openzipkin-attic / sleuth-webmvc-example Goto Github PK

I just pulled the add-mysql-tracing. The only thing I've changed is the username and password for the mysql database. Ran it, it traces the /api endpoint but shows no mysql spans.

Does this require root access for mysql?

hello~ i use rabbitMQ in my code. And i want to use 'brave' to trace program.

i want to know how to transfer trace conttext by rabbitMQ.

is it headers?
But transferring trace context by rabbitMQ‘s headers destructive rabbitMQ‘s Headers Exchange characteristic

I test the dubbo tracing instrument and it's ok because it uses the denpendency is

<dependency>
      <groupId>com.alibaba.spring.boot</groupId>
      <artifactId>dubbo-spring-boot-starter</artifactId>
      <version>2.0.0</version>
</dependency>

but this dubbo version has not been maintained,and now offical version is

<dependency>
      <groupId>com.alibaba.boot</groupId>
      <artifactId>dubbo-spring-boot-starter</artifactId>
      <version>0.2.0</version>
</dependency>

when use this one,the brave may not adapt to it and can not trace the PRC call chain which I have tested.
Appendix:
https://mvnrepository.com/artifact/com.alibaba.boot/dubbo-spring-boot-starter/0.2.0
https://mvnrepository.com/artifact/com.alibaba.spring.boot/dubbo-spring-boot-starter/2.0.0
https://github.com/alibaba/dubbo-spring-boot-starter
https://github.com/apache/incubator-dubbo-spring-boot-project
https://github.com/apache/incubator-dubbo-spring-boot-project
https://github.com/alibaba/dubbo-spring-boot-starter

With the changes to docker containers from @abesto I started seeing the following exception when running the scripted acceptance tests:

08:52:17 
Creating docker_frontend_1 ... done

Creating zipkin            ... done

Creating docker_backend_1  ... done
the input device is not a TTY
08:52:17 Fail #1/10... will try again in [5] seconds
08:52:23 the input device is not a TTY
08:52:23 Fail #2/10... will try again in [5] seconds
08:52:29 the input device is not a TTY
08:52:29 Fail #3/10... will try again in [5] seconds
08:52:35 the input device is not a TTY
08:52:35 Fail #4/10... will try again in [5] seconds
08:52:41 the input device is not a TTY
08:52:41 Fail #5/10... will try again in [5] seconds
08:52:46 the input device is not a TTY
08:52:46 Fail #6/10... will try again in [5] seconds
08:52:52 the input device is not a TTY
08:52:52 Fail #7/10... will try again in [5] seconds
08:52:58 the input device is not a TTY
08:52:58 Fail #8/10... will try again in [5] seconds
08:53:04 the input device is not a TTY
08:53:04 Fail #9/10... will try again in [5] seconds
08:53:09 the input device is not a TTY
08:53:09 Fail #10/10... will try again in [5] seconds
08:53:09 Failed to start the app...
08:53:10 Killing docker_backend_1  ... 
08:53:10 Killing docker_frontend_1 ... 
08:53:10 Killing zipkin            ... 
08:53:11 
Killing docker_frontend_1 ... done

Killing docker_backend_1  ... done

Killing zipkin            ... done
Build step 'Execute shell' marked build as failure
08:53:14 Finished: FAILURE

Obviously, the problematic line is the input device is not a TTY. Has anybody noticed such an issue on their side?

After following the Readme.md,I spent some times finding the instrument about dubbo tracing.
it needs git checkout add-dubbo-tracing to see the codes,can we polish up the doc to make this way explicitly.if so,I can give a PR.

Currently, when I do performance troubleshooting, I start frontend and backend locally and use wrk -t4 -c128 -d100s http://localhost:8081 --latency against our docker-compose setup which includes grafana. I then look at the results.

Some of our stuff uses cloud and also docker internal networks are a bit weird. It seems like we could create and openzipkin/example-frontend and openzipkin/example-backend image which could then be deployed and tweaked as necessary. These literally could be added to our default docker-compose file for people who don't already have apps also.. or at least an additional compose file could add them.

I'm big in favor of literally using this repo vs something less used, relevant or tested. Sleuth report more concerns than many people do, and we have a lot of branches here which means we can switch to working and relevant config somewhat quickly. I want to re-use energy we already spend here (referencing this repo several times a week) vs creating something similar that people don't use today.

cc @openzipkin/core

Span span = ThreadLocalSpan.CURRENT_TRACER.next(); span is null

Using the sleuth-webmvc-example with Jetty HttpClient as per example from https://www.baeldung.com/spring-log-webclient-calls (point 4.1) for customized logging purposes. Getting the following:

1. First logging output on executing the frontend controller the logging from the RequestLogEnhancer for both the request (request.onRequestxxx) and the response (reqeust.onResponsexxx) contains no trace id in the logs
   eg: 2020-05-17 12:03:16.919 INFO [frontend,,,] 18616 --- [ @298b64f7-67]

2. After the first attempt executing the frontend controller there after it appears only the response logging has no trace id and the request logging have trace id information eg: 2020-05-17 12:10:06.562 INFO [frontend,5ec080de84f5b416f3397e3cc503c6e1,f3397e3cc503c6e1,false] 9496 --- [io-8081-exec-10]

Enabled debug and all debug logs have the trace ids in them, only the logging I implemented for Jetty HttpClient logging gives issues with point 1 and 2 above. Any help would be appreciated.

Configuration:
org.springframework.boot 2.2.7.RELEASE
springCloudVersion Hoxton.SR4
spring-cloud-starter-sleuth
spring-boot-starter-web and spring-boot-starter-webflux included

Hi! Spring Kafka instrumentation has improved since the last two example branches to include out of the box support for Spring Kafka (following spring-cloud/spring-cloud-sleuth#1001) and Kafka Streams (following spring-cloud/spring-cloud-sleuth#1365, pending 2.2.0.M1).

While this project already contains two kafka branches, these seem outdated or have a different scope:

• https://github.com/openzipkin/sleuth-webmvc-example/tree/add-kafka-tracing (duplicates Sleuth instrumentation)
• https://github.com/openzipkin/sleuth-webmvc-example/tree/kafka (span reporting over kafka)

As suggested by @adriancole in spring-cloud/spring-cloud-sleuth#1365 I've added a sample here to show Spring Kafka instrumentation and use:
master...timtebeek:add-spring-kafka-tracing

Not really sure how to target a new branch through a fork/pull request, but perhaps you can help out there. :) Once Sleuth 2.2.0.M1 is out I can provide a similar example (or expand this one) to include Kafka Streams.

Hi,

When trying the add-jms-tracing branch, i've got this issue at startup for both application (backend and frontend) :

***************************
APPLICATION FAILED TO START
***************************

Description:

The bean 'configureTracing', defined in class path resource [org/springframework/cloud/sleuth/instrument/messaging/TraceMessagingAutoConfiguration$SleuthJmsConfiguration.class], could not be registered. A bean with that name has already been defined in sleuth.webmvc.JmsTracingConfiguration and overriding is disabled.

Action:

Consider renaming one of the beans or enabling overriding by setting spring.main.allow-bean-definition-overriding=true

Adding this configuration inside application.properties makes it work :

spring.sleuth.messaging.jms.enabled=false

I was wondering if this is the way to go or maybe if sleuth.version is outdated (2.1.0.BUILD-SNAPSHOT on this branch and 2.1.2.RELEASE on master).

I have tested with no change on code, only modifications on application.properties :

• spring.activemq.broker-url to match my broker url
• spring.zipkin.baseUrl to match my zipkin url
• spring.sleuth.messaging.jms.enabled=false to turn off the error as startup

With this, i'm able :

• to send message to the queue after calling frontend api
• to observe the message reaching the queue (active mq from docker webcenter/activemq)
• to observe producer processing the message
• to observe beautiful graphic into zipkin

Also, i don't know if it is normal or not but my message get processed twice.