go.d.plugin

go.d.plugin is a Netdata external plugin:

• What it does: collects metrics from databases, middleware, services, APIs, and infrastructure components using built-in collectors.
• How you use it:
  • set collector-specific jobs in config/go.d/*.conf,
  • restart Netdata and the collector jobs start automatically.
• Operational behavior:
  • runs as a separate process from Netdata core (visible in ps fax),
  • lifecycle is managed by Netdata (start/stop/restart),
  • communicates through the external plugin pipe.
• Scale model:
  • supports many collectors,
  • each collector can run multiple jobs,
  • collectors can be added/extended independently.

Required Linux capabilities

All capabilities are set automatically during Netdata installation using the official installation method.

Capability	Required by
CAP_NET_RAW	Ping
CAP_NET_ADMIN	Wireguard
CAP_DAC_READ_SEARCH	Filecheck

Available collectors

Data Collection Collectors

Name	Monitors
adaptec_raid	Adaptec Hardware RAID
activemq	ActiveMQ
ap	Wireless AP
apache	Apache
apcupsd	UPS (APC)
azure_monitor	Azure Monitor
beanstalk	Beanstalk
bind	ISC Bind
boinc	BOINC
cassandra	Cassandra
cato_networks	Cato Networks
ceph	Ceph
chrony	Chrony
clickhouse	ClickHouse
cloudwatch	Amazon CloudWatch
cockroachdb	CockroachDB
consul	Consul
coredns	CoreDNS
couchbase	Couchbase
couchdb	CouchDB
dmcache	DMCache
dnsdist	Dnsdist
dnsmasq	Dnsmasq DNS Forwarder
dnsmasq_dhcp	Dnsmasq DHCP
dns_query	DNS Query RTT
docker	Docker Engine
docker_engine	Docker Engine
dockerhub	Docker Hub
dovecot	Dovecot
elasticsearch	Elasticsearch/OpenSearch
envoy	Envoy
exim	Exim
fail2ban	Fail2Ban Jails
filecheck	Files and Directories
fluentd	Fluentd
freeradius	FreeRADIUS
gearman	Gearman
hddtemp	Disks temperature
hdfs	HDFS
hpssa	HPE Smart Array
httpcheck	Any HTTP Endpoint
icecast	Icecast
intelgpu	Intel integrated GPU
ipfs	IPFS
isc_dhcpd	ISC DHCP
jetson	NVIDIA Jetson
k8s_kubelet	Kubelet
k8s_kubeproxy	Kube-proxy
k8s_state	Kubernetes cluster state
lighttpd	Lighttpd
litespeed	Litespeed
logind	systemd-logind
logstash	Logstash
lvm	LVM logical volumes
maxscale	MaxScale
megacli	MegaCli Hardware Raid
memcached	Memcached
mongoDB	MongoDB
monit	Monit
mysql	MySQL
nats	NATS
nginx	NGINX
nginxplus	NGINX Plus
nginxunit	NGINX Unit
nginxvts	NGINX VTS
nsd	NSD (NLnet Labs)
ntpd	NTP daemon
nvidia_smi	Nvidia SMI
nvme	NVMe devices
openldap	OpenLDAP
openvpn	OpenVPN
openvpn_status_log	OpenVPN
panos	Palo Alto Networks PAN-OS
pgbouncer	PgBouncer
oracledb	Oracle DB
phpdaemon	phpDaemon
phpfpm	PHP-FPM
pihole	Pi-hole
pika	Pika
ping	Any network host
prometheus	Any Prometheus Endpoint
portcheck	Any TCP Endpoint
postgres	PostgreSQL
postfix	Postfix
powerdns	PowerDNS Authoritative Server
powerdns_recursor	PowerDNS Recursor
proxysql	ProxySQL
pulsar	Apache Pulsar
puppet	Puppet
rabbitmq	RabbitMQ
redfish	Redfish BMCs
redis	Redis
rethinkdb	RethinkDB
riakkv	Riak KV
rspamd	Rspamd
s3check	S3 Compatible Object Storage
samba	Samba
scaleio	Dell EMC ScaleIO
SNMP	SNMP
squid	Squid
squidlog	Squid
smartctl	S.M.A.R.T Storage Devices
smbios_memory	SMBIOS Memory Inventory
spigotmc	SpigotMC
storcli	Broadcom Hardware RAID
supervisord	Supervisor
systemdunits	Systemd unit state
tengine	Tengine
tomcat	Tomcat
tor	Tor
traefik	Traefik
typesense	Typesense
unbound	Unbound
upsd	UPSd (Nut)
uwsgi	uWSGI
varnish	Varnish
vcsa	vCenter Server Appliance
vernemq	VerneMQ
vsphere	VMware vCenter Server
w1sensor	1-Wire Sensors
web_log	Apache/NGINX
wireguard	WireGuard
whoisquery	Domain Expiry
x509check	Digital Certificates
yugabytedb	YugabyteDB
zfspool	ZFS Pools
zookeeper	ZooKeeper

Configuration

Edit the go.d.conf configuration file using edit-config from the Netdata config directory, which is typically at /etc/netdata.

cd /etc/netdata # Replace this path with your Netdata config directory
sudo ./edit-config go.d.conf

Configurations are written in YAML.

• plugin configuration
• specific collector configuration

Enable a collector

To enable a collector you should edit go.d.conf to uncomment the collector in question and change it from no to yes.

For example, to enable the example plugin you would need to update go.d.conf from something like:

modules:
#  example: no 

to

modules:
  example: yes

Then restart netdata for the change to take effect.

Troubleshooting

HTTP response-size errors

If a collector logs HTTP response body exceeds the 64 MiB limit, reading a response crossed its HTTP response-size limit. Gzip decompression does not bypass the limit. Depending on the collector, this can fail the collection or leave only the metrics from other successful requests. Some SDK integrations report a sanitized error instead of this exact message.

• For HTTPCheck, use a dedicated health-check endpoint with a smaller response. A body-read failure reports bad_content and a response length of zero, even without a response_match rule, and can trigger its existing alerts.
• For Prometheus endpoints, reduce the metrics exposed by the exporter or target. Collector-side selectors, relabeling and series limits do not shrink the downloaded response.
• For other API collectors, reduce the data returned by the target using its supported API controls, while preserving the response format the collector needs. If the API cannot return a smaller response, contact Netdata support with the collector name and failing API operation.

There is no option to raise this limit; increasing timeout does not change it. The limit applies while reading HTTP responses, so a parser that finishes early may accept a response without reading all of its trailing content. Some collectors have stricter limits. PAN-OS and independently managed cloud, vSphere and Kubernetes SDK clients are outside this shared limit; local-file and FastCGI inputs are also outside it.

Plugin CLI

Usage:
  orchestrator [OPTIONS] [update every]

Application Options:
  -m, --modules=    module name to run (default: all)
  -c, --config-dir= config dir to read
  -w, --watch-path= config path to watch
  -d, --debug       debug mode
  -v, --version     display the version and exit

Help Options:
  -h, --help        Show this help message

Debugging a Specific Collector

To debug a particular collector, first switch to the Netdata user:

sudo su -s /bin/bash netdata

Then run the plugin in debug mode, specifying your target collector:

# For standard installations
/usr/libexec/netdata/plugins.d/go.d.plugin -d -m <collector_name>

# For static installations (e.g., in /opt)
/opt/netdata/usr/libexec/netdata/plugins.d/go.d.plugin -d -m <collector_name>

Replace <collector_name> with the specific collector you wish to debug.

Debug mode (-d) collects data continuously and does not exit on its own — press Ctrl+C to stop it when you are done.

Output appears once per collection interval (update_every), which varies by collector (commonly 1–60 seconds). Wait at least one full interval before concluding the collector is broken.

If you see no output after waiting, the collector likely failed to connect to its target or has no configured job — check the log lines printed above for connection or autodetection errors.