(columns={'index':'city'});                         # some column renaming;
          df.pivot(columns='city').mean().to_frame().reset_index();    # force to be one row per city;
          df.rename(columns={0:'degrees'});                            # some column renaming;
          pd.concat([df, df['city']+'_'+df['level_0']], axis=1);       # add new column combining city and summary measurement label;
          df.rename(columns={0:'measurement'});                        # some column renaming;
          df[['measurement', 'degrees']].set_index('measurement');     # just take two columns we want;
          df.sort_index();                                             # sort by city name;
          df.transpose();                                              # transpose so its just one wide row;          
      - name: "temperature_current_by_city"
        title: "Temperature By City - Current"
        family: "temperature.current"
        context: "pandas.temperature"
        type: "line"
        units: "Celsius"
        df_steps: >
          pd.DataFrame.from_dict(
              {city: requests.get(f'https://api.open-meteo.com/v1/forecast?latitude={lat}&longitude={lng}&current_weather=true').json()['current_weather']
              for (city,lat,lng)
              in [
                  ('dublin', 53.3441, -6.2675),
                  ('athens', 37.9792, 23.7166),
                  ('london', 51.5002, -0.1262),
                  ('berlin', 52.5235, 13.4115),
                  ('paris', 48.8567, 2.3510),
                  ('madrid', 40.4167, -3.7033),
                  ('new_york', 40.71, -74.01),
                  ('los_angeles', 34.05, -118.24),
                  ]
              }
              );
          df.transpose();
          df[['temperature']];
          df.transpose();          

API CSV Example

example showing a read_csv from a url and some light pandas data wrangling.

example_csv:
    name: "example_csv"
    update_every: 2
    chart_configs:
      - name: "london_system_cpu"
        title: "London System CPU - Ratios"
        family: "london_system_cpu"
        context: "pandas"
        type: "line"
        units: "n"
        df_steps: >
          pd.read_csv('https://london.my-netdata.io/api/v1/data?chart=system.cpu&format=csv&after=-60', storage_options={'User-Agent': 'netdata'});
          df.drop('time', axis=1);
          df.mean().to_frame().transpose();
          df.apply(lambda row: (row.user / row.system), axis = 1).to_frame();
          df.rename(columns={0:'average_user_system_ratio'});
          df*100;          

API JSON Example

example showing a read_json from a url and some light pandas data wrangling.

example_json:
    name: "example_json"
    update_every: 2
    chart_configs:
      - name: "london_system_net"
        title: "London System Net - Total Bandwidth"
        family: "london_system_net"
        context: "pandas"
        type: "area"
        units: "kilobits/s"
        df_steps: >
          pd.DataFrame(requests.get('https://london.my-netdata.io/api/v1/data?chart=system.net&format=json&after=-1').json()['data'], columns=requests.get('https://london.my-netdata.io/api/v1/data?chart=system.net&format=json&after=-1').json()['labels']);
          df.drop('time', axis=1);
          abs(df);
          df.sum(axis=1).to_frame();
          df.rename(columns={0:'total_bandwidth'});          

XML Example

example showing a read_xml from a url and some light pandas data wrangling.

example_xml:
    name: "example_xml"
    update_every: 2
    line_sep: "|"
    chart_configs:
      - name: "temperature_forcast"
        title: "Temperature Forecast"
        family: "temp"
        context: "pandas.temp"
        type: "line"
        units: "celsius"
        df_steps: >
          pd.read_xml('http://metwdb-openaccess.ichec.ie/metno-wdb2ts/locationforecast?lat=54.7210798611;long=-8.7237392806', xpath='./product/time[1]/location/temperature', parser='etree')|
          df.rename(columns={'value': 'dublin'})|
          df[['dublin']]|          

SQL Example

example showing a read_sql from a postgres database using sqlalchemy.

sql:
    name: "sql"
    update_every: 5
    chart_configs:
      - name: "sql"
        title: "SQL Example"
        family: "sql.example"
        context: "example"
        type: "line"
        units: "percent"
        df_steps: >
          pd.read_sql_query(
            sql='\
                select \
                    random()*100 as metric_1, \
                    random()*100 as metric_2 \
              ',
            con=create_engine('postgresql://localhost/postgres?user=netdata&password=netdata')
            );          

Metrics

Metrics grouped by scope.

The scope defines the instance that the metric belongs to. An instance is uniquely identified by a set of labels.

This collector is expecting one row in the final pandas DataFrame. It is that first row that will be taken as the most recent values for each dimension on each chart using (df.to_dict(orient='records')[0]). See pd.to_dict()."

Per Pandas instance

These metrics refer to the entire monitored application.

This scope has no labels.

Metrics:

Metric Dimensions Unit

Alerts

There are no alerts configured by default for this integration.

Troubleshooting

Debug Mode

To troubleshoot issues with the pandas collector, run the python.d.plugin with the debug option enabled. The output should give you clues as to why the collector isn’t working.

  • Navigate to the plugins.d directory, usually at /usr/libexec/netdata/plugins.d/. If that’s not the case on your system, open netdata.conf and look for the plugins setting under [directories].

    cd /usr/libexec/netdata/plugins.d/
    
  • Switch to the netdata user.

    sudo -u netdata -s
    
  • Run the python.d.plugin to debug the collector:

    ./python.d.plugin pandas debug trace
    

Getting Logs

If you’re encountering problems with the pandas collector, follow these steps to retrieve logs and identify potential issues:

  • Run the command specific to your system (systemd, non-systemd, or Docker container).
  • Examine the output for any warnings or error messages that might indicate issues. These messages should provide clues about the root cause of the problem.

System with systemd

Use the following command to view logs generated since the last Netdata service restart:

journalctl _SYSTEMD_INVOCATION_ID="$(systemctl show --value --property=InvocationID netdata)" --namespace=netdata --grep pandas

System without systemd

Locate the collector log file, typically at /var/log/netdata/collector.log, and use grep to filter for collector’s name:

grep pandas /var/log/netdata/collector.log

Note: This method shows logs from all restarts. Focus on the latest entries for troubleshooting current issues.

Docker Container

If your Netdata runs in a Docker container named “netdata” (replace if different), use this command:

docker logs netdata 2>&1 | grep pandas

The observability platform companies need to succeed

Sign up for free

Want a personalised demo of Netdata for your use case?

Book a Demo