Hello,

We would like to use this plugin in some kind of "self service manner" netbox wise. Meaning that we would essentially use one http_sd_config (technically two for vms and server in our case) with something like: /api/plugins/prometheus-sd/devices/?tags=prom-autoconfig and then use config context to configure metrics_path, ports etc.

The idea is to have something similar as what we used to configure Kubernetes (essentially before or if prometheus-operator is not used) with annotation on pods like: prometheus.io/scheme, prometheus.io/path, prometheus.io/port.

I propose to put something like that in the config context:

prometheus-plugin-prometheus-sd:
  - metrics_path: /not/metrics
    port: 4242
    scheme: https
  - port: 4243

And then the plugin would discover this and use it for the http sd json essentially. Of course I am happy to work on the implementation, in case that's something that could be included!

Before I start, thank you for this work and I fully admit this could be me. No experience with Prometheus at all here.
prometheus-2.42.0.linux-amd64
netbox 3.4.5

API seems to return good data, validly formatted. Ran output through json lint and looks valid. All elements are strings excep for the id.

          "__meta_netbox_id": 81,

Prometheus never shows the target at all and logs this error:

Feb 23 08:35:13 prometheus.netfortris.com prometheus[11055]: ts=2023-02-23T16:35:13.317Z caller=refresh.go:99 level=error component="discovery manager scrape" discovery=http config=http_sd_devices msg="Unable to refresh target groups" err="json: cannot unmarshal number into Go struct field .labels of type model.LabelValue"

Prometheus config:

  - job_name: "http_sd_devices"
#    metrics_path: /netbox
    scheme: https
#    static_configs:
#      - targets: ["localhost:9090"]
    http_sd_configs:
    - url: https://netbox.netfortris.net/api/plugins/prometheus-sd/devices/?has_primary_ip=true&name=atl-bdr1
      refresh_interval: 15s
      authorization:
        type: "Token"
        credentials: "blahblahblahforthelength of the token."

Validated with curl that the data is returned properly from the url with the proper api key. I get different issues if I try to use the metrics_path so I let it default with the other job into /metrics.

In Netbox 3.6, "device_role" has been renamed to "role". This gives the following exception:

Internal Server Error: /api/plugins/prometheus-sd/devices/

ValueError at /api/plugins/prometheus-sd/devices/
'device_role' does not resolve to an item that supports prefetching - this is an invalid parameter to prefetch_related().

Request Method: GET
Request URL: https://netbox.example.net/api/plugins/prometheus-sd/devices/?tag=prom_snmp&status=active
Django Version: 4.2.4
Python Executable: /opt/netbox/venv/bin/python3.8
Python Version: 3.8.10
Python Path: ['/opt/netbox/netbox', '/opt/netbox', '/opt/netbox/venv/bin', '/usr/lib/python38.zip', '/usr/lib/python3.8', '/usr/lib/python3.8/lib-dynload', '/opt/netbox/venv/lib/python3.8/site-packages']
Server time: Thu, 31 Aug 2023 10:08:31 +0000
Installed Applications:
['django.contrib.admin',
 'django.contrib.auth',
 'django.contrib.contenttypes',
 'django.contrib.sessions',
 'django.contrib.messages',
 'django.contrib.staticfiles',
 'django.contrib.humanize',
 'corsheaders',
 'debug_toolbar',
 'graphiql_debug_toolbar',
 'django_filters',
 'django_tables2',
 'django_prometheus',
 'graphene_django',
 'mptt',
 'rest_framework',
 'social_django',
 'taggit',
 'timezone_field',
 'core',
 'account',
 'circuits',
 'dcim',
 'ipam',
 'extras',
 'tenancy',
 'users',
 'utilities',
 'virtualization',
 'wireless',
 'django_rq',
 'drf_spectacular',
 'drf_spectacular_sidecar',
 'netbox_prometheus_sd.PrometheusSD']
Installed Middleware:
['graphiql_debug_toolbar.middleware.DebugToolbarMiddleware',
 'django_prometheus.middleware.PrometheusBeforeMiddleware',
 'corsheaders.middleware.CorsMiddleware',
 'django.contrib.sessions.middleware.SessionMiddleware',
 'django.middleware.common.CommonMiddleware',
 'django.middleware.csrf.CsrfViewMiddleware',
 'django.contrib.auth.middleware.AuthenticationMiddleware',
 'django.contrib.messages.middleware.MessageMiddleware',
 'django.middleware.clickjacking.XFrameOptionsMiddleware',
 'django.middleware.security.SecurityMiddleware',
 'netbox.middleware.RemoteUserMiddleware',
 'netbox.middleware.CoreMiddleware',
 'netbox.middleware.MaintenanceModeMiddleware',
 'django_prometheus.middleware.PrometheusAfterMiddleware']


Traceback (most recent call last):
  File "/opt/netbox/venv/lib/python3.8/site-packages/django/core/handlers/exception.py", line 55, in inner
    response = get_response(request)
  File "/opt/netbox/venv/lib/python3.8/site-packages/django/core/handlers/base.py", line 197, in _get_response
    response = wrapped_callback(request, *callback_args, **callback_kwargs)
  File "/opt/netbox/venv/lib/python3.8/site-packages/django/views/decorators/csrf.py", line 56, in wrapper_view
    return view_func(*args, **kwargs)
  File "/opt/netbox/venv/lib/python3.8/site-packages/rest_framework/viewsets.py", line 125, in view
    return self.dispatch(request, *args, **kwargs)
  File "/opt/netbox/netbox/netbox/api/viewsets/__init__.py", line 91, in dispatch
    return super().dispatch(request, *args, **kwargs)
  File "/opt/netbox/venv/lib/python3.8/site-packages/rest_framework/views.py", line 509, in dispatch
    response = self.handle_exception(exc)
  File "/opt/netbox/venv/lib/python3.8/site-packages/rest_framework/views.py", line 469, in handle_exception
    self.raise_uncaught_exception(exc)
  File "/opt/netbox/venv/lib/python3.8/site-packages/rest_framework/views.py", line 480, in raise_uncaught_exception
    raise exc
  File "/opt/netbox/venv/lib/python3.8/site-packages/rest_framework/views.py", line 506, in dispatch
    response = handler(request, *args, **kwargs)
  File "/opt/netbox/netbox/netbox/api/viewsets/mixins.py", line 95, in list
    return super().list(request, *args, **kwargs)
  File "/opt/netbox/venv/lib/python3.8/site-packages/rest_framework/mixins.py", line 46, in list
    return Response(serializer.data)
  File "/opt/netbox/venv/lib/python3.8/site-packages/rest_framework/serializers.py", line 768, in data
    ret = super().data
  File "/opt/netbox/venv/lib/python3.8/site-packages/rest_framework/serializers.py", line 253, in data
    self._data = self.to_representation(self.instance)
  File "/opt/netbox/venv/lib/python3.8/site-packages/rest_framework/serializers.py", line 686, in to_representation
    return [
  File "/opt/netbox/venv/lib/python3.8/site-packages/django/db/models/query.py", line 398, in __iter__
    self._fetch_all()
  File "/opt/netbox/venv/lib/python3.8/site-packages/django/db/models/query.py", line 1883, in _fetch_all
    self._prefetch_related_objects()
  File "/opt/netbox/venv/lib/python3.8/site-packages/django/db/models/query.py", line 1273, in _prefetch_related_objects
    prefetch_related_objects(self._result_cache, *self._prefetch_related_lookups)
  File "/opt/netbox/venv/lib/python3.8/site-packages/django/db/models/query.py", line 2310, in prefetch_related_objects
    raise ValueError(

Exception Type: ValueError at /api/plugins/prometheus-sd/devices/
Exception Value: 'device_role' does not resolve to an item that supports prefetching - this is an invalid parameter to prefetch_related().
Raised during: netbox_prometheus_sd.api.views.DeviceViewSet
Request information:
USER: admin

GET:
tag = 'prom_snmp'
status = 'active'

POST: No POST data

FILES: No FILES data

Hi all, Firstly GREAT project :D

With netbox having the ability to build hierarchies of Locations, and then racks living in a location.

Eg

Currently the prom_sd exporter returns the rack and the location on the next level up, so using the above example, you can get the rack and the racks first level parent and the site.

What you cant get is the top level parent of the rack.

i have had a look in the code, but I am unable to work out the best way to get that data, I think we are going to have to make another call to pull in the parent of the first level parent.

Any suggestions on how this might be implemented, im happy to take a stab at it, but im struggling to work out the best way forward.

Hey.

It would be great if the API could expose both IPv6 and IPv4 primary addresses instead/on top of __meta_netbox_primary_ip.

Hello,

Currently the sd plugin creates an api call for each job following the configured refresh_interval under the http_sd_config in prometheus (http_sd_config).

This might create some load if you have quite some jobs for different device types running over different prometheus instances.
It would be nice to let this plugin do the initial scrape on prometheus request and subscribe to changes through Netbox webhooks and dynamically update the sd result for prometheus based on the received changes from the Netbox webhook.

This creates a kind of cache for netbox limiting the amount of database calls happening.

I am unsure how in scope this is but it would be nice to have it as a feature that can be enabled on request (giving both options between in memory cache or not)

Hi,
I am testing using netbox v4.0.5 on docker got error :

example-netbox-1 | ⏳ Waiting on DB... (15s / 30s)
example-netbox-1 | 🧬 loaded config '/etc/netbox/config/configuration.py'
example-netbox-1 | 🧬 loaded config '/etc/netbox/config/extra.py'
example-netbox-1 | 🧬 loaded config '/etc/netbox/config/logging.py'
example-netbox-1 | 🧬 loaded config '/etc/netbox/config/plugins.py'
example-netbox-1 | Traceback (most recent call last):
example-netbox-1 | File "/opt/netbox/netbox/./manage.py", line 10, in
example-netbox-1 | execute_from_command_line(sys.argv)
example-netbox-1 | File "/opt/netbox/venv/lib/python3.11/site-packages/django/core/management/init.py", line 442, in execute_from_command_line
example-netbox-1 | utility.execute()
example-netbox-1 | File "/opt/netbox/venv/lib/python3.11/site-packages/django/core/management/init.py", line 436, in execute
example-netbox-1 | self.fetch_command(subcommand).run_from_argv(self.argv)
example-netbox-1 | File "/opt/netbox/venv/lib/python3.11/site-packages/django/core/management/base.py", line 413, in run_from_argv
example-netbox-1 | self.execute(*args, **cmd_options)
example-netbox-1 | File "/opt/netbox/venv/lib/python3.11/site-packages/django/core/management/base.py", line 454, in execute
example-netbox-1 | self.check()
example-netbox-1 | File "/opt/netbox/venv/lib/python3.11/site-packages/django/core/management/base.py", line 486, in check
example-netbox-1 | all_issues = checks.run_checks(
example-netbox-1 | ^^^^^^^^^^^^^^^^^^
example-netbox-1 | File "/opt/netbox/venv/lib/python3.11/site-packages/django/core/checks/registry.py", line 88, in run_checks
example-netbox-1 | new_errors = check(app_configs=app_configs, databases=databases)
example-netbox-1 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
example-netbox-1 | File "/opt/netbox/venv/lib/python3.11/site-packages/django/core/checks/model_checks.py", line 18, in check_all_models
example-netbox-1 | models = apps.get_models()
example-netbox-1 | ^^^^^^^^^^^^^^^^^
example-netbox-1 | File "/opt/netbox/venv/lib/python3.11/site-packages/django/apps/registry.py", line 181, in get_models
example-netbox-1 | self.check_models_ready()
example-netbox-1 | File "/opt/netbox/venv/lib/python3.11/site-packages/django/apps/registry.py", line 143, in check_models_ready
example-netbox-1 | raise AppRegistryNotReady("Models aren't loaded yet.")
example-netbox-1 | django.core.exceptions.AppRegistryNotReady: Models aren't loaded yet.
example-netbox-1 | ⏳ Waiting on DB... (18s / 30s)

Thank you

Hi,

There are some breaking changes in Netbox 4.0, including the following:

from extras.plugins import PluginConfig should be replaced by from netbox.plugins import PluginConfig.

I'm not sure what else needs to be changed, but is an update planned to be compatible with Netbox 4.0 ? Thanks :)

P.S.: Here are the changes from Netbox Doc.

I'm not positive when this broke, but noticed it this morning, so possibly related to netbox 4.0.9 changes but will need to look.

            "__meta_netbox_rack_u_position": 31.0,

This causes:

ts=2024-08-16T14:25:05.851Z caller=refresh.go:90 level=error component="discovery manager scrape" discovery=http config=nb_snmp msg="Unable to refresh target groups" err="json: cannot unmarshal number into Go struct field .labels of type model.LabelValue"

Easy fix in utils to force it back to a string. PR to come.

Goal: Provide releases as fast as possible and less errors.

https://github.com/semantic-release/semantic-release

• Pull a release based on conventional commits
• Update version strings in __init.py

While this plugin works great with a small number of devices, it becomes very slow at scale. For example, with 1158 devices on my test, the following query takes 23 seconds.

This instance is running v3.7.0 on EC2 m5.2xlarge (8 vCPU and 32GB memory), connected to Postgres 15 in RDS Aurora.

/api/plugins/prometheus-sd/devices/?status=active&cf_prometheus_blackbox_module__n=null

I did a little investigation and found that 99% of that time is spent querying the database. The HTTP download is only 10ms. Postgres log shows that this one API call causes 6157 SELECT statements, about 2/3 of which return 0 rows. You can see this log here: prometheus-sd-query.psql.zip

To easily replicate this issue, you can launch a fresh instance of netbox-docker and run the following code in nbshell. This way it's a bit faster, taking 13 seconds, perhaps because the database is local.

# >>> exec(open('/etc/netbox/config/scripts/init-dummy-data.py').read())

import ipaddress
from django.db import transaction

s = Site(name='Dummy Site', slug='dummy-site')
s.full_clean()
s.save()

m = Manufacturer(name='Generic', slug='generic')
m.full_clean()
m.save()

dt = DeviceType(model='Generic Device', manufacturer_id=1, slug='generic-device')
dt.full_clean()
dt.save()

it = InterfaceTemplate(name='eth0', type='virtual', device_type_id=1)
it.full_clean()
it.save()

dr = DeviceRole(name='Generic Role', slug='generic-role')
dr.full_clean()
dr.save()

print(f"Created Dummy Site, Manufacturer, Device Type, and Device Role")
print("Creating {num_hosts} dummy devices...")

hosts = list(ipaddress.IPv4Network('127.0.0.0/16'))
num_hosts= 2000
for i in range(num_hosts):
    # if any nested atomic transactions fail then device will not be created
    with transaction.atomic():
        # create device (and interface)
        d = Device(
            site=s,
            name = f'Dummy Device {i:03}',
            device_type=dt, 
            role=dr
        )
        d.full_clean()
        d.save()
        
        # create IP and assign to interface
        ip = IPAddress(address=f"{hosts[i]}/32")
        ip.assigned_object = d.interfaces.first()
        with transaction.atomic():
            ip.full_clean()
            ip.save()
        
        # assign device primary IP
        d.primary_ip4 = ip
        with transaction.atomic():
            d.full_clean()
            d.save()

print(f"Done!")

Hello,
Would it be possible to exclude devices/VMs being shown in the api/plugins/prometheus-sd/devices/ API endpoint please?
For context - we're looking at making a low weight config context with something like

{
    "prometheus-plugin-prometheus-sd": {
        "monitor": "false"
    }
}

in it. Then a higher weight one that we can apply to certain roles, sites etc. that has more useful information:

{
    "prometheus-plugin-prometheus-sd": {
        "monitor": "true",
        "metrics_path": "/metrics",
        "port": 9100,
        "scheme": "http"
    }
}

This would then allow us to get just the devices we'd like to monitor, without having to filter by roles in the API and setting the correct config context for roles inside Netbox for instance. Effectively, by only changing config contexts in Netbox your list of devices to monitor is automatically updated without you needing to change any Prometheus config at all!

e.g. you wouldn't need to do api/plugins/prometheus-sd/devices/?role=x&role=y any more because you can do it all via Netbox.
What do you think?
Thanks!

The current list of __meta_netbox labels is good, but I think it would be helpful to include Site and Region.

Forgive me if I have just missed them or if there is a good reason not to include them.

Hi, the last release has no version provided so the latest version is still .6.1.

Version

# /opt/netbox/venv/bin/pip freeze | grep prometheus-sd
netbox-plugin-prometheus-sd @ git+https://github.com/FlxPeters/netbox-plugin-prometheus-sd@686cb3ba3cd726a163659f2303aad822414bb6a7

running under Netbox 3.7.4

Issues

I have a couple of problems with labels from a custom field, where the value it contains is a list. The example here is a custom field called "snmp_module" of type "multiple selection", linked to a choice set with strings.

Problem 1: the label value returned is a Python-style repr().

# curl -sS -H "Authorization: Token $TOKEN" "$NETBOX/api/plugins/prometheus-sd/devices/?cf_snmp_module__ic=i" | jq .
...
      "__meta_netbox_custom_field_snmp_module": "['hrDevice', 'hrStorage', 'if_mib', 'mikrotik']",

In order to get this into a simple comma-separated list (as required by snmp_exporter with multiple modules) I have to use a very ugly rewrite:

      - source_labels: [__meta_netbox_custom_field_snmp_module]
        target_label: __param_module
      # Ugh: multiselect is of form ['foo', 'bar'] and we need foo,bar. There is no gsub.
      - source_labels: [__param_module]
        regex: "\\['(.*)'\\]"
        target_label: __param_module
      - source_labels: [__param_module]
        regex: "(.*)', *'(.*)"
        replacement: "$1,$2"
        target_label: __param_module
      - source_labels: [__param_module]
        regex: "(.*)', *'(.*)"
        replacement: "$1,$2"
        target_label: __param_module
      - source_labels: [__param_module]
        regex: "(.*)', *'(.*)"
        replacement: "$1,$2"
        target_label: __param_module

... and this only works up to a fixed maximum number of values (4 here), as there's no "gsub" in prometheus rewriting.

Problem 2: when the custom field is unset the label gets the string value "None", whereas I would expect either empty string or the entire label to be omitted.

...
      "__meta_netbox_custom_field_snmp_module": "None",

(This is minor: I could always match this value explicitly in a rewrite rule, and I'm unlikely to use "None" as a genuine choice value)

Proposal

For values of a multiple selection type, I'd like to see such a list returned as a plain comma-separated list of strings.

What about other types of custom field that might return a list? I can only see two that might make a difference:

• JSON fields. These can stay as is. JSON is JSON, after all.
• Multiple Object Selection. I haven't tested what these do today. Maybe simplest is to generate a comma-separated list of the str(...) representation of each object, so you get whatever the default display is. (I suppose it could also be a comma-separated list of object ID, or of slug for object classes which have slugs, but there's no obvious one-size-fits-all)

The question is whether the prometheus-sd plugin would have to dig into the definition of a custom field to determine whether it's JSON or not.

References

Relates to #34, #65

Is there a way to include Netbox location field as a label to Prometheus? I organize devices using a site > room hierarchy, so I'm using location to mean the room. Site represents a physical building here.

I can't seem to wrap my head around how this works in general. Netbox api call (with slight mod to change id to string from other thread) returns a boatload of items out of my netbox.

Those 758 objects all show as "down" because prometheus then tries to contact each of them on port 80 (of which most do not respond). I assume that I should be adding something to the netbox side so that prometheus then knows what to look for in an object? i.e. use snmp, snmp community, address to use instead of the name?

I tried to install this plugin under netbox v3.2 that was released a few days ago. Unfortunately the plugin fails with:

netbox-worker_1        | Traceback (most recent call last):
netbox-worker_1        |   File "/opt/netbox/netbox/manage.py", line 10, in <module>
netbox-worker_1        |     execute_from_command_line(sys.argv)
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/core/management/__init__.py", line 446, in execute_from_command_line
netbox-worker_1        |     utility.execute()
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/core/management/__init__.py", line 440, in execute
netbox-worker_1        |     self.fetch_command(subcommand).run_from_argv(self.argv)
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/core/management/base.py", line 414, in run_from_argv
netbox-worker_1        |     self.execute(*args, **cmd_options)
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/core/management/base.py", line 455, in execute
netbox-worker_1        |     self.check()
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/core/management/base.py", line 487, in check
netbox-worker_1        |     all_issues = checks.run_checks(
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/core/checks/registry.py", line 88, in run_checks
netbox-worker_1        |     new_errors = check(app_configs=app_configs, databases=databases)
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/core/checks/urls.py", line 14, in check_url_config
netbox-worker_1        |     return check_resolver(resolver)
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/core/checks/urls.py", line 24, in check_resolver
netbox-worker_1        |     return check_method()
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/urls/resolvers.py", line 480, in check
netbox-worker_1        |     for pattern in self.url_patterns:
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/utils/functional.py", line 49, in __get__
netbox-worker_1        |     res = instance.__dict__[self.name] = self.func(instance)
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/urls/resolvers.py", line 696, in url_patterns
netbox-worker_1        |     patterns = getattr(self.urlconf_module, "urlpatterns", self.urlconf_module)
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/utils/functional.py", line 49, in __get__
netbox-worker_1        |     res = instance.__dict__[self.name] = self.func(instance)
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/django/urls/resolvers.py", line 689, in urlconf_module
netbox-worker_1        |     return import_module(self.urlconf_name)
netbox-worker_1        |   File "/usr/lib/python3.9/importlib/__init__.py", line 127, in import_module
netbox-worker_1        |     return _bootstrap._gcd_import(name[level:], package, level)
netbox-worker_1        |   File "<frozen importlib._bootstrap>", line 1030, in _gcd_import
netbox-worker_1        |   File "<frozen importlib._bootstrap>", line 1007, in _find_and_load
netbox-worker_1        |   File "<frozen importlib._bootstrap>", line 986, in _find_and_load_unlocked
netbox-worker_1        |   File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
netbox-worker_1        |   File "<frozen importlib._bootstrap_external>", line 855, in exec_module
netbox-worker_1        |   File "<frozen importlib._bootstrap>", line 228, in _call_with_frames_removed
netbox-worker_1        |   File "/opt/netbox/netbox/netbox/urls.py", line 9, in <module>
netbox-worker_1        |     from extras.plugins.urls import plugin_admin_patterns, plugin_patterns, plugin_api_patterns
netbox-worker_1        |   File "/opt/netbox/netbox/extras/plugins/urls.py", line 35, in <module>
netbox-worker_1        |     urlpatterns = import_object(f"{plugin_path}.api.urls.urlpatterns")
netbox-worker_1        |   File "/opt/netbox/netbox/extras/plugins/utils.py", line 31, in import_object
netbox-worker_1        |     spec.loader.exec_module(module)
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/netbox_prometheus_sd/api/urls.py", line 2, in <module>
netbox-worker_1        |     from .views import VirtualMachineViewSet, DeviceViewSet, IPAddressViewSet
netbox-worker_1        |   File "/opt/netbox/venv/lib/python3.9/site-packages/netbox_prometheus_sd/api/views.py", line 4, in <module>
netbox-worker_1        |     from extras.api.views import CustomFieldModelViewSet
netbox-worker_1        | ImportError: cannot import name 'CustomFieldModelViewSet' from 'extras.api.views' (/opt/netbox/netbox/extras/api/views.py)

It seems some changes are needed to make it compatible with newer netbox.

There seems to be a ticket with changes that were made to the plugin framework.

Dear @FlxPeters,

thank you very much for bringing this plugin to life!

For our usage, it would be beneficial if the following two features were available:

1. Read a custom field boolean prometheus-sd_enable and only show the entry if the box is ticked (default: true).
2. Read a custom field integer prometheus-sd_port that will specify the exporter's port for this entry (default: 9100).

Personally, I'd leave it up to the users to decide wether they want to add the custom fields by hand. As both fields have sane defaults, existing installations should not face any issues.

What do you think? Would these be helpful additions to the plugin? :)

Best, phpfs

curl -Li ${NETBOX_URL}/api/plugins/prometheus-sd/devices?role=server&status=active&limit=100 -H "Authorization: Token $NETBOX_TOKEN" -H "Content-Type: application/json; indent=4" -o /dev/null
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
100 47924 100 47924 0 0 873 0 0:00:54 0:00:54 --:--:-- 10523

real 0m54.860s
user 0m0.078s
sys 0m0.015s
root@instance-20240429-121233:~/netbox-docker#

I have ca 50 devices listed, netbox version is 4.0.5

Hi, below is the Netbox API return from /api/dcim/devices/?name=router-01.example.com and I would like to know if you could implement in your API a filter on the model or model slug (using the model__iregex) that will return all devices matching this model as well as an option to return all custom_fields as labels, please.

{
    "count": 1,
    "next": null,
    "previous": null,
    "results": [
        {
            "id": 8,
            "url": "https://netbox.example.com/api/dcim/devices/8/",
            "display": "router-01.example.com",
            "name": "router-01.example.com",
            "device_type": {
                "id": 61,
                "url": "https://netbox.example.com/api/dcim/device-types/61/",
                "display": "Dlink xSeries",
                "manufacturer": {
                    "id": 12,
                    "url": "https://netbox.example.com/api/dcim/manufacturers/12/",
                    "display": "DLINK",
                    "name": "DLINK",
                    "slug": "dlink"
                },
                "model": "Dlink xSeries",
                "slug": "dlink-xseries"
            },
            "device_role": {
                "id": 5,
                "url": "https://netbox.example.com/api/dcim/device-roles/5/",
                "display": "ROUTER",
                "name": "ROUTER",
                "slug": "ROUTER"
            },
            "tenant": {
                "id": 1,
                "url": "https://netbox.example.com/api/tenancy/tenants/1/",
                "display": "HOME",
                "name": "HOME",
                "slug": "home"
            },
            "platform": null,
            "serial": "01234567890",
            "asset_tag": null,
            "site": {
                "id": 1,
                "url": "https://netbox.example.com/api/dcim/sites/1/",
                "display": "Malakoff",
                "name": "Paris",
                "slug": "paris"
            },
            "location": null,
            "rack": {
                "id": 8,
                "url": "https://netbox.example.com/api/dcim/racks/8/",
                "display": "RACK 11",
                "name": "RACK 11"
            },
            "position": 6,
            "face": {
                "value": "front",
                "label": "Front"
            },
            "parent_device": null,
            "status": {
                "value": "active",
                "label": "Active"
            },
            "airflow": {
                "value": "side-to-rear",
                "label": "Side to rear"
            },
            "primary_ip": null,
            "primary_ip4": null,
            "primary_ip6": null,
            "cluster": null,
            "virtual_chassis": null,
            "vc_position": null,
            "vc_priority": null,
            "comments": "",
            "local_context_data": null,
            "tags": [],
            "custom_fields": {
                "IP": "10.1.1.10",
                "snmp_communaute": "read",
                "snmp_communaute_alternate": null,
                "snmp_version": "2c",
                "snmp_port": "161",
                "ping_frequence": "60",
                "snmp_frequence": "60",
                "client": null,
                "exploitant": "Operator",
                "partenaire": null
            },
            "config_context": {},
            "created": "2022-07-12T12:21:22.461170Z",
            "last_updated": "2022-07-12T12:23:18.688284Z"
        }
    ]
}

Thank you in advance.

is it possible to scrape multiple exporters on one target?
i.e. node_exporter on port 2000, mariadb_exporter on port 2001

how can i define exporter port?

Hey Felix,

are you pushing the latest version up onto PyPi? I've noticed the build there is not the current version in git. 0.4?

Many Thanks.
Pete

Hello,

First thank you for this project, it is very useful!

I needed to add the Rack label when requesting a Device as Prometheus target.

But by looking the code I saw 2 possibilities of implementing this.
So I forked the project and created 2 branches here :

• main...geg347:netbox-plugin-prometheus-sd:adding-rack-label-device-serializer-1

• main...geg347:netbox-plugin-prometheus-sd:adding-rack-label-device-serializer-2

Also I tried to blindly implement Unit test but I was not able to try it.

Can you tell me if we can implement this, please? If yes which one of the 2 possibilities seem the best (I will create the PR based on the best branch)?
Feel free to ask for any improvements.

Thanks and have a great day!
geg347

https://demo.netbox.dev/static/docs/core-functionality/services/

I think one of the great use cases could be if netbox services could be discovered as well.

Critical services can be defined in the netbox, to be checked by blackbox exporter later on.

Hi everybody,
I'm trying Netbox 3.4.0 with netbox-plugin-prometheus-sd and I can't see any device using the /api/plugins/prometheus-sd/devices API even if I have at least 1 device.

This is the request:

GET http://netbox.local:8000/api/plugins/prometheus-sd/devices/

and this is the answer:

`
HTTP 200 OK
Allow: GET, POST, HEAD, OPTIONS
Content-Type: application/json
Vary: Accept

[]
`

But I have at least 1 device (Dell R630 in state Active)

Hey @FlxPeters.

I got this up and running and solved a few problems in testing. Ace work!

I was just testing this out and noticed that the custom fields are not coming out where the netbox-prometheus-sd does pull them out and turn the fields into labels. (I'd patched a bug there with label txt). I think it would be nice to export them as I'd based some functionality around it based on the other tool and it allows people to pass some params outbound for things like escalation ownership etc.

from the other closed issue about this - I'm just testing using tags etc, services looks like a right pain and not a fast way to turn it on.

I'll have a look at the code and see if I can figure out something and use your logic from the other application. might look at a url option for customlablels=true to turn them on and off etc.

happy to bounce a few things about.

For a years NetBox users requested for OOB IP in DCIM netbox-community/netbox#8137
Now this PR merged and will avail in NetBox 3.6 netbox-community/netbox#13013

netbox-plugin-prometheus-sd shoud have oob_ip field accordingly

How hard would it be to make a single endpoint which exposes both devices and virtual-machines?

Reason: AFAICS, currently every Prometheus scrape job needs to be duplicated, since prometheus http-sd can only scrape a single URL. If you've got a lot of rewriting rules, that can be painful.

A union endpoint would solve this, and __meta_netbox_model can be used to distinguish between the objects if necessary.

However, I can see there is much Django ORM queryset and serializer magic going on, so it may not be easy. Maybe this is relevant.