Styjun

Why did the population of Bhutan drop by 70% between 2007 and 2008?

Are (c#) dictionaries an Anti Pattern?

Is there an in-universe explanation given to the senior Imperial Navy Officers as to why Darth Vader serves Emperor Palpatine?

Is it unusual for a math department not to have a mail/web server?

Why is there not a willingness from the world to step in between Pakistan and India?

To what extent should we fear giving offense?

Count the number of shortest paths to n

Videos of surgery

Federal Pacific 200a main panel problem with oversized 100a 2pole breaker

How many petaflops does it take to land on the moon? What does Artemis need with an Aitken?

Find most "academic" implementation of doubly linked list

Find feasible point in polynomial time in linear programming

Is a Centaur PC considered an animal when calculating carrying capacity for vehicles?

What is the name of this plot that has rows with two connected dots?

What stops you from using fixed income in developing countries?

Was a star-crossed lover

Does trying to charm an uncharmable creature cost a spell slot?

Does NASA use any type of office/groupware software and which is that?

Are there any to-scale diagrams of the TRAPPIST-1 system?

How do I insert two edge loops equally spaced from the edges?

Book featuring a child learning from a crowdsourced AI book

Notice period 60 days but I need to join in 45 days

Did ancient peoples ever hide their treasure behind puzzles?

Stolen MacBook should I worry about my data?

Apache response time on custom chart on Stackdriver dashboard

Stackdriver Dashboard - Cannot save a dashboard with a custom log-based metricHow are log-based metrics priced in Google Stackdriver?Want to create StackDriver dashboard from templateStackdriver custom multiline logging, time formatstackdriver change line colour on Dashboard GraphsStackdriver Dashboard there is not Share option for Charts based on custom dataCount number of GCP log entries during a specified timeMaking logs available to Stackdriver from a Custom Kubernetes docker container running Apache and PHP-FPMStackdriver Logs-Based Metrics - need sum over alignment periodDoes Stackdriver convert different time units in logs to milliseconds?

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty margin-bottom:0;

0

We would like to have a line chart (or distribution) in our custom Stackdriver dashboard with the response time returned by the apache logs. Simple line chart, respone time or latency called by others coming from structured logs.

We setup our logging agent, added structured logs to fluentd as you can see on the image it's working on the log screen.

in the httpRequest we have the latency.

 httpRequest: 
 latency: "0.081215s" 
 referer: "-" 
 requestMethod: "GET" 
 requestUrl: "/v1/call/match-in?q=spdif&fields=&limit=200&ra..." 
 responseSize: "636" 
 serverIp: "3.89.69.139" 
 status: 200 
 userAgent: "HTTPClient/1.0 (2.8.3, ruby 2.2.3 (2015-08-18))" 
 

We tried creating a custom log metric by picking the field, and setting the expression as ([0-9.]+)s to eliminate the ending 's'. Our values are in the 79ms 0.079800s not sure if we need to modify the bucket configuration.

But the chart is coming back empty:

How do we plot it as a line chart on a custom dashboard in Stackdriver?
Is there a way to plot an extracted time field from logs on the screen?

Update 1:

How can we actually check if we created a metric, that has too many time series? In the Troubleshooting section there are three cases, but how do we verify them?

Update 2:

So we found that this doesn't work with values below 1. Like we had chart where we had numbers sub second, and looks like while it's between 0 and 1 it doesn't plot.

The below image clearly shows that while the latency was below 1 it didn't plot on the distribution start, and once it was more than 1 it was plotting on the right chart.

So based on these finding what is the real work around? What me missed?

here is our config:

stackdriver

share|improve this question

edited Mar 28 at 13:06

Pentium10

asked Mar 27 at 20:21

Pentium10Pentium10

135k105105 gold badges366366 silver badges443443 bronze badges

add a comment | 

0

We would like to have a line chart (or distribution) in our custom Stackdriver dashboard with the response time returned by the apache logs. Simple line chart, respone time or latency called by others coming from structured logs.

We setup our logging agent, added structured logs to fluentd as you can see on the image it's working on the log screen.

in the httpRequest we have the latency.

 httpRequest: 
 latency: "0.081215s" 
 referer: "-" 
 requestMethod: "GET" 
 requestUrl: "/v1/call/match-in?q=spdif&fields=&limit=200&ra..." 
 responseSize: "636" 
 serverIp: "3.89.69.139" 
 status: 200 
 userAgent: "HTTPClient/1.0 (2.8.3, ruby 2.2.3 (2015-08-18))" 
 

We tried creating a custom log metric by picking the field, and setting the expression as ([0-9.]+)s to eliminate the ending 's'. Our values are in the 79ms 0.079800s not sure if we need to modify the bucket configuration.

But the chart is coming back empty:

How do we plot it as a line chart on a custom dashboard in Stackdriver?
Is there a way to plot an extracted time field from logs on the screen?

Update 1:

How can we actually check if we created a metric, that has too many time series? In the Troubleshooting section there are three cases, but how do we verify them?

Update 2:

So we found that this doesn't work with values below 1. Like we had chart where we had numbers sub second, and looks like while it's between 0 and 1 it doesn't plot.

The below image clearly shows that while the latency was below 1 it didn't plot on the distribution start, and once it was more than 1 it was plotting on the right chart.

So based on these finding what is the real work around? What me missed?

here is our config:

stackdriver

share|improve this question

edited Mar 28 at 13:06

Pentium10

asked Mar 27 at 20:21

Pentium10Pentium10

135k105105 gold badges366366 silver badges443443 bronze badges

add a comment | 

We would like to have a line chart (or distribution) in our custom Stackdriver dashboard with the response time returned by the apache logs. Simple line chart, respone time or latency called by others coming from structured logs.

We setup our logging agent, added structured logs to fluentd as you can see on the image it's working on the log screen.

in the httpRequest we have the latency.

 httpRequest: 
 latency: "0.081215s" 
 referer: "-" 
 requestMethod: "GET" 
 requestUrl: "/v1/call/match-in?q=spdif&fields=&limit=200&ra..." 
 responseSize: "636" 
 serverIp: "3.89.69.139" 
 status: 200 
 userAgent: "HTTPClient/1.0 (2.8.3, ruby 2.2.3 (2015-08-18))" 
 

We tried creating a custom log metric by picking the field, and setting the expression as ([0-9.]+)s to eliminate the ending 's'. Our values are in the 79ms 0.079800s not sure if we need to modify the bucket configuration.

But the chart is coming back empty:

How do we plot it as a line chart on a custom dashboard in Stackdriver?
Is there a way to plot an extracted time field from logs on the screen?

Update 1:

How can we actually check if we created a metric, that has too many time series? In the Troubleshooting section there are three cases, but how do we verify them?

Update 2:

So we found that this doesn't work with values below 1. Like we had chart where we had numbers sub second, and looks like while it's between 0 and 1 it doesn't plot.

The below image clearly shows that while the latency was below 1 it didn't plot on the distribution start, and once it was more than 1 it was plotting on the right chart.

So based on these finding what is the real work around? What me missed?

here is our config:

stackdriver

share|improve this question

edited Mar 28 at 13:06

Pentium10

asked Mar 27 at 20:21

Pentium10Pentium10

135k105105 gold badges366366 silver badges443443 bronze badges

 httpRequest: 
 latency: "0.081215s" 
 referer: "-" 
 requestMethod: "GET" 
 requestUrl: "/v1/call/match-in?q=spdif&fields=&limit=200&ra..." 
 responseSize: "636" 
 serverIp: "3.89.69.139" 
 status: 200 
 userAgent: "HTTPClient/1.0 (2.8.3, ruby 2.2.3 (2015-08-18))" 
 

share|improve this question

edited Mar 28 at 13:06

Pentium10

asked Mar 27 at 20:21

Pentium10Pentium10

135k105105 gold badges366366 silver badges443443 bronze badges

share|improve this question

edited Mar 28 at 13:06

Pentium10

asked Mar 27 at 20:21

Pentium10Pentium10

135k105105 gold badges366366 silver badges443443 bronze badges

asked Mar 27 at 20:21

Pentium10Pentium10

135k105105 gold badges366366 silver badges443443 bronze badges

asked Mar 27 at 20:21

Pentium10Pentium10

135k105105 gold badges366366 silver badges443443 bronze badges