Page History

General information

The

To view the system status data, you need to:

...

web interface of the self-diagnostics service is available for monitoring the system statuses and analyzing its performance.

Access to the self-diagnostics service

To go to the monitoring interface, do the following:

1. Open the web browser.
2. In the address line, enter: http://127.0.0.1:20040/
.
Image Removed
3. .
4. Click the Enter button.

Interface and queries execution

The service interface allows viewing metrics as a table or graphs. To run a query, do the following:

1. Select the

required

2. metric

in

3. from the drop-down list

(

4. 1

)

5. or enter the query manually in the Expression field.

Expandtitle Description of useful metrics

6. You can:

   1. Use several metrics at a time. The system has the following available metrics:
      

      Metric	Description
      ALERTS_FOR_STATE

Troubleshooting by self-diagnostics service

7. 1. Found and fixed malfunctions. Contains the alertname parameter with the problem type. Code Block language go title Example ALERTS_FOR_STATE{alertname="ipint_is_not_activated",ep_name="hosts/Server1/DeviceIpint.99",instance="127.0.0.1:20108",job="ngp_

export

8. 1. exporter",ngp_alert="true"}

Possible values of the alert name parameter

9. 1. Decryption of the alertname values (see General information about the self-diagnostics service) for the ALERTS_FOR_STATE metric: low_os_memory

− out

10. 1. —out of RAM

;

11. 1. . ipint_is_not_activated

− camera

12. 1. —camera is connected

but does not send data;

13. 1. , but there is no data from it. no_samples_in_detector

− no

14. 1. —no events from

a detection tool;

15. 1. the detector. restart_services_when_archive_source_not_activated

− the archive is not working;

16. 1. —the archive recording isn't working. restart_services_when_no_samples_in_archive

− recording

17. 1. —recording to archive with 0

FPS;

18. 1. fps. restart_services_when_no_ping_from_detector_to_

archive − no footage is recorded on detection;

19. 1. archive—no recording to the archive at the event from the detector. logs_disk_space_is_low / db_disk_space_is_low

− out

20. 1. —out of system disk space.
       ngp_archive_channel_fps	The frame rate of all

video

21. 1. cameras when recording to

archive.

22. 1. the archive
       ngp_archive_volume_size	The current total size of the archive (in bytes)
       ngp_cpu_total_usage	The

percentage of

23. 1. CPU load

on a Server.

24. 1. of the server
       ngp_fps	The frame rate of all

Server video

25. 1. server cameras,

all detection tools

26. 1. detectors and

their decoders.

The request allows for:

27. Using multiple metrics.

Using expressions to find issues. For example, a query like ngp_fps <17 will return all metrics where fps was less than 17

28. 1. decoders
       ngp_people_count	The last captured number of people in the frame by the Crowd estimation VA detector
       ngp_errors	Number of errors in the detectors' operation: Neural tracker Neural counter Human pose detector Neural classifier Stopped object detector Smoke detector Fire detector Water level detector with a neural network Barcode detector Meta-detector Person-based privacy masking Object tracker (with a Neural filter) Crowd estimation VA
       ngp_skipped_pp	Number of missed frames by the Crowd estimation VA detector due to the lack of resources for processing

    2. Apply logic and arithmetic operators for anomaly searching. The full list of
logical
29. 1. logic and arithmetic operators is
listed
30. 1. specified in
the Filtering by any of the parameters. For example, a query like
31. 1. the official Prometheus documentation
.  
32. 1. .
       

       Code Block
       language go title Example. All metrics where fps is less than 17
       ngp_fps < 17

    2. Faltering by metrics parameters using curly brackets.
       

       Code Block
       language go title Example. Fps values only for the specified source

33. 1. ngp_fps{ep_name=~"hosts/

V-BELYAKOV

34. 1. TEST/DeviceIpint.2/SourceEndpoint.video:0:0"}

will return fps values only for the specified source
35. 1. Image Added
36. If necessary, set the time range for the data.

37. Click the Execute

 (2).

38. button.

Viewing results:

• The Console tab

...

• displays the current metrics values in the table format.
  Image Modified
  When you

...

• specify the date and time in

...

• the calendar, the data

...

• is updated.
  Image Modified
  

...

• On the Graph tab you can create the graph of selected metrics at the specified period.

  ...

  • • The 1 field—sets the graph time interval.
    • The 2 field—specifies the end graph point.
    • The 3 field—sets

  ...

  • • the interval between data points.
    • The 4 checkbox—enables the display mode with accumulation (filling the areas under the graph).
      Image Added
      

  Examples of useful queries for Windows OS

  1. The CPU loading graph (analog of the System monitor):
     

     Code Block
     language go
     sum by (process_id) (100 / scalar(wmi_cs_logical_processors) * (irate(wmi_process_cpu_time_total{process="AppHost"}[10m]))) or ngp_cpu_total_usage

  2. RAM usage by the AppHost processes and a total memory space:
     

     Code Block
     language go
     sum by (process_id) (avg_over_time(wmi_process_working_set{process="AppHost"}[5m])) / 1024 or avg_over_time(wmi_os_virtual_memory_bytes[5m]) / 1024

  3. The percentage of RAM usage:
     

     Code Block
     100.0 - 100 * avg_over_time(wmi_os_virtual_memory_free_bytes[5m]) / avg_over_time(wmi_os_virtual_memory_bytes[5m])

  Examples of useful queries for Linux OS

  1. The total RAM usage by the AppHost processes:
     

     Code Block
     language go
     sum by (groupname) (namedprocess_namegroup_memory_bytes{memtype="resident"})

  2. The percentage of RAM usage:
     

     Code Block
     language go
     100 - node_memory_MemAvailable_bytes * 100 / node_memory_MemTotal_bytes

  3. The CPU load by the AppHost processes as a percentage:
     

     Code Block
     sum by (object_id) (rate(namedprocess_namegroup_cpu_seconds_total{groupname="AppHost"}[1m])) * 100

  4. The total CPU load as a percentage:
     

     Code Block
     language go
     100 * avg without (cpu) (1 - rate(node_cpu_seconds_total{mode="idle"}[1m]))

  5. RAM usage by the AppHost processes to detect the memory leak:
     

     Code Block
     language go
     namedprocess_namegroup_memory_bytes{object_id=~"APP_HOST.*",memtype="proportionalResident"}

  To fill the chart, select the stacked (4) checkbox.