Integrate Sercomm Speedport Plus and ZTE Speedport Entry 2i (VDSL2 modems) with Home assistant using the command_line sensor platform.

Exported statistics

Choose the appropriate script for your modem:

• speedport_plus.py for Speedport Plus
• speedport_entry2i.py for Speedport Entry 2i

1. Download the python scripts

Create a folder named scripts_cli under your /config Home assistant folder and download there the speedport_plus.py or speedport_entry2i.py script from this repo. If you are new to Home assistant, consider the following ways to do that:

• If you have installed the "Samba share" addon, you may download the script first in your computer and then create the folder and copy-paste the file by accessing the "/config" share.
• If you have installed the "Terminal & SSH addon", you may mkdir config/scripts_cli && cd config/scripts_cli and download the script using wget and the raw github path to the script. You may also use git and clone this repo there.
• If you have installed the "File editor" addon, you may manually create the folder from the menu (top left folder icon) and then create a new file, name it speedport_plus.py / speedport_entry2i.py and copy-paste the contents from this repo.

All 3 addons are recommended for productive use and management of Home assistant, so you may try any of the suggestions above.

Note: If you are just using Home assistant core installed in a server, then you have to adapt your path to /path/to/homeassistant/config/scripts_cli

2. Configure the new sensor in configuration.yaml

After you have successfully added the script under config/scripts_cli/speedport_plus.py (config/scripts_cli/speedport_entry2i.py for the Entry 2i), you may edit configuration.yaml and create a command_line sensor:

sensor:
    - platform: command_line
      name: Speedport Plus status
      scan_interval: 60
      json_attributes:
        - vdsl_atnu
        - vdsl_atnd
        - dsl_crc_errors
        - dsl_fec_errors
        - dsl_snrd
        - dsl_snru
        - dsl_downstream
        - dsl_upstream
        - dsl_max_downstream
        - dsl_max_upstream
        - uptime
        - uptime_online
        - dsl_online_status
        - dsl_transmission_mode
        - firmware_version
      command: 'python3 /config/scripts_cli/speedport_plus.py "http://192.168.1.1"'
      value_template: '{{ value_json.dsl_link_status }}'

Change speedport_plus.py with speedport_entry2i.py above if you have this modem. Also change the path to /path/to/homeassistant/config/scripts_cli/ if you just use Home assistant core installed in a custom server location.

You may change the name: Speedport Plus status above with name: Speedport Entry2i status (or whatever you desire) if you have the Entry 2i modem. Just be aware that the dashboards shared in this repo have been created with this specific entity name (sensor.speedport_plus_status) therefore you will have to search/replace entity names to make them work.

Configure the router IP

If the IP of your router is not 192.168.1.1 or if you want to use a hostname, then change command: above with the correct http base url (including http:// but without a trailing slash /) as the first argument (quoted). For instance if the IP of your router is 10.0.50.1 change the configuration above as follows:

      command: 'python3 /config/scripts_cli/speedport_plus.py "http://10.0.50.1"'

for the Entry2i:

      command: 'python3 /config/scripts_cli/speedport_entry2i.py "http://10.0.50.1"'

Polling interval

The configuration suggested above will use a default scan_interval: 60 which means that you will get statistics from your modem every 60 seconds. You can lower it below 60 seconds but the amount of storage required to store the statistics will increase accordingly. If you only care about long term statistics and don't care to catch short term spikes, you may want to increase it to 300 seconds (5 minutes).

DSL status Vs Internet status

The configuration suggested above will use the DSL synchronization status as the "online/offline" status of the sensor. You may want to change it to indicate full IP connectivity instead (i.e. you have an active PPPoE session and an assigned IP address). You can do this by changing the value_template field as:

      value_template: '{{ value_json.dsl_online_status }}'

(i.e. dsl_link_status replaced by dsl_online_status)

3. Restart Home assistant and then create widgets or automations

Restart Home assistant and the new sensor will be available as the entity speedport_plus_status. You may add the new sensor to your dashboards or create automations.

The new sensor entity speedport_plus_status (or speedport_entry2i_status) will have the values "online" or "offline". The DSL line metrics (attenuation, snr, sync speed etc.) will be available as attributes of this entity.

You may use the sensor status (online/offline) in automations (for instance, restart your router with a smart plug when disconnected). You may also use the numeric attributes using trigger numeric_state and selecting an attribute from the list.

Monitoring your Internet connection long term may be more practical with a tool such as Grafana. A sample dashboard is included in this repo. Read the following sections for more.

If you want to make some of the DSL attributes available as individual sensors (useful for UI widgets), you may accomplish this with template sensors:

sensor:
    - platform: template
      sensors:
          dsl_sync_downstream:
              friendly_name: DSL Sync downstream
              value_template: >-
                  {{state_attr("sensor.speedport_plus_status", "dsl_downstream") | float | multiply(0.001) | round(2) }}
              unit_of_measurement: "Mbit/s"
          dsl_sync_upstream:
              friendly_name: DSL Sync upstream
              value_template: >-
                  {{state_attr("sensor.speedport_plus_status", "dsl_upstream") | float | multiply(0.001) | round(2) }}
              unit_of_measurement: "Mbit/s"
          dsl_errors_crc:
              friendly_name: DSL CRC Errors
              value_template: >-
                  {{state_attr("sensor.speedport_plus_status", "dsl_crc_errors") | int }}
          dsl_errors_fec:
              friendly_name: DSL FEC Errors
              value_template: >-
                  {{state_attr("sensor.speedport_plus_status", "dsl_fec_errors") | int }}

(Change sensor.speedport_plus_status to sensor.speedport_entry2i_status in the above snippet if you have the Entry 2i and have chosen this name)

You may also want to compute error rates which are more useful than totals for detecting spikes during the day. We can use the Home assistant derivative platform to accomplish that.

(Make sure you have added the invidual sensors as suggested above first)

sensor:
    - platform: derivative
      source: sensor.dsl_errors_crc
      name: DSL Error rate (CRC)
      round: 0
      unit_time: h
      unit: "Err/h"
      time_window: "00:05:00"  # we look at the change over the last 5 minutes
    - platform: derivative
      source: sensor.dsl_errors_fec
      name: DSL Error rate (FEC)
      round: 0
      unit_time: h
      unit: "Err/h"      
      time_window: "00:05:00"  # we look at the change over the last 5 minutes

InfluxDB and Grafana are available as Home assistant (community) addons. Check the Add-on Store under the Supervisor section.

If you have integrated Home assistant with InfluxDB, the entity attributes (DSL speed, snr, attenuation etc.) will be also available as seperate "fields" in the InfluxDB "measurement" time-series created for this entity.

The dashboard displayed above is available in this repo under the grafana folder. You may import it into your Grafana instance while choosing your InfluxDB data source. Read the related section further below for more details.

Sample queries

In Grafana, if you want to plot the downstream SNR, then you have to choose in the Visual editor:

FROM default state WHERE entity_id = speedport_plus_status
SELECT field(dsl_snrd) last()
GROUP BY time($__interval) tag(entity_id) fill(none)

(this is not an actual InfluxDB query, just the way Grafana UI editor displays the query parts)

In text query mode, this is:

SELECT last("dsl_snrd") FROM "state" WHERE ("entity_id" = 'speedport_plus_status') AND $timeFilter GROUP BY time($__interval) fill(none)

(AND $timeFilter is added by Grafana to allow filtering by time using the top right dropdown menu)

This assumes that you use state as the default measurement name when a metric is "unitless". i.e. in configuration.yaml you have:

influxdb:
    ...
    default_measurement: state
    ...

If it's not the case, then you should adjust the queries above accordingly (you may need to change FROM "state" with FROM "sensor.speedport_plus_status" or whatever entity name you have defined).

Rates instead of total stats

You may compute rates (e.g. errors per minute) instead of totals using InfluxDB transformation functions such as: non_negative_difference() and non_negative_derivative().
Or just use the Home assistant derivative platform as described in the previous section.

The dashboard displayed above is available in this repo under the grafana folder. When you import it into your Grafana instance you will be asked to choose your InfluxDB data source.

Notes:

• The dashboard has been created for the Speedport Plus modem. If you have an Entry 2i modem and you have renamed the name of the sensor as "Speedport Entry2i status" you'll have to search and replace within the json dashboard file all instances of speedport_plus_status to speedport_entry2i_status.
• If you haven't set default_measurement: state in the InfluxDB integration as mentioned in the previous section, you'll have to search and replace within the json dashboard file all instances of "measurement": "state" to "measurement": "sensor.speedport_plus_status" (or sensor.speedport_entry2i_status).

A lovelace view (tab in a dashboard) has been shared in this repo under the lovelace_dashboard_views folder. You have to place it within an existing dashboard (under the views: key) using the Raw configuration editor (or by editing the yaml file if you use yaml mode).

The lovelace dashboard depends on the following custom cards (available in Hacs): layout card, mini graph card, entity attributes card.

It also includes widgets that depend on the extra integrations listed in the following section (UPnP, Ping binary sensor for latency, Speedtest.net). Of course, if you don't want to use these extra integrations or if you want to use only builtin cards you may adapt the dashboard to your needs.

Finally, you will have to replace the dummy values with real user IDs under the visible: key. You may do that in the tab options in edit mode using the UI or you may copy-paste your user IDs from another dashboard yaml configuration.

UPnP/Internet Gateway Device (IGD)

If you have UPnP enabled on your router, Home assistant will detect it and offer to add the integration (or you may check manually from the integrations page). This will output traffic statistics from your home network. It's not very consistent with Speedport Plus but when it works it may provide an indication of (part of) your traffic.

Ping (ICMP) Binary sensor

The ping integration is usually used for presence detection. But you may also use it to check your Internet latency.

binary_sensor:
  # Ping an IP of your ISP
  - platform: ping
    host: hostname-or-ip-of-your-isp-to-ping
    name: "Ping ISP"
    count: 3
    scan_interval: 3600
  # Ping an IP in your city
  - platform: ping
    host: hostname-or-ip-of-other-isp-to-ping
    name: "Ping City"
    count: 3
    scan_interval: 3600
  # Ping an IP abroad
  - platform: ping
    host: hostname-or-ip-from-other-country-to-ping
    name: "Ping Abroad"
    count: 3
    scan_interval: 3600 

The examples above will make HA ping 3 different IP addresses (3 times every 1 hour) and measure the latency.

The binary sensor (connected/not connected) exposes the round trip time values measured by ping as attributes: round trip time avg, round trip time min, round trip time max.

If you want to expose the latencies as separate sensors you can use template sensors (required by the shared Lovalace dashboard view shared in this repo):

sensor:
    - platform: template
      sensors:
          net_latency_isp:
              friendly_name: Internet Latency (ISP)
              value_template: >-
                  {{state_attr("binary_sensor.ping_isp", "round_trip_time_avg") | float | round(1) }}
              unit_of_measurement: "ms"
          net_latency_city:
              friendly_name: Internet Latency (City)
              value_template: >-
                  {{state_attr("binary_sensor.ping_city", "round_trip_time_avg") | float | round(1) }}
              unit_of_measurement: "ms"
          net_latency_abroad:
              friendly_name: Internet Latency (Abroad)
              value_template: >-
                  {{state_attr("binary_sensor.ping_abroad", "round_trip_time_avg") | float | round(1) }}
              unit_of_measurement: "ms"

Speedtest.net Integration

You may measure your true download and upload speeds and your ping (latency) with the Speedtest.net integration. This way you can compare your DSL sync speed with your actual speed.

Configure the integration from the UI per the instructions in the linked page above.

• It is advised to not run the test very frequently because it will clog your connection on unpredictable times (I have changed mine to 360 minutes - 6 hours).
• It is suggested to set a specific server close to you (see the dropdown in the "Options" menu) so you have consistent results to compare long term.

Note that a Raspberry PI (other than 4B or newer) may limit the reported maximum speed to 100Mbps or 300Mbps (Raspberry PI 3+) due to the slower LAN adapter. Also note that this integration uses speedtest-cli which may be a little slower than the results you get on Speedtest.net. Your true latency may be 7-10ms lower than what is reported from this tool. Read more here. The ping integration mentioned previously is more accurate for measuring latency.

Alternatively you may use the Iperf3 or the Fast.com integrations.

Make sure that the admin page of your modem (http://192.168.1.1 by default) is accessible from the Home assistant host. If you have ssh enabled, you may use it to ping the IP from the command line.

You may test the Python script from your PC (if you have python3 installed).

Run without arguments to retrieve data from the default IP address (http://192.168.1.1):

# for Speedport Plus
python3 speedport_plus.py

# for Speedport Entry 2i
python3 speedport_entry2i.py

If the IP of your router is not 192.168.1.1 or if you want to use a hostname, then supply the http base url (including "http://" but without a trailing slash) as the first argument (quoted). For example:

# for Speedport Plus
python3 speedport_plus.py "http://10.0.50.1"

# for Speedport Entry 2i
python3 speedport_entry2i.py "http://10.0.50.1"

The expected response should be similar to:

{"firmware_version": "09022001.00.031_OTE1", "dsl_link_status": "online", "dsl_downstream": 104254, "dsl_upstream": 10996, "dsl_max_downstream": 80692, "dsl_max_upstream": 48300, "dsl_transmission_mode": "...", "dsl_crc_errors": 10825, "dsl_fec_errors": 21, "vdsl_atnu": 10.0, "vdsl_atnd": 6.5, "dsl_snrd": 10.6, "dsl_snru": 31.0, "uptime_online": 285555, "dsl_online_status": "online", "uptime": 285562}

You may increase the verbosity of the logs for the command_line platform by adding the logger: platform at configuration.yaml:

logger:
  default: warning
  logs:
    homeassistant.components.command_line: debug

If the configuration works properly you should see entries like the following in the logs:

2021-04-15 23:30:54 DEBUG (SyncWorker_16) [homeassistant.components.command_line.sensor] Running command: python3 /config/scripts_cli/speedport_plus.py "http://192.168.1.1"
....
2021-04-15 23:31:54 DEBUG (SyncWorker_9) [homeassistant.components.command_line.sensor] Running command: python3 /config/scripts_cli/speedport_plus.py "http://192.168.1.1"

If there are any warnings it may help tracking down the cause. When you finish debugging, you may revert the debug logging level to reduce the messages in the logs.

The python scripts have been tested with the following firmware versions:

• Speedport Plus: 09022001.00.031_OTE2, 09022001.00.031_OTE1, 09022001.00.030_OTE5
• Speedport Entry 2i: V1.0.0_OTET14MAPEA

If the integration does not work and the command line testing fails or produces different results than suggested above, check whether the device has a new firmware and please open an issue.