New Relic CLI is a command line tool which is used to operate New Relic objects (Synthetic monitors, alert policies, conditions, account users etc). You can use it easily to list/get/create/delete these objects. It can be used to backup your New Relic configuration data and restore in the future. It is easy to be used other than calling different REST API endpoints.
| Command | Subcommand | Resource | arguments | flag |
|---|---|---|---|---|
| nr | get | users | - | |
| nr | get | user | <id> | |
| nr | get | monitors | - | |
| nr | get | monitor | <id> | |
| nr | get | labels | - | |
| nr | get | labelsmonitors | <category:label> | |
| nr | get | alertspolicies | - | |
| nr | get | alertsconditions | - | |
| nr | get | alertschannels | - | |
| nr | get | dashboards | - | |
| nr | get | dashboard | <id> | |
| nr | create | monitor | - | -f <monitor_sample.json> |
| nr | create | alertspolicies | - | -f <alertspolicies_sample.json> |
| nr | create | alertsconditions | - | -f <alertsconditions_sample.json> |
| nr | create | alertschannels | - | -f <alertschannels_sample.json> |
| nr | add | alertschannels | <id> <category:label> | |
| nr | update | monitor | - | -f <monitor_sample.json> |
| nr | update | alertspolicies | - | -f <alertspolicies_sample.json> |
| nr | update | alertsconditions | - | -f <alertsconditions_sample.json> |
| nr | update | alertschannels | - | -f <alertschannels_sample.json> |
| nr | patch | monitor | - | -f <monitor_sample.json> |
| nr | delete | monitor | <id> | |
| nr | delete | alertspolicies | <id> | |
| nr | delete | alertsconditions | <id> | |
| nr | delete | alertschannels | <id> | |
| nr | delete | labelsmonitors | <id> <category:label> | |
| nr | insert | customevents | - | -f <custom_events.json> -i <New Relic insert key> -a <New Relic account ID> |
| nr | backup | monitors | - | -d <backup_folder> -r <result_file.log> |
| nr | backup | alertsconditions | - | -d <backup_folder> -r <result_file.log> |
| nr | backup | dashboards | - | -d <backup_folder> -r <result_file.log> |
| nr | restore | monitors | - | -d <monitors_folder> -f <monitor_filenames> -F <file_contains_names> -m [skip|override|clean] -r <result_file.log> |
| nr | restore | alertsconditions | - | -d <alertsconditions_folder> -f <alertscondition_filenames> -F <file_contains_names> -m [skip|override|clean] -r <result_file.log> |
| nr | restore | dashboards | - | -d <dashboards_folder> -f <dashboard_filenames> -F <file_contains_names> -m [skip|override|clean] -r <result_file.log> |
| nr | take | template | <template type name> |
- Set environment variable
NEW_RELIC_APIKEY
Define New Relic admin API key in environment by export cmd on Linux OS like this:
export NEW_RELIC_APIKEY=xxxx-xxxxxxx-xxxxx-xxxxxx
xxxx-xxxxxxx-xxxxx-xxxxxx is the New Relic admin API key. How to find the admin API key in your New Relic account, reference this doc, Activate Admin user's API key: REST API keys
- Get all users info in current New Relic account
Use nr get users command like this:
$ nr get users
ID FirstName LastName Email Role
2071178 Tom Smith [email protected] admin
2000900 Jack Xi [email protected] admin
Define the output format as JSON using -o json argument
$ nr get users -o json
{
"users": [
{
"id": 2071178,
"first_name": "Tom",
"last_name": "Smith",
"email": "[email protected]",
"role": "admin"
},
......
Define the output format as JSON using -o YAML argument
$ nr get users -o yaml
users:
- email: [email protected]
first_name: Tom
id: 2071178
last_name: Smith
role: admin
......
- Use proxy
Can configure proxy if the target machine can not directly connect to newrelic.com
export NEW_RELIC_PROXY=http://<user>:<password>@<ip>:<port>
Like:
export NEW_RELIC_PROXY=http://user1:[email protected]:3128
- Configure retries
Can configure retries for calling NewRelic REST API while some network issues, all NewRelic REST API callings in CLI would follow the retries. The default retries value is 3 if RETRIES not configure
export RETRIES=<times>
Like:
export RETRIES=5
- Return codes
The nr CLI uses exit codes, which help with scripting and confirming that a command has run successfully. For example, after you run a nr CLI command, you can retrieve its return code by running echo $? (on Windows, echo %ERRORLEVEL%). If the return code is 0, the command was successful.
A sample to use return code in shell scripts for CI/CD pipeline
To restore monitors, we check if successful by the return code, if failed, it would output the monitor file names to the log file fail-restore-monitors.log(the file name you can customize by -r argument), we continue to retry to restore all monitors that failed, the monitor names were stored in fail-restore-monitors.log and we use -F argument to tell nr what monitors we want to retry, we also add a counter, if retry times exceed 3, it would end.
shell scripts to restore monitors with retry:
#!/bin/bash
noNRKey="No NEW_RELIC_APIKEY detected."
noNRKey=${noNRKey}"\n\n"
noNRKey=${noNRKey}"Please export New Relic API key."
noNRKey=${noNRKey}"\n\n"
noNRKey=${noNRKey}"Example:\n"
noNRKey=${noNRKey}" export NEW_RELIC_APIKEY=xxx-xxxxx-xx"
noNRKey=${noNRKey}"\n"
if [ $NEW_RELIC_APIKEY"" == "" ];then
echo -e "${noNRKey}"
exit 1
fi
basepath=$(cd `dirname $0`; pwd)
${basepath}/nr restore monitors -d ${basepath}/backup_monitors_folder -r fail-restore-monitors.log
exitCode=$?""
if [ $exitCode == "0" ];then
echo ""
echo "Success, restore end."
exit 0
else
echo ""
echo "Some monitors to restore failed, begin to retry..."
fi
counter=0
while [ $exitCode != "0" ]
do
counter=`expr $counter + 1`
${basepath}/nr restore monitors -F ${basepath}/fail-restore-monitors.log -r fail-restore-monitors.log
exitCode=$?""
if [ $exitCode != "0" ];then
echo ""
echo "Some monitors to restore failed in this retry: "$counter"."
if [ $counter -ge 3 ];then
echo ""
echo "Retry 3 times, restore end."
exit 1
fi
else
echo ""
echo "After retry, no failed, restore end."
exit 0
fi
done
- nr help command
The nr help command lists the nr CLI commands and a brief description of each. Passing the -h flag to any command lists detailed help, including any aliases. For example, to see detailed help for nr get, run:
$ nr get -h
Display one or many NewRelic resources.
Usage:
nr get [command]
Available Commands:
alertschannels Display all alerts_channels.
alertsconditions Display alert conditions by alert id.
alertspolicies Display all alerts_policies.
labels Display all labels.
labelsmonitors Display monitors by label.
monitor Display a single monitor by id.
monitors Display all synthetics monitors.
user Display a single user by id.
users Display all users.
Flags:
-h, --help help for get
-o, --output string Output format. table/json/yaml are supported (default "table")
-t, --type-condition string Alert condition type. Only used for 'alertsconditions' command. all|conditions|synthetics|ext|plugin|nrql are supported (default "all")
for nr get users, run:
$ nr get users -h
Display all users.
Usage:
nr get users [flags]
Examples:
* nr get users
* nr get users -o json
* nr get users -o yaml
* nr get users -i 2102902
* nr get users -i 2102902,+801314
Flags:
-e, --email string email to filter returned result. can't specify emails
-h, --help help for users
-i, --id string user id(s) to filter returned result. use ',+' to separate ids
Global Flags:
-o, --output string Output format. table/json/yaml are supported (default "table")
-t, --type-condition string Alert condition type. Only used for 'alertsconditions' command. all|conditions|synthetics|ext|plugin|nrql are supported (default "all")
-
Golang 1.9 or 1.9+
-
Golang
dep
Installdep:
go get -u github.com/golang/dep/cmd/dep
- git clone
newrelic-cli - Enter project root folder
- Run
make deps - Run
make build
- Install
gox - Enter project root folder
- Run
gox -os "windows linux darwin" -arch "amd64"
export NEW_RELIC_APIKEY=<Your NewRelic API Key>make test
Special thanks to Huang Wei who proposed this good idea and developed the initial version.