Your data. Anywhere you go.

New Relic for iOS or Android


Download on the App Store    Android App on Google play


New Relic Insights App for iOS


Download on the App Store


Learn more

Close icon

Shared Dashboards Contest—Add Your Queries Here!

shared-dashboards

#25

My simplest dashboard that do not require much anonymization. I just want the cool t-shirt :smiley:

(Edited for organization:)

This particular dashboard would easily show drastic changes if an authentication solution went down or if one node went offline. The application delivery controllers would stop traffic to a node that responds with 5xx error codes, so that should not show up in the graph :slight_smile:

I usually create dashboards that create a full understanding of specific situations (“why are there so few new orders in the system?”), or a dashboard that shows the performance and information flow within a solution that consists of many services that speak with each other.

This particular dashboard consists of two NRQLs (obviously);
SELECT count(*) FROM Transaction WHERE appName LIKE ‘%(Production)’ FACET httpResponseCode SINCE 30 MINUTES AGO TIMESERIES
SELECT count(host) FROM Transaction where host IN(‘host1’, ‘host2’, ‘host3’) FACET host SINCE 30 MINUTES AGO TIMESERIES


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#26

Simplicity is great, @Geir.Arne.Aasen, isn’t it?

You’re going to want to add some of your associated NRQL queries and a value statement if you want to earn that t-shirt! And, you’re right: it is a really cool t-shirt, by the way. :wink: Thanks!


#27

Infra Costs Monitoring

(even without NewRelic Infrastructure)

This dashboard helps us to monitor and optimize Infrastructure costs by visualizing NewRelic ComputeUnits and UniqueHosts. No other monitoring tool (we also have SysDig and Prometheus) or even AWS itself allows us such deep insights into the resource consumption of each single app running in our Kubernetes clusters across the world.

And the best: Even without Insights Pro the data is available for years in the past

Resource Consumption per Application

SELECT sum(apmComputeUnits) FROM NrDailyUsage FACET apmAppName LIMIT 20 WHERE `productLine` = 'APM' and `usageType` = 'Application' since 7 days ago
Application ComputeUnits consumed by each app (also works in KUBERNETES CLUSTER !!!)

Monthly runrate of NewRelic costs

SELECT SUM(apmComputeUnits) * 0.015 * 30 as NRCosts FROM NrDailyUsage WHERE `productLine` = 'APM' AND `usageType` = 'Host' FACET dateOf(timestamp) since 7 days ago
Based on Host ComputeUnits it’s possible to calculate the monthly NewRelic costs by multiplying with whatever your CU price is

Long term Application ComputeUnits usage

SELECT sum(apmComputeUnits) FROM NrDailyUsage timeseries 1 day LIMIT 40 WHERE `productLine` = 'APM' and `usageType` = 'Application' facet apmAppName since 12 months ago
APPLICATION ComputeUnits consumed by each app per day over 12 months (even without Insights Pro!!!)

Long term Host ComputeUnits usage

SELECT SUM(apmComputeUnits) as usage FROM NrDailyUsage WHERE `productLine` = 'APM' AND `usageType` = 'Host' TIMESERIES 1 day SINCE 12 months ago facet cases(where cloudZone LIKE 'ap-southeast%' as Singapore, where cloudZone LIKE 'eu-central%' as Frankfurt, where cloudZone LIKE 'eu-west%' as Dublin, where cloudZone LIKE 'ap-southeast%' or cloudZone LIKE 'eu-central%' or cloudZone LIKE 'eu-west%' as Others) LIMIT 20
HOSTS ComputeUnits provided for each AWS region over 12 months (even without Insights Pro!!!)

Kubenetes Cluster: Pods in % per app

SELECT filter(uniqueCount(host), where appName LIKE 'xxx') / uniqueCount(host) * 100 as PodPercentage from Transaction TIMESERIES since 7 days ago
Percentage of Kubernetes Pods running for single app

Kubenetes Cluster: Usage in % per app

SELECT filter(sum(apmComputeUnits), where apmAppName LIKE 'xxx') / sum(apmComputeUnits) * 100 as ClusterUsagePercentage FROM NrDailyUsage where `productLine` = 'APM' and `usageType` = 'Application' TIMESERIES 1 day since 1 month ago
Percentage of Kubernetes Cluster Usage by single app

Hosts per day

SELECT uniqueCount(cloudInstanceId) as usage FROM NrDailyUsage WHERE `productLine` = 'APM' timeseries 1 day since 12 months ago FACET cases(where cloudZone LIKE '%eu-central%' as Frankfurt, where cloudZone LIKE '%eu-west%' as Dublin, where cloudZone LIKE '%ap-%' as Singapore)
Unique Hosts seen per day in each AWS region - over 12 months (even without Insights Pro!!!)

EC2 instance types

SELECT uniquecount(agentHostname) FROM NrDailyUsage LIMIT 40 WHERE `productLine` = 'APM' and `usageType` = 'Host' and cloudZone LIKE 'ap-%' facet cloudInstanceType since 7 days ago
EC2 Instance type running in Asia AWS account

Some more tips I shared recently:

https://tech.deliveryhero.com/blog_post/newrelic-tricks-we-wish-we-would-have-known/


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#29

This is a SpringBoot microservices API dashboard - that is beautifully designed to capture the key metrics such as -

  • API Response codes.
  • API Average response time.
  • External web services that this API calls.
  • Database, Redis data dependencies this API relly on.
  • Top 5 Errors and Error rate.
  • CPU and Memory usage by each host.

My favorite NRQL to get the transaction counts for web service call this api makes (we are using CXF library).

SELECT count(library.CXF.soap.Message Name) FROM Transaction FACET library.CXF.soap.Message Name SINCE 12 HOURS AGO TIMESERIES

NRQL to get API response codes:

Blockquote

SELECT count(httpResponseCode) FROM Transaction FACET httpResponseCode SINCE yesterday WHERE appName = ‘your-app-name’

Blockquote


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#30

Some associated queries:

Applications Summary

SELECT count(*) FROM Transaction since 1 day ago FACET appName

Busiest Days

SELECT count(*) FROM Transaction since 7 day ago FACET weekdayOf(timestamp)

Response Codes

SELECT count(*) FROM Transaction since 1 day ago FACET httpResponseCode

Load Balancing

SELECT count(*) FROM Transaction since 1 day ago FACET host

Average Response Times

SELECT average(duration) FROM Transaction since 1 day ago TIMESERIES AUTO

Response Times Percentile Breakdown

SELECT percentile(duration,99,95,75,50) FROM PageView SINCE 1 day ago

Performance by Host

SELECT average(duration) FROM Transaction since 1 day ago TIMESERIES AUTO FACET host


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#31

This is my production dashboard. It is projected on a TV so I can easily see if something wrong happens.
Database connections, Java things, etc. Also, During our web server release with Jenkins, I use this dashboard to check if there are still clients connected before to bring the server out of production and apply for the release.

Everything else about the infrastructure such as storage getting full, high CPU for too long, etc, I receive alerts so there is no need for me to add them to a dashboard.

The queries are specific so I cannot share them.


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#32

We use the above dashboard on a regular basis to monitor throughput and server errors (500s) on our APIs.

500s by Endpoint
SELECT count(*) FROM Transaction WHERE response.status >= ‘500’ and errorType not in (‘BusinessRuleException’, ‘System.Reflection.TargetInvocationException’, ‘System.ArgumentException’, ‘System.ArgumentOutOfRangeException’) and databaseDuration >= 0 and appName in (‘AppName’) FACET name

Hosts making the database calls
SELECT count(*), average(databaseDuration) FROM Transaction WHERE databaseDuration >= 0 and appName in (‘AppName’) FACET host limit 100

** Slowest Transactions**
SELECT average(databaseDuration ) FROM Transaction where databaseDuration >= 0 and appName in (‘AppName’) FACET name

500s
SELECT count(*) as Errors FROM Transaction WHERE response.status >= ‘500’ and errorType not in (‘BusinessRuleException’, ‘System.Reflection.TargetInvocationException’, ‘System.ArgumentException’, ‘System.ArgumentOutOfRangeException’) and databaseDuration >= 0 and appName in (‘AppName’)

Transactions Per Server
SELECT count(*) FROM Transaction WHERE databaseDuration >= 0 and appName in (‘AppName’) FACET host TIMESERIES 1 minute since 6 hours ago

Errors by Name
SELECT count(*) FROM Transaction WHERE (response.status like ‘5%’ ) and appName in (‘Edgecast Production API Servers’) and errorType not in (‘EdgeCast.Business.BusinessRuleException’, ‘System.Reflection.TargetInvocationException’, ‘System.ArgumentException’, ‘System.ArgumentOutOfRangeException’) since 30 minutes ago FACET name


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#34

Purpose

This dashboard is for performance testing environment and helps us to monitor the app’s performance for every build. It visualizes the high level performance of all the transactions mainly response time and resource utilization.

Dashboard Details

Required products: compatible browser and New Relic license

Level of Effort: Moderate

Settings: No additional config required

Following charts have been added to the dashboard:

1. Web Transactions Response time

This chart visualizes the transaction response time of all the hits.

2. Icon Load Time

This chart visualizes the response time of an image file.

3. Throughput

This chart lists the throughput of the application.

4. CPU

This chart visualizes the current CPU load of the system under test.

5. Free Memory

This chart visualizes the free memory of the system under test.

6. Max Response Time

This chart visualizes the maximum response time of the transactions.

7. Std Deviation

This chart visualizes the standard deviation of the response time.

New Relic Insights

This query lists out the response time, time and the app.

SELECT timestamp, duration, error FROM Transaction WHERE appName = 'My Application'


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#35

A New Relic dashboard used in the business to track the below areas:

  1. Top 10 Page Visits ;
  2. Duration by all webs;
  3. How App is used;
  4. Unique accounts;
  5. User Flow;
  6. Response Status;
  7. Location of web visits;
  8. Page view histogram;
  9. Number of accounts

It’s both useful and good looking, indicating what’s happening in application, and also a beautiful piece of dashboard work, containing multiple different kind of chart types, Histograms, Stacked Area Chart, Line Chart, Pie Chart, and User Flow chart.


(a few areas are covered on purpose for security reasons)

Two interesting queries

  1. Histogram:
    SELECT histogram(duration,60,30 ) from PageView SINCE 1 day ago FACET appName limit 20

  2. A normal pie / bar chart:
    SELECT uniqueCount(session) FROM PageView FACET city SINCE 7 DAY AGO WHERE appName = ‘type your appName here’


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#36

New Relic is useful for monitoring the Login Failures and Application availability. This helps to spot the issue before our customers do.

INSIGHTS

NRQLs:
SELECT average(duration), apdex(duration, t:100000) FROM SyntheticCheck FACET monitorName
SELECT count() FROM SyntheticCheck SINCE yesterday
SELECT count(
) from SyntheticRequest
SELECT sum(duration) FROM SyntheticCheck FACET monitorName


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#37

Hi there, I am throwing Santa’s KPI Dashboard into the pool. The message is straight forward, showing that even Santa is working with measuring the KPI of hardware, software and environment to be successful. The ease of use and the value he gets out of real-time information is big to him and his customers who are having a satisfied Christmas.


#38


This dashboard is created to view the transactions that are occurring within the application server and their distribution over 3 weeks period of time. It also gives an overview on the CPU usage when these transactions occur.
Required Products: Browser, Application Server running on one or more than one host
Level of Effort: Low
This dashboard will work on the standard New Relic data models and does not require any custom attributes or events.
The various charts in the dashboard is showing

Last 7 days transaction

SELECT count(*) FROM Transaction since 7 day ago FACET weekdayOf(timestamp)

Service Transactions

SELECT count(*) FROM Transaction since 3 week ago FACET name limit 100

Hourly Comparision of Transactions

SELECT count(*) FROM Transaction since 3 weeks ago FACET hourOf(timestamp) WITH TIMEZONE ‘Asia/Kolkata’

Average Duration of Transactions

SELECT average(duration) FROM Transaction since 3 weeks ago TIMESERIES AUTO

Server Host transactions

SELECT average(duration) FROM Transaction since 3 weeks ago TIMESERIES AUTO FACET host

CPU Usage

SELECT average(cpuPercent) as ‘% CPU Usage’ FROM SystemSample SINCE 3 weeks ago


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#39


#40

Dashboard: Care Operations (Production)

This dashboard determines production care channel performance and provides end user performance of specific BUC’s

Level of effort: low
Audience: Operations and Management
Usage: Heavy

Initial Load Time Avg (today vs. 7d ago) : Todays page load time of Home page compared to 7 days ago
Prepaid Cart Session Funnel: Funnel diagram of cart sessions vs completion
Avg Page Load Time (orderconfirm): page performance of order completion
Initial Load Time by Browser Version: Load time by Browser Version
Page Load Detail by Browser Version: Page Load Detail by Browser Version

Dashboard Queries:

Initial Load Time:
SELECT average(duration) as ‘Page Load’ FROM BrowserInteraction where appName=‘Rebellion_Prod_Care’ and category = ‘Initial page load’ SINCE 1 day ago until 1 minute ago TIMESERIES auto COMPARE WITH 7 days ago

Browse and search page URL Counts:
SELECT count(*) from BrowserInteraction where appName = ‘Rebellion_Prod_Care’ facet cases(WHERE targetUrl like ‘%prepaid/browse%’ as ‘Browse’, WHERE targetUrl like ‘%search%’ as ‘Search’) since 1 day ago

Page Load Counts:
SELECT count(*) as ‘orderconfirm’ FROM BrowserInteraction where appName=‘Rebellion_Prod_Care’ where targetUrl like ‘https://care.t-mobile.com/prepaid/cart/orderconfirm’ SINCE 1 day ago COMPARE WITH 7 days ago

Cart Session Funnel:
select funnel(session, where targetUrl like ‘%prepaid/cart/landing’ as ‘Cart Landing’, where targetUrl like ‘%prepaid/cart/orderpayment’ as ‘Order Payment’, where targetUrl like ‘%prepaid/cart/orderreview’ as ‘Order Review’, where targetUrl like ‘%prepaid/cart/orderconfirm’ as ‘Order Confirm’) from BrowserInteraction where appName = ‘Rebellion_Prod_Care’ since 1 day ago

Page load detail by Browser Session:

SELECT average(duration) as ‘Avg PageLoad’, count(*) as ‘Initial Pageload Count’, average(timeToFetchStart-timeToRedirectStart) as ‘Redirect Duration’, average(timetoDomainLookupStart-timeToFetchStart) as ‘Cache Duration’,average(timeToDomainLookupEnd-timeToDomainLookupStart) as ‘DNS Duration’,average(timeToRequestStart-timeToConnectStart) as ‘TCP Duration’, average(timeToDomLoading - timeToRequestStart) as ‘Request Duration’, average(timeToLoadEventStart-timeToDomLoading) as ‘Dom Duration’, average(timeToLoadEventEnd-timeToLoadEventStart) as ‘OnLoad Duration’, average(duration-timeToLoadEventEnd) as ‘XHR Duration’ from BrowserInteraction where appName = ‘Rebellion_Prod_Care’ and category = ‘Initial page load’ facet userAgentName, userAgentVersion limit 100 since 1 day ago until 1 minute ago


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#41

We maintain 3 major APIs those are used by bunch of clients.
It is important to monitor how the throughput vs Response times of those apis behave with time.

The idea is no matter how much the throughput increases we should observe a almost constant response time. If we ever see a considerable deviation in the response times that is when we should act upon it.

Below screenshot shows such a case where the blue line represents current time and dotted yellow line shows same information one week earlier.
We can see the response times are in similar range even though the throughput values are changing a lot.

Below are throughput queries
SELECT count() AS TransactionCount FROM Transaction WHERE appName = ‘MerchantApi’ TIMESERIES COMPARE WITH 1 week ago
SELECT count(
) AS TransactionCount FROM Transaction WHERE appName = ‘WebApi’ TIMESERIES COMPARE WITH 1 week ago
SELECT count(*) AS TransactionCount FROM Transaction WHERE appName = ‘AdminApi’ TIMESERIES COMPARE WITH 1 week ago

Below are response time queries
SELECT average(duration1000) AS AvgDuration FROM Transaction WHERE appName = ‘MerchantApi’ TIMESERIES COMPARE WITH 1 week ago
SELECT average(duration
1000) AS AvgDuration FROM Transaction WHERE appName = ‘WebApi’ TIMESERIES COMPARE WITH 1 week ago
SELECT average(duration*1000) AS AvgDuration FROM Transaction WHERE appName = ‘AdminApi’ TIMESERIES COMPARE WITH 1 week ago


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#42

The below dashboard shows the appdex for 3 services we have and the response time for them. The response time also includes the past data too so that we can track any anomalies and compare them over the time.


The apdex graph you can directly copy to the dashboard from the overview. The query for the response time is below.

SELECT average(duration*1000) AS AvgDuration FROM Transaction WHERE appName = TIMESERIES COMPARE WITH 1 week ago


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#43

Dashboard: - Infrastructure Overview

This dashboard represents the servers status from an infrastructure perspective, and shows information on below 4 critical areas,

  • Alerts :- 3 Policy screens whose facets are linked to the same dashboard to enable drilling down of the most common alerts .
  • Current Server performance: CPU/Memory/Disk/Network
  • Servers approching limits
  • Infrastructure events.

The dashboard provide a high level status of the platform from a hosting team perspective.

Alerts:
Alerts events are ingested into insights via a webhook
SELECT count() FROM ALERTS FACET policy_name SINCE 1 day ago
SELECT count(
) FROM ALERTS FACET condition_name SINCE 1 day ago
SELECT timestamp, incident_id, policy_name, condition_name, details, severity FROM ALERTS SINCE 1 day ago

Servers using 20% and above of their CPU:
SELECT average(cpuPercent) AS ‘CPU %’
FROM SystemSample WHERE ec2Tag_Monitoring LIKE ‘xxxxxx’
FACET hostname where cpuPercent > 10
SINCE 5 minutes ago LIMIT 100

Servers with less than 20% free disk
SELECT average(cpuPercent) AS ‘CPU %’
FROM SystemSample WHERE ec2Tag_Monitoring LIKE ‘xxxxxx’
FACET hostname where cpuPercent > 10
SINCE 5 minutes ago LIMIT 100

The rest are standard NRQL for infrastructure


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#44

Here is anonymized dashboard for one our k8s clusters

Below are the queries:
SELECT average(cpuPercent) FROM SystemSample TIMESERIES FACET label.displayName
SELECT average(memoryUsedBytes/memoryTotalBytes*100) FROM SystemSample TIMESERIES FACET label.displayName
SELECT average(diskUsedPercent) FROM SystemSample TIMESERIES FACET label.displayName
SELECT timestamp, incident_id, policy_name, condition_name, details, severity FROM alert WHERE policy_name = ‘Engg Alert Policy’ SINCE yesterday


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#45

Hello Every One,

This Dashboard gives the information about New Relic Exceptions across the environments like Production, UAT, QC & Dev by Sprint wise.

Advantages:
1 Find Common Exceptions in all the environment.
2 Always take top errors from UAT and Give Fix before client face the same issue in production
3 Using this dash board, Find out issues before client raise CLI’s.

Below are the sample NRQL Queries for this dashboard

UAT: Sprint Start and End Date(27th Oct-02 Nov)
SELECT count(*) FROM TransactionError where appName LIKE 'UAT Appname ’ SINCE ‘2018-10-27 00:00:00’ UNTIL ‘2018-11-02 00:00:00’ FACET error.message

QC: Sprint Start and End Date(13th Oct- 26th Oct)

SELECT count(*) FROM TransactionError where appName LIKE ‘QC Appname’ SINCE ‘2018-10-13 00:00:00’ UNTIL ‘2018-10-26 00:00:00’ FACET error.message

Dev: Sprint Start and End Date(24th Sep- 12th Oct)

SELECT count(*) FROM TransactionError where appName LIKE ‘Dev’ SINCE ‘2018-09-2400:00:00’ UNTIL ‘2018-10-12 00:00:00’ FACET error.message


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters


#46

Screenshoot

Dashboard Details

Required Products: APM
Level of Effort: **Low**
This dashboard will work on the standard New Relic data models and does not require any custom attributes or events.

Dashboard: - All E-commerce Performance-Prod

This dashboard represent the e-commerces status from a performance perspective, and shows information on below:

  • Number of users on each webstore
  • Total of orders (All stores).
  • Number of errors 5xx and 4xx by store.
  • Orders by store on today
  • Page load / City
  • Funnel
  • And more

The dashboard provide a high level status of the application platform to support, development and management teams.


Vote for this Dashboard!

  • :star: :star: :star:
  • :star: :star:
  • :star:

0 voters