| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [View Raw Code] [Original HTTPS Page] |
Every module should be compatible with python2 and python3. All third party libraries should be installed system-wide or in python_modules directory. Module configurations are written in YAML and pyYAML is required.
Every configuration file must have one of two formats:
update_every : 2 # update frequency
retries : 1 # how many failures in update() is tolerated
priority : 20000 # where it is shown on dashboard
other_var1 : bla # variables passed to module
other_var2 : alb# module defaults:
update_every : 2
retries : 1
priority : 20000
local: # job name
update_every : 5 # job update frequency
other_var1 : some_val # module specific variable
other_job:
priority : 5 # job position on dashboard
retries : 20 # job retries
other_var2 : val # module specific variableupdate_every, retries, and priority are always optional.
The following python.d modules are supported:
This module will monitor one or more apache servers depending on configuration.
Requirements:
It produces the following charts:
Needs only url to server's server-status?auto
Here is an example for 2 servers:
update_every : 10
priority : 90100
local:
url : 'http://localhost/server-status?auto'
retries : 20
remote:
url : 'http://www.apache.org/server-status?auto'
update_every : 5
retries : 4Without configuration, module attempts to connect to http://localhost/server-status?auto
Module monitors apache mod_cache log and produces only one chart:
cached responses in percent cached
Sample:
update_every : 10
priority : 120000
retries : 5
log_path : '/var/log/apache2/cache.log'If no configuration is given, module will attempt to read log file at /var/log/apache2/cache.log
Module provides server and tube level statistics:
Requirements:
Server statistics:
Per tube statistics:
Sample:
host : '127.0.0.1'
port : 11300If no configuration is given, module will attempt to connect to beanstalkd on 127.0.0.1:11300 address
Module parses bind dump file to collect real-time performance metrics
Requirements:
It produces:
Sample:
local:
named_stats_path : '/var/log/bind/named.stats'If no configuration is given, module will attempt to read named.stats file at /var/log/bind/named.stats
This module monitors task counts for the Berkely Open Infrastructure Networking Computing (BOINC) distributed computing client using the same RPC interface that the BOINC monitoring GUI does.
It provides charts tracking the total number of tasks and active tasks, as well as ones tracking each of the possible states for tasks.
BOINC requires use of a password to access it's RPC interface. You can find this password in the gui_rpc_auth.cfg file in your BOINC directory.
By default, the module will try to auto-detect the password by looking in /var/lib/boinc for this file (this is the location most Linux distributions use for a system-wide BOINC installation), so things may just work without needing configuration for the local system.
You can monitor remote systems as well:
remote:
hostname: some-host
password: some-passwordThis module monitors the precision and statistics of a local chronyd server.
It produces:
Requirements: Verify that user netdata can execute chronyc tracking. If necessary, update /etc/chrony.conf, cmdallow.
Sample:
# data collection frequency:
update_every: 1
# chrony query command:
local:
command: 'chronyc -n tracking'This module monitors the ceph cluster usage and consuption data of a server.
It produces:
Requirements:
# chmod 640 /etc/ceph/ceph.client.admin.keyringSample:
local:
config_file: '/etc/ceph/ceph.conf'
keyring_file: '/etc/ceph/ceph.client.admin.keyring'This module monitors vital statistics of a local Apache CouchDB 2.x server, including:
Sample for a local server running on port 5984:
local:
user: 'admin'
pass: 'password'
node: 'couchdb@127.0.0.1'Be sure to specify a correct admin-level username and password.
You may also need to change the node name; this should match the value of -name NODENAME in your CouchDB's etc/vm.args file. Typically this is of the form couchdb@fully.qualified.domain.name in a cluster, or couchdb@127.0.0.1 / couchdb@localhost for a single-node server.
If you want per-database statistics, these need to be added to the configuration, separated by spaces:
local:
...
databases: 'db1 db2 db3 ...'This module shows the current CPU frequency as set by the cpufreq kernel module.
Requirement: You need to have CONFIG_CPU_FREQ and (optionally) CONFIG_CPU_FREQ_STAT enabled in your kernel.
This module tries to read from one of two possible locations. On initialization, it tries to read the time_in_state files provided by cpufreq_stats. If this file does not exist, or doesn't contain valid data, it falls back to using the more inaccurate scaling_cur_freq file (which only represents the current CPU frequency, and doesn't account for any state changes which happen between updates).
It produces one chart with multiple lines (one line per core).
Sample:
sys_dir: "/sys/devices"If no configuration is given, module will search for cpufreq files in /sys/devices directory. Directory is also prefixed with NETDATA_HOST_PREFIX if specified.
This module monitors the usage of CPU idle states.
Requirement: Your kernel needs to have CONFIG_CPU_IDLE enabled.
It produces one stacked chart per CPU, showing the percentage of time spent in each state.
This module provides dns query time statistics.
Requirement:
It produces one aggregate chart or one chart per dns server, showing the query time.
Module monitor dnsdist performance and health metrics.
Following charts are drawn:
localhost:
name : 'local'
url : 'http://127.0.0.1:5053/jsonstat?command=stats'
user : 'username'
pass : 'password'
header:
X-API-Key: 'dnsdist-api-key'This module provides statistics information from dovecot server. Statistics are taken from dovecot socket by executing EXPORT global command. More information about dovecot stats can be found on project wiki page.
Requirement: Dovecot unix socket with R/W permissions for user netdata or dovecot with configured TCP/IP socket.
Module gives information with following charts:
Sample:
localtcpip:
name : 'local'
host : '127.0.0.1'
port : 24242
localsocket:
name : 'local'
socket : '/var/run/dovecot/stats'If no configuration is given, module will attempt to connect to dovecot using unix socket localized in /var/run/dovecot/stats
Module monitor elasticsearch performance and health metrics
It produces:
Sample:
local:
host : 'ipaddress' # Server ip address or hostname
port : 'password' # Port on which elasticsearch listed
cluster_health : True/False # Calls to cluster health elasticsearch API. Enabled by default.
cluster_stats : True/False # Calls to cluster stats elasticsearch API. Enabled by default.If no configuration is given, module will fail to run.
Simple module executing exim -bpc to grab exim queue. This command can take a lot of time to finish its execution thus it is not recommended to run it every second.
It produces only one chart:
Configuration is not needed.
Module monitor fail2ban log file to show all bans for all active jails
Requirements:
It produces one chart with multiple lines (one line per jail)
Sample:
local:
log_path: '/var/log/fail2ban.log'
conf_path: '/etc/fail2ban/jail.local'
exclude: 'dropbear apache'If no configuration is given, module will attempt to read log file at /var/log/fail2ban.log and conf file at /etc/fail2ban/jail.local. If conf file is not found default jail is ssh.
Uses the radclient command to provide freeradius statistics. It is not recommended to run it every second.
It produces:
Sample:
local:
host : 'localhost'
port : '18121'
secret : 'adminsecret'
acct : False # Freeradius accounting statistics.
proxy_auth : False # Freeradius proxy authentication statistics.
proxy_acct : False # Freeradius proxy accounting statistics.Freeradius server configuration:
The configuration for the status server is automatically created in the sites-available directory. By default, server is enabled and can be queried from every client. FreeRADIUS will only respond to status-server messages, if the status-server virtual server has been enabled.
To do this, create a link from the sites-enabled directory to the status file in the sites-available directory:
and restart/reload your FREERADIUS server.
The go_expvar module can monitor any Go application that exposes its metrics with the use of expvar package from the Go standard library.
go_expvar produces charts for Go runtime memory statistics and optionally any number of custom charts. Please see the wiki page for more info.
For the memory statistics, it produces the following charts:
Please see the wiki page for detailed info about module configuration.
Module monitors frontend and backend metrics such as bytes in, bytes out, sessions current, sessions in queue current. And health metrics such as backend servers status (server check should be used).
Plugin can obtain data from url OR unix socket.
Requirement: Socket MUST be readable AND writable by netdata user.
It produces:
Sample:
via_url:
user : 'username' # ONLY IF stats auth is used
pass : 'password' # # ONLY IF stats auth is used
url : 'http://ip.address:port/url;csv;norefresh'OR
via_socket:
socket : 'path/to/haproxy/sock'If no configuration is given, module will fail to run.
Module monitors disk temperatures from one or more hddtemp daemons.
Requirement: Running hddtemp in daemonized mode with access on tcp port
It produces one chart Temperature with dynamic number of dimensions (one per disk)
Sample:
update_every: 3
host: "127.0.0.1"
port: 7634If no configuration is given, module will attempt to connect to hddtemp daemon on 127.0.0.1:7634 address
Module monitors remote http server for availability and response time.
Following charts are drawn per job:
Sample configuration and their default values.
server:
url: 'http://host:port/path' # required
status_accepted: # optional
- 200
timeout: 1 # optional, supports decimals (e.g. 0.2)
update_every: 3 # optional
regex: 'REGULAR_EXPRESSION' # optional, see https://docs.python.org/3/howto/regex.html
redirect: yes # optionalThis module will monitor number of listeners for active sources.
Requirements:
It produces the following charts:
Needs only url to server's /status-json.xsl
Here is an example for remote server:
remote:
url : 'http://1.2.3.4:8443/status-json.xsl'Without configuration, module attempts to connect to http://localhost:8443/status-json.xsl
Module monitors IPFS basic information.
Only url to IPFS server is needed.
Sample:
localhost:
name : 'local'
url : 'http://localhost:5001'Module monitor leases database to show all active leases for given pools.
Requirements:
It produces:
Sample:
local:
leases_path : '/var/lib/dhcp/dhcpd.leases'
pools : '192.168.3.0/24 192.168.4.0/24 192.168.5.0/24'In case of python2 you need to install py2-ipaddress to make plugin work. The module will not work If no configuration is given.
Module monitor litespeed web server performance metrics.
It produces:
local:
path : 'PATH'If no configuration is given, module will use "/tmp/lshttpd/".
THis module monitors active sessions, users, and seats tracked by systemd-logind or elogind.
It provides the following charts:
This module needs no configuration. Just make sure the netdata user can run the loginctl command and get a session list without having to specify a path.
This will work with any command that can output data in the exact same format as loginctl list-sessions --no-legend. If you have some other command you want to use that outputs data in this format, you can specify it using the command key like so:
command: '/path/to/other/command'This module's ability to track logins is dependent on what PAM services are configured to register sessions with logind. In particular, for most systems, it will only track TTY logins, local desktop logins, and logins through remote shell connections.
The users chart counts usernames not UID's. This is potentially important in configurations where multiple users have the same UID.
The users chart counts any given user name up to once for each type of login. So if the same user has a graphical and a console login on a system, they will show up once in the graphical count, and once in the console count.
Because the data collection process is rather expensive, this plugin is currently disabled by default, and needs to be explicitly enabled in /etc/netdata/python.d.conf before it will run.
Module monitor /proc/mdstat
It produces:
Health Number of failed disks in every array (aggregate chart).
Disks stats
No configuration is needed.
Module collects adapter, physical drives and battery stats.
Requirements:
To grab stats it executes:
It produces:
Adapter State
Physical Drives Media Errors
Physical Drives Predictive Failures
Battery Relative State of Charge
Battery Cycle Count
Battery stats disabled by default in the module configuration file.
Memcached monitoring module. Data grabbed from stats interface.
Sample:
localtcpip:
name : 'local'
host : '127.0.0.1'
port : 24242If no configuration is given, module will attempt to connect to memcached instance on 127.0.0.1:11211 address.
Module monitor mongodb performance and health metrics
Requirements:
You need to install it manually.
Number of charts depends on mongodb version, storage engine and other features (replication):
Sample:
local:
name : 'local'
host : '127.0.0.1'
port : 27017
user : 'netdata'
pass : 'netdata'
If no configuration is given, module will attempt to connect to mongodb daemon on 127.0.0.1:27017 address
Monit monitoring module. Data is grabbed from stats XML interface (exsists for a long time, but not mentioned in official documentation). Mostly this plugin shows statuses of monit targets, i.e. statuses of specified checks.
Sample:
local:
name : 'local'
url : 'http://localhost:2812'
user: : admin
pass: : monitIf no configuration is given, module will attempt to connect to monit as http://localhost:2812.
Module monitors one or more mysql servers
Requirements:
It will produce following charts (if data is available):
You can provide, per server, the following:
Here is an example for 3 servers:
update_every : 10
priority : 90100
retries : 5
local:
'my.cnf' : '/etc/mysql/my.cnf'
priority : 90000
local_2:
user : 'root'
pass : 'blablablabla'
socket : '/var/run/mysqld/mysqld.sock'
update_every : 1
remote:
user : 'admin'
pass : 'bla'
host : 'example.org'
port : 9000
retries : 20If no configuration is given, module will attempt to connect to mysql server via unix socket at /var/run/mysqld/mysqld.sock without password and with username root
This module will monitor one or more nginx servers depending on configuration. Servers can be either local or remote.
Requirements:
Example nginx configuration can be found in 'python.d/nginx.conf'
It produces following charts:
Needs only url to server's stub_status
Here is an example for local server:
update_every : 10
priority : 90100
local:
url : 'http://localhost/stub_status'
retries : 10Without configuration, module attempts to connect to http://localhost/stub_status
This module will monitor one or more nginx_plus servers depending on configuration. Servers can be either local or remote.
Example nginx_plus configuration can be found in 'python.d/nginx_plus.conf'
It produces following charts:
For every server zone:
For every upstream:
For every cache:
Needs only url to server's status
Here is an example for local server:
local:
url : 'http://localhost/status'Without configuration, module fail to start.
Module uses the nsd-control stats_noreset command to provide nsd statistics.
Requirements:
It produces:
Configuration is not needed.
Module monitors the system variables of the local ntpd daemon (optional incl. variables of the polled peers) using the NTP Control Message Protocol via UDP socket, similar to ntpq, the standard NTP query program.
Requirements:
# Local users may interrogate the ntp server more closely. restrict 127.0.0.1 restrict ::1
It produces:
configuration
Sample:
update_every: 10
host: 'localhost'
port: '123'
show_peers: yes
# hide peers with source address in ranges 127.0.0.0/8 and 192.168.0.0/16
peer_filter: '(127\..*)|(192\.168\..*)'
# check for new/changed peers every 60 updates
peer_rescan: 60Sample (multiple jobs):
Note: ntp.conf on the host otherhost must be configured to allow queries from our local host by including a line like restrict <IP> nomodify notrap nopeer.
local:
host: 'localhost'
otherhost:
host: 'otherhost'If no configuration is given, module will attempt to connect to ntpd on ::1:123 or 127.0.0.1:123 and show charts for the systemvars. Use show_peers: yes to also show the charts for configured peers. Local peers in the range 127.0.0.0/8 are hidden by default, use peer_filter: '' to show all peers.
Module monitor openvpn-status log file.
Requirements:
If you are running multiple OpenVPN instances out of the same directory, MAKE SURE TO EDIT DIRECTIVES which create output files so that multiple instances do not overwrite each other's output files.
Make sure NETDATA USER CAN READ openvpn-status.log
Update_every interval MUST MATCH interval on which OpenVPN writes operational status to log file.
It produces:
Sample:
default
log_path : '/var/log/openvpn-status.log'This module will monitor one or more php-fpm instances depending on configuration.
Requirements:
It produces following charts:
Needs only url to server's status
Here is an example for local instance:
update_every : 3
priority : 90100
local:
url : 'http://localhost/status'
retries : 10Without configuration, module attempts to connect to http://localhost/status
Module monitors a remote TCP service.
Following charts are drawn per host:
server:
host: 'dns or ip' # required
port: 22 # required
timeout: 1 # optional
update_every: 1 # optionalSimple module executing postfix -p to grab postfix queue.
It produces only two charts:
Configuration is not needed.
Module monitors one or more postgres servers.
Requirements:
Following charts are drawn:
socket:
name : 'socket'
user : 'postgres'
database : 'postgres'
tcp:
name : 'tcp'
user : 'postgres'
database : 'postgres'
host : 'localhost'
port : 5432When no configuration file is found, module tries to connect to TCP/IP socket: localhost:5432.
Module monitor powerdns performance and health metrics.
Powerdns charts:
Powerdns Recursor charts:
local:
name : 'local'
url : 'http://127.0.0.1:8081/api/v1/servers/localhost/statistics'
header :
X-API-Key: 'change_me'Monitor status of Puppet Server and Puppet DB.
Following charts are drawn:
puppetdb:
url: 'https://fqdn.example.com:8081'
tls_cert_file: /path/to/client.crt
tls_key_file: /path/to/client.key
autodetection_retry: 1
retries: 3600
puppetserver:
url: 'https://fqdn.example.com:8140'
autodetection_retry: 1
retries: 3600When no configuration is given then https://fqdn.example.com:8140 is tried without any retries.
Module monitor rabbitmq performance and health metrics.
Following charts are drawn:
socket:
name : 'local'
host : '127.0.0.1'
port : 15672
user : 'guest'
pass : 'guest'
When no configuration file is found, module tries to connect to: localhost:15672.
Get INFO data from redis instance.
Following charts are drawn:
socket:
name : 'local'
socket : '/var/lib/redis/redis.sock'
localhost:
name : 'local'
host : 'localhost'
port : 6379When no configuration file is found, module tries to connect to TCP/IP socket: localhost:6379.
Performance metrics of Samba file sharing.
It produces the following charts:
Requires that smbd has been compiled with profiling enabled. Also required that smbd was started either with the -P 1 option or inside smb.conf using smbd profiling level.
This plugin uses smbstatus -P which can only be executed by root. It uses sudo and assumes that it is configured such that the netdata user can execute smbstatus as root without password.
For example:
netdata ALL=(ALL) NOPASSWD: /usr/bin/smbstatus -P
update_every : 5 # update frequencySystem sensors information.
Charts are created dynamically.
For detailed configuration information please read sensors.conf file.
There have been reports from users that on certain servers, ACPI ring buffer errors are printed by the kernel (dmesg) when ACPI sensors are being accessed. We are tracking such cases in issue #827. Please join this discussion for help.
This module does some really basic monitoring for Spigot Minecraft servers.
It provides two charts, one tracking server-side ticks-per-second in 1, 5 and 15 minute averages, and one tracking the number of currently active users.
This is not compatible with Spigot plugins which change the format of the data returned by the tps or list console commands.
host: localhost
port: 25575
password: passBy default, a connection to port 25575 on the local system is attempted with an empty password.
This module will monitor one or more Java Spring-boot applications depending on configuration.
It produces following charts:
Please see the Monitoring Java Spring Boot Applications page for detailed info about module configuration.
This module will monitor one or more squid instances depending on configuration.
It produces following charts:
priority : 50000
local:
request : 'cache_object://localhost:3128/counters'
host : 'localhost'
port : 3128Without any configuration module will try to autodetect where squid presents its counters data
Module monitor smartd log files to collect HDD/SSD S.M.A.R.T attributes.
It produces following charts (you can add additional attributes in the module configuration file):
Read Error Rate attribute 1
Start/Stop Count attribute 4
Reallocated Sectors Count attribute 5
Seek Error Rate attribute 7
Power-On Hours Count attribute 9
Power Cycle Count attribute 12
Load/Unload Cycles attribute 193
Temperature attribute 194
Current Pending Sectors attribute 197
Off-Line Uncorrectable attribute 198
Write Error Rate attribute 200
local:
log_path : '/var/log/smartd/'If no configuration is given, module will attempt to read log files in /var/log/smartd/ directory.
Present tomcat containers memory utilization.
Charts:
localhost:
name : 'local'
url : 'http://127.0.0.1:8080/manager/status?XML=true'
user : 'tomcat_username'
pass : 'secret_tomcat_password'Without configuration, module attempts to connect to http://localhost:8080/manager/status?XML=true, without any credentials. So it will probably fail.
Module uses the health API to provide statistics.
It produces:
Detailed Response Codes requests/s (number of responses for each response code family individually)
Requests/s
Average response time
Average response time per iteration
Uptime
Needs only url to server's health
Here is an example for local server:
update_every : 1
priority : 60000
local:
url : 'http://localhost:8080/health'
retries : 10Without configuration, module attempts to connect to http://localhost:8080/health.
Monitoring uses the remote control interface to fetch statistics.
Provides the following charts:
If extended stats are enabled, also provides:
Unbound must be manually configured to enable the remote-control protocol. Check the Unbound documentation for info on how to do this. Additionally, if you want to take advantage of the autodetection this plugin offers, you will need to make sure your unbound.conf file only uses spaces for indentation (the default config shipped by most distributions uses tabs instead of spaces).
Once you have the Unbound control protocol enabled, you need to make sure that either the certificate and key are readable by Netdata (if you're using the regular control interface), or that the socket is accessible to Netdata (if you're using a UNIX socket for the contorl interface).
By default, for the local system, everything can be auto-detected assuming Unbound is configured correctly and has been told to listen on the loopback interface or a UNIX socket. This is done by looking up info in the Unbound config file specified by the ubconf key.
To enable extended stats for a given job, add extended: yes to the definition.
You can also enable per-thread charts for a given job by adding per_thread: yes to the definition. Note that the numbe rof threads is only checked on startup.
A basic local configuration with extended statistics and per-thread charts looks like this:
local:
ubconf: /etc/unbound/unbound.conf
extended: yes
per_thread: yesWhile it's a bit more complicated to set up correctly, it is recommended that you use a UNIX socket as it provides far better performance.
Module uses the varnishstat command to provide varnish cache statistics.
It produces:
No configuration is needed.
Data from 1-Wire sensors. On Linux these are supported by the wire, w1_gpio, and w1_therm modules. Currently temperature sensors are supported and automatically detected.
Charts are created dynamically based on the number of detected sensors.
For detailed configuration information please read w1sensor.conf file.
Tails the apache/nginx/lighttpd/gunicorn log files to collect real-time web-server statistics.
It produces following charts:
Detailed Response Codes requests/s (number of responses for each response code family individually)
Bandwidth KB/s
Request per url requests/s (configured by user)
Http Methods requests/s (requests per http method)
Http Versions requests/s (requests per http version)
IP protocols requests/s (requests per ip protocol version)
Curent Poll Unique Client IPs unique ips/s (unique client IPs per data collection iteration)
All Time Unique Client IPs unique ips/s (unique client IPs since the last restart of netdata)
nginx_log:
name : 'nginx_log'
path : '/var/log/nginx/access.log'
apache_log:
name : 'apache_log'
path : '/var/log/apache/other_vhosts_access.log'
categories:
cacti : 'cacti.*'
observium : 'observium'Module has preconfigured jobs for nginx, apache and gunicorn on various distros.
| Back | FazBrowse Home | New Git URL |