kieker-users — Kieker user list

Re: [Kieker-users] Issues with setting up Kieker in Spring PetClinic REST Maven project

[Dr. D. G. R. <dav...@da...>]

Hi Syed,

Thanks for your question.

In general, Kieker 1.16 is not something we plan to release; we are 
currently working on the 2.0.0 release (as a Snapshot), and 1.15.2 is 
the latest stable release. For your use case, I'd recommend to use the 
stable 1.15.2 release (and in this case, you do not need to add the 
repository).

For monitoring with Kieker, you'll need to insert the probes and define 
the configuration via kieker.monitoring.properties. While it seems that 
you've done the latter correctly, it seems like you configured the 
aop.xml for the instrumentation, but didn't define the  javaagent (which 
is necessary for adding the instrumentation; adding the jar as 
dependency if not sufficient).

To do so, you'll need to define something like 
-javaagent:$PATH_TO_KIEKER, which could be 
-javaagent:~/.m2/repository/net/kieker-monitoring/kieker/1.15.2/kieker-1.15.2-aspectj.jar 
(or, if you're starting in an IDE, you'll need to pass the path to the 
AspectJ jar).

If adding the -javaagent this way doesn't work for you, the easiest 
solution would be that you send your full configuration. You could 
therefore fork the PetClinic, make your changes for instrumentation, and 
share your fork in this mailing list.

Another option would be to use Kieker source instrumentation 
(https://github.com/DaGeRe/kieker-source-instrumentation-dagere/), which 
reduces the overhead, but changes the source code of your application 
(described in Reichelt, David Georg, Stefan Kühne, and Wilhelm 
Hasselbring. "Towards Solving the Challenge of Minimal Overhead 
Monitoring." /Companion of the 2023 ACM/SPEC International Conference on 
Performance Engineering/. 2023.)

If you have any further questions, don't hesitate to ask.

Best regards,
David

Am 15.06.23 um 08:20 schrieb Syed Ghulam Mustufa:
>
> Good Morning,
>
> I am trying Kieker Monitoring Framework as part of my masters thesis. 
> Basically, I need to monitor the Spring PetClinic REST application 
> using Kieker Monitoring Framework as part of my current task. I am 
> trying for weeks now but I cannot get pass a specific point. Here is 
> what I am trying to do:
>
> In my pom.xml file, I have added the following kieker dependencies:
>
> <dependency>
>
> <groupId>net.kieker-monitoring</groupId>
>
> <artifactId>kieker</artifactId>
>
> <version>1.16-SNAPSHOT</version>
>
> <classifier>aspectj</classifier>
>
> </dependency>
> and a repository:
>
> <repository>
>
> <id>sonatype.oss.snapshots</id>
>
> <url>https://oss.sonatype.org/content/repositories/snapshots/</url>
>
> </repository>
>
> I have setup kieker.monitoring.properties and aop.xml files under 
> classpath => src/main/resources/META-INF folder since kieker tries to 
> look under this folder.
>
> I have also setup aop.xml and added different pointcuts and aspects 
> for probes that I am using in my application.
>
> But I am unable to create logs or see kieker debug logs when I try to 
> run the application using the IDE => Run as => Java Application option.
>
> If I try to add a code to enable monitoring at the start of the 
> application, I can see KIEKER .dat and .map files created but I can 
> ONLY see KIEKER SINGLETON logs created when the application starts. I 
> do not see any probe related logs. If I see the application logs, I 
> see that the Monitoring Controller is enabled BUT ProbeController is 
> disabled.
>
> I cannot get past this issue and there is no such documentation 
> available in Kieker documentation. Can you please help me understand 
> what is going wrong? It would be much appreciated since I am stuck on 
> this for weeks now.
>
> Thank you and kind regards,
>
>
>
> _______________________________________________
> Kieker-users mailing list
> Kie...@li...
> https://lists.sourceforge.net/lists/listinfo/kieker-users

[Kieker-users] Issues with setting up Kieker in Spring PetClinic REST Maven project

[Syed G. M. <sye...@st...>]

Good Morning,

I am trying Kieker Monitoring Framework as part of my masters thesis. Basically, I need to monitor the Spring PetClinic REST application using Kieker Monitoring Framework as part of my current task. I am trying for weeks now but I cannot get pass a specific point. Here is what I am trying to do:

In my pom.xml file, I have added the following kieker dependencies:
<dependency>
<groupId>net.kieker-monitoring</groupId>
                <artifactId>kieker</artifactId>
                <version>1.16-SNAPSHOT</version>
                <classifier>aspectj</classifier>
</dependency>
and a repository:
<repository>
                <id>sonatype.oss.snapshots</id>
                <url>https://oss.sonatype.org/content/repositories/snapshots/</url>
</repository>


I have setup kieker.monitoring.properties and aop.xml files under classpath => src/main/resources/META-INF folder since kieker tries to look under this folder.
I have also setup aop.xml and added different pointcuts and aspects for probes that I am using in my application.

But I am unable to create logs or see kieker debug logs when I try to run the application using the IDE => Run as => Java Application option.

If I try to add a code to enable monitoring at the start of the application, I can see KIEKER .dat and .map files created but I can ONLY see KIEKER SINGLETON logs created when the application starts. I do not see any probe related logs. If I see the application logs, I see that the Monitoring Controller is enabled BUT ProbeController is disabled.

I cannot get past this issue and there is no such documentation available in Kieker documentation. Can you please help me understand what is going wrong? It would be much appreciated since I am stuck on this for weeks now.

Thank you and kind regards,

[Kieker-users] SLF4J Logging when using Kieker agent

[Simon E. <sim...@un...>]

Dear Kieker team,

we are currently using Kieker 1.15 with AspectJ as a Java agent to  
monitor the TeaStore (https://github.com/DescartesResearch/TeaStore).

The tracing seems to work fine, but upon startup, we get the following  
message:

    SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder".
    SLF4J: Defaulting to no-operation (NOP) logger implementation
    SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for  
further details.

We are currently loading the Kieker agent by adding the following java option:

    export JAVA_OPTS="-javaagent:/kieker/agent/kieker-1.15-aspectj.jar"

Based on my understanding, this error message is caused by the  
inclusion of slf4j-api, without an underlying implementation. Do you  
have any advice on how to provide the underlying logger when using  
Kieker as an agent? Or maybe we are doing this entirely wrong?

Best regards,
Simon


------------------------------------------------------------------------------------------
M.Sc. Simon Eismann
Doctoral Researcher

Chair of Software Engineering
Department of Computer Science
University of Würzburg

Am Hubland, Informatikgebäude (M2), Room 107/108
97074 Würzburg, Germany

Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
http://go.uni-wuerzburg.de/eismann/
------------------------------------------------------------------------------------------

[Kieker-users] Kieker 1.15 released

[Andre v. H. <and...@un...>]

Hi all,

today, we released version 1.15 of our Kieker framework for application
performance monitoring and dynamic software analysis. As usual, the 
release is available for download at 
https://kieker-monitoring.net/download/.

- New features (selection)
   - Migration of analysis stages to the TeeTime-based Kieker analysis.
   - Added new analysis stages from the iObserve research project
- Improvements and refactorings
   - Improved AspectJ probes, e.g., by introducing before/after advices
     for cflow pointcuts, OpenJDK 11 support
   - Refactored the file system writer
- Build scripts and infrastructure
   - Replaced additional local Jar dependencies by using the Maven
     dependencies
   - Converted Kieker tools to multi-project builds
   - Use of GitHub actions to test common platforms (Linux, Mac, Windows)
     and Java versions
- Documentation
   - Migrated the deprecated LaTeX-based user guide to Read The Docs
     (https://kieker-monitoring.readthedocs.io/en/latest/)

-- 
Dr.-Ing. Andre van Hoorn

University of Hamburg, Department of Informatics
Software Engineering and Construction Methods

and...@un...

Re: [Kieker-users] Using Kieker with Java 11

[Reiner J. <rei...@em...>]

Hi Simon,

thanks for your reply. When you are done with your setup, it would be
nice when you could shortly describe it so that we can add this
information to the documentation on Kieker. Or you could make the
contribution yourself and create a pull request.

Kind regards
Reiner

On Thu, 12 Aug 2021 08:04:03 +0000
Simon Eismann <sim...@un...> wrote:

> Hi Reiner,
> 
> thank you for your support, the complete log helped reveal the issue.
> We were missing some dependencies while compiling Kieker which caused
> the AMQP writer to fail. After fixing this, it works like a charm :)
> 
> Best regards,
> Simon
> 
> Quoting Reiner Jung <rei...@em...>:
> 
> > Hi Simon, André, David,
> >
> > I had a look into the source code and it seems the a writer is not
> > instantiated correctly. Maybe due to a missing property or a missing
> > value. There should be an error message somewhere in the log.
> >
> > I would suggest that you try to create a complete log with
> > your-program > log 2>&1
> >
> > That should collect all output. Then look for on of the following
> > messages:
> > - Class '{}' has to implement '{}'
> > - {}: Class '{}' not found",
> > - {}: Class '{}' has to implement a (public) constructor that
> > accepts a single Configuration
> > - {}: Failed to load class for name '{}'
> >
> > Also you could look into your Kieker configuration file for the
> > following setting:
> > kieker.monitoring.writer.collector.ChunkingCollector.writer
> >
> > You can use any Class implementing the IRawDataWriter interface.
> > Depending on the selected writer additional parameters are required.
> >
> > @andre @david For Kieker, we should better fail early when a
> > parameter is missing and report that accordingly.
> >
> > Kind regards
> > Reiner
> >
> > On Wed, 04 Aug 2021 11:02:41 +0000
> > Simon Eismann <sim...@un...> wrote:
> >  
> >> Hi Andre,
> >>
> >> as I'm not sure which version we used as I can't find the branch
> >> anymore, we recompiled it with the current master branch, but
> >> unfortunately, this did not solve the issue. I've attached a
> >> screenshot of the error message we are currently getting.
> >>
> >> Best regards,
> >> Simon
> >>
> >> Quoting Andre van Hoorn <and...@is...>:
> >>  
> >> > Hi Simon,
> >> >
> >> > thanks for reaching out to us.
> >> >
> >> > Which version did you use? In the past weeks, we have conducted a
> >> > bunch of fixes (also related to AspectJ and Java 11) that can be
> >> > found in the master branch. The current master branch basically
> >> > reflects the next Kieker release to be released very soon.
> >> >
> >> > I am CC'ing Reiner and David who worked on several issues in the
> >> > last weeks.
> >> >
> >> > Best, André
> >> >
> >> > Am 29.07.21 um 20:52 schrieb Simon Eismann:  
> >> >> Dear kieker-users,
> >> >>
> >> >> we have been working on migrating the teastore project
> >> >> (https://github.com/DescartesResearch/TeaStore) to Java 11 and
> >> >> we are running into some issues with Kieker and Java 11.
> >> >>
> >> >> We compiled Kieker from a pull request for an AspectJ/Java11 fix
> >> >> (at the time of writing I can not find this pull request
> >> >> anymore, I guess it?) and are using the AMQP writer to collect
> >> >> the messages. After the migration to Java 11, the kieker client
> >> >> does no longer connect to the AMQP server at all. The service
> >> >> logs show a NullPointerException at ChunkWriterTask.java:111 in
> >> >> kieker-monitoring, which seems to originate from its writer
> >> >> being null.
> >> >>
> >> >> Does anybody know what we might be doing wrong here?
> >> >>
> >> >> I've also included instructions on how to reproduce the issue
> >> >> below.
> >> >>
> >> >> Best regards,
> >> >> Simon Eismann
> >> >>
> >> >> ## How to reproduce
> >> >>
> >> >> Create docker-compose_kieker.yaml:
> >> >>
> >> >> ```
> >> >> version: '3'
> >> >> services:
> >> >>   rabbitmq:
> >> >>     image: descartesresearch/teastore-kieker-rabbitmq
> >> >>     expose:
> >> >>       - "5672"
> >> >>     ports:
> >> >>       - "8081:8080"
> >> >>       - "5672:5672"
> >> >>       - "15672:15672"
> >> >>   registry:
> >> >>     image: simont1105/teastore-registry
> >> >>     expose:
> >> >>       - "8080"
> >> >>   db:
> >> >>     image: descartesresearch/teastore-db
> >> >>     expose:
> >> >>       - "3306"
> >> >>     ports:
> >> >>       - "3306:3306"
> >> >>   persistence:
> >> >>     image: simont1105/teastore-persistence
> >> >>     expose:
> >> >>       - "8080"
> >> >>     environment:
> >> >>       HOST_NAME: "persistence"
> >> >>       REGISTRY_HOST: "registry"
> >> >>       DB_HOST: "db"
> >> >>       DB_PORT: "3306"
> >> >>       RABBITMQ_HOST: "rabbitmq"
> >> >>   auth:
> >> >>     image: simont1105/teastore-auth
> >> >>     expose:
> >> >>       - "8080"
> >> >>     environment:
> >> >>       HOST_NAME: "auth"
> >> >>       REGISTRY_HOST: "registry"
> >> >>       RABBITMQ_HOST: "rabbitmq"
> >> >>   image:
> >> >>     image: simont1105/teastore-image
> >> >>     expose:
> >> >>       - "8080"
> >> >>     environment:
> >> >>       HOST_NAME: "image"
> >> >>       REGISTRY_HOST: "registry"
> >> >>       RABBITMQ_HOST: "rabbitmq"
> >> >>   recommender:
> >> >>     image: simont1105/teastore-recommender
> >> >>     expose:
> >> >>       - "8080"
> >> >>     environment:
> >> >>       HOST_NAME: "recommender"
> >> >>       REGISTRY_HOST: "registry"
> >> >>       RABBITMQ_HOST: "rabbitmq"
> >> >>   webui:
> >> >>     image: simont1105/teastore-webui
> >> >>     environment:
> >> >>       HOST_NAME: "webui"
> >> >>       REGISTRY_HOST: "registry"
> >> >>       RABBITMQ_HOST: "rabbitmq"
> >> >>     ports:
> >> >>       - "8080:8080"
> >> >> ```
> >> >>
> >> >> Then execute:
> >> >>
> >> >> ```shell
> >> >> docker-compose -f docker-compose_kieker.yaml up -d
> >> >> ```
> >> >>
> >> >> After startup, open http://localhost:8080 (TeaStore),
> >> >> http://localhost:15672 (RabbitMQ management, credentials:
> >> >> admin:nimda) and http://localhost:8081 (RabbitMQ data) in
> >> >> browser.
> >> >>
> >> >> ## Building the TeaStore
> >> >>
> >> >> Requirements: openjdk11 and maven >3.6, docker and
> >> >> docker-compose.
> >> >>
> >> >> ```shell
> >> >> mvn clean install -Dmaven.test.skip=true
> >> >> ./examples/docker_build.sh -p -r simont1105 -v java11
> >> >> # IMPORTANT: change simont1105 to your your docker repo in this
> >> >> and the docker-compose file!
> >> >> ```
> >> >>
> >> >> This would allow to try other kieker packages, just replace all
> >> >> kieker 1.15 jars with new ones of the same kind.
> >> >>
> >> >>
> >> >>    
> >> ------------------------------------------------------------------------------------------
> >>  
> >> >> M.Sc. Simon Eismann
> >> >> Doctoral Researcher
> >> >>
> >> >> Chair of Software Engineering
> >> >> Department of Computer Science
> >> >> University of Würzburg
> >> >>
> >> >> Am Hubland, Informatikgebäude (M2), Room 107/108
> >> >> 97074 Würzburg, Germany
> >> >>
> >> >> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
> >> >> http://go.uni-wuerzburg.de/eismann/
> >> >>    
> >> ------------------------------------------------------------------------------------------
> >>  
> >> >> _______________________________________________
> >> >> Kieker-users mailing list
> >> >> Kie...@li...
> >> >> https://lists.sourceforge.net/lists/listinfo/kieker-users  
> >> >
> >> > --
> >> > Dr.-Ing. André van Hoorn
> >> >
> >> > University of Stuttgart, Inst. of Software Engineering
> >> > Software Quality and Architecture Group
> >> > Universitätsstraße 38, D-70569 Stuttgart, Germany
> >> >
> >> > Room: 1.332
> >> > Phone:  +49 (0)711 685-88-252, Fax: -472
> >> > E-Mail: van...@in...
> >> >
> >> > https://www.iste.uni-stuttgart.de/institute/team/van-Hoorn/
> >> > http://kieker-monitoring.net/, http://research.spec.org/  
> >>
> >>
> >>
> >> ------------------------------------------------------------------------------------------
> >> M.Sc. Simon Eismann
> >> Doctoral Researcher
> >>
> >> Chair of Software Engineering
> >> Department of Computer Science
> >> University of Würzburg
> >>
> >> Am Hubland, Informatikgebäude (M2), Room 107/108
> >> 97074 Würzburg, Germany
> >>
> >> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
> >> http://go.uni-wuerzburg.de/eismann/
> >> ------------------------------------------------------------------------------------------
> >>  
> >
> >
> >
> > --
> > Dr.-Ing. Reiner Jung, Software Engineering Group
> > Project OceanDSL
> > Dept. Computer Science, Kiel University, D-24118 Kiel, Germany
> > Tel: +49 (0)431 880-2776, Fax: -7617
> > Email: rei...@em...
> > http://se.informatik.uni-kiel.de/  
> 
> 
> 
> ------------------------------------------------------------------------------------------
> M.Sc. Simon Eismann
> Doctoral Researcher
> 
> Chair of Software Engineering
> Department of Computer Science
> University of Würzburg
> 
> Am Hubland, Informatikgebäude (M2), Room 107/108
> 97074 Würzburg, Germany
> 
> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
> http://go.uni-wuerzburg.de/eismann/
> ------------------------------------------------------------------------------------------



-- 
Dr.-Ing. Reiner Jung, Software Engineering Group
Project OceanDSL
Dept. Computer Science, Kiel University, D-24118 Kiel, Germany
Tel: +49 (0)431 880-2776, Fax: -7617
Email: rei...@em...
http://se.informatik.uni-kiel.de/

Re: [Kieker-users] Using Kieker with Java 11

[Simon E. <sim...@un...>]

Hi Reiner,

thank you for your support, the complete log helped reveal the issue.  
We were missing some dependencies while compiling Kieker which caused  
the AMQP writer to fail. After fixing this, it works like a charm :)

Best regards,
Simon

Quoting Reiner Jung <rei...@em...>:

> Hi Simon, André, David,
>
> I had a look into the source code and it seems the a writer is not
> instantiated correctly. Maybe due to a missing property or a missing
> value. There should be an error message somewhere in the log.
>
> I would suggest that you try to create a complete log with
> your-program > log 2>&1
>
> That should collect all output. Then look for on of the following
> messages:
> - Class '{}' has to implement '{}'
> - {}: Class '{}' not found",
> - {}: Class '{}' has to implement a (public) constructor that accepts a
>   single Configuration
> - {}: Failed to load class for name '{}'
>
> Also you could look into your Kieker configuration file for the
> following setting:
> kieker.monitoring.writer.collector.ChunkingCollector.writer
>
> You can use any Class implementing the IRawDataWriter interface.
> Depending on the selected writer additional parameters are required.
>
> @andre @david For Kieker, we should better fail early when a parameter
> is missing and report that accordingly.
>
> Kind regards
> Reiner
>
> On Wed, 04 Aug 2021 11:02:41 +0000
> Simon Eismann <sim...@un...> wrote:
>
>> Hi Andre,
>>
>> as I'm not sure which version we used as I can't find the branch
>> anymore, we recompiled it with the current master branch, but
>> unfortunately, this did not solve the issue. I've attached a
>> screenshot of the error message we are currently getting.
>>
>> Best regards,
>> Simon
>>
>> Quoting Andre van Hoorn <and...@is...>:
>>
>> > Hi Simon,
>> >
>> > thanks for reaching out to us.
>> >
>> > Which version did you use? In the past weeks, we have conducted a
>> > bunch of fixes (also related to AspectJ and Java 11) that can be
>> > found in the master branch. The current master branch basically
>> > reflects the next Kieker release to be released very soon.
>> >
>> > I am CC'ing Reiner and David who worked on several issues in the
>> > last weeks.
>> >
>> > Best, André
>> >
>> > Am 29.07.21 um 20:52 schrieb Simon Eismann:
>> >> Dear kieker-users,
>> >>
>> >> we have been working on migrating the teastore project
>> >> (https://github.com/DescartesResearch/TeaStore) to Java 11 and we
>> >> are running into some issues with Kieker and Java 11.
>> >>
>> >> We compiled Kieker from a pull request for an AspectJ/Java11 fix
>> >> (at the time of writing I can not find this pull request anymore,
>> >> I guess it?) and are using the AMQP writer to collect the
>> >> messages. After the migration to Java 11, the kieker client does
>> >> no longer connect to the AMQP server at all. The service logs show
>> >> a NullPointerException at ChunkWriterTask.java:111 in
>> >> kieker-monitoring, which seems to originate from its writer being
>> >> null.
>> >>
>> >> Does anybody know what we might be doing wrong here?
>> >>
>> >> I've also included instructions on how to reproduce the issue
>> >> below.
>> >>
>> >> Best regards,
>> >> Simon Eismann
>> >>
>> >> ## How to reproduce
>> >>
>> >> Create docker-compose_kieker.yaml:
>> >>
>> >> ```
>> >> version: '3'
>> >> services:
>> >>   rabbitmq:
>> >>     image: descartesresearch/teastore-kieker-rabbitmq
>> >>     expose:
>> >>       - "5672"
>> >>     ports:
>> >>       - "8081:8080"
>> >>       - "5672:5672"
>> >>       - "15672:15672"
>> >>   registry:
>> >>     image: simont1105/teastore-registry
>> >>     expose:
>> >>       - "8080"
>> >>   db:
>> >>     image: descartesresearch/teastore-db
>> >>     expose:
>> >>       - "3306"
>> >>     ports:
>> >>       - "3306:3306"
>> >>   persistence:
>> >>     image: simont1105/teastore-persistence
>> >>     expose:
>> >>       - "8080"
>> >>     environment:
>> >>       HOST_NAME: "persistence"
>> >>       REGISTRY_HOST: "registry"
>> >>       DB_HOST: "db"
>> >>       DB_PORT: "3306"
>> >>       RABBITMQ_HOST: "rabbitmq"
>> >>   auth:
>> >>     image: simont1105/teastore-auth
>> >>     expose:
>> >>       - "8080"
>> >>     environment:
>> >>       HOST_NAME: "auth"
>> >>       REGISTRY_HOST: "registry"
>> >>       RABBITMQ_HOST: "rabbitmq"
>> >>   image:
>> >>     image: simont1105/teastore-image
>> >>     expose:
>> >>       - "8080"
>> >>     environment:
>> >>       HOST_NAME: "image"
>> >>       REGISTRY_HOST: "registry"
>> >>       RABBITMQ_HOST: "rabbitmq"
>> >>   recommender:
>> >>     image: simont1105/teastore-recommender
>> >>     expose:
>> >>       - "8080"
>> >>     environment:
>> >>       HOST_NAME: "recommender"
>> >>       REGISTRY_HOST: "registry"
>> >>       RABBITMQ_HOST: "rabbitmq"
>> >>   webui:
>> >>     image: simont1105/teastore-webui
>> >>     environment:
>> >>       HOST_NAME: "webui"
>> >>       REGISTRY_HOST: "registry"
>> >>       RABBITMQ_HOST: "rabbitmq"
>> >>     ports:
>> >>       - "8080:8080"
>> >> ```
>> >>
>> >> Then execute:
>> >>
>> >> ```shell
>> >> docker-compose -f docker-compose_kieker.yaml up -d
>> >> ```
>> >>
>> >> After startup, open http://localhost:8080 (TeaStore),
>> >> http://localhost:15672 (RabbitMQ management, credentials:
>> >> admin:nimda) and http://localhost:8081 (RabbitMQ data) in browser.
>> >>
>> >> ## Building the TeaStore
>> >>
>> >> Requirements: openjdk11 and maven >3.6, docker and docker-compose.
>> >>
>> >> ```shell
>> >> mvn clean install -Dmaven.test.skip=true
>> >> ./examples/docker_build.sh -p -r simont1105 -v java11
>> >> # IMPORTANT: change simont1105 to your your docker repo in this
>> >> and the docker-compose file!
>> >> ```
>> >>
>> >> This would allow to try other kieker packages, just replace all
>> >> kieker 1.15 jars with new ones of the same kind.
>> >>
>> >>
>> >>  
>> ------------------------------------------------------------------------------------------
>> >> M.Sc. Simon Eismann
>> >> Doctoral Researcher
>> >>
>> >> Chair of Software Engineering
>> >> Department of Computer Science
>> >> University of Würzburg
>> >>
>> >> Am Hubland, Informatikgebäude (M2), Room 107/108
>> >> 97074 Würzburg, Germany
>> >>
>> >> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
>> >> http://go.uni-wuerzburg.de/eismann/
>> >>  
>> ------------------------------------------------------------------------------------------
>> >> _______________________________________________
>> >> Kieker-users mailing list
>> >> Kie...@li...
>> >> https://lists.sourceforge.net/lists/listinfo/kieker-users
>> >
>> > --
>> > Dr.-Ing. André van Hoorn
>> >
>> > University of Stuttgart, Inst. of Software Engineering
>> > Software Quality and Architecture Group
>> > Universitätsstraße 38, D-70569 Stuttgart, Germany
>> >
>> > Room: 1.332
>> > Phone:  +49 (0)711 685-88-252, Fax: -472
>> > E-Mail: van...@in...
>> >
>> > https://www.iste.uni-stuttgart.de/institute/team/van-Hoorn/
>> > http://kieker-monitoring.net/, http://research.spec.org/
>>
>>
>>
>> ------------------------------------------------------------------------------------------
>> M.Sc. Simon Eismann
>> Doctoral Researcher
>>
>> Chair of Software Engineering
>> Department of Computer Science
>> University of Würzburg
>>
>> Am Hubland, Informatikgebäude (M2), Room 107/108
>> 97074 Würzburg, Germany
>>
>> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
>> http://go.uni-wuerzburg.de/eismann/
>> ------------------------------------------------------------------------------------------
>
>
>
> --
> Dr.-Ing. Reiner Jung, Software Engineering Group
> Project OceanDSL
> Dept. Computer Science, Kiel University, D-24118 Kiel, Germany
> Tel: +49 (0)431 880-2776, Fax: -7617
> Email: rei...@em...
> http://se.informatik.uni-kiel.de/



------------------------------------------------------------------------------------------
M.Sc. Simon Eismann
Doctoral Researcher

Chair of Software Engineering
Department of Computer Science
University of Würzburg

Am Hubland, Informatikgebäude (M2), Room 107/108
97074 Würzburg, Germany

Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
http://go.uni-wuerzburg.de/eismann/
------------------------------------------------------------------------------------------

Re: [Kieker-users] Using Kieker with Java 11

[Reiner J. <rei...@em...>]

Hi Simon, André, David,

I had a look into the source code and it seems the a writer is not
instantiated correctly. Maybe due to a missing property or a missing
value. There should be an error message somewhere in the log.

I would suggest that you try to create a complete log with
your-program > log 2>&1

That should collect all output. Then look for on of the following
messages:
- Class '{}' has to implement '{}'
- {}: Class '{}' not found",
- {}: Class '{}' has to implement a (public) constructor that accepts a
  single Configuration
- {}: Failed to load class for name '{}'

Also you could look into your Kieker configuration file for the
following setting:
kieker.monitoring.writer.collector.ChunkingCollector.writer

You can use any Class implementing the IRawDataWriter interface.
Depending on the selected writer additional parameters are required.

@andre @david For Kieker, we should better fail early when a parameter
is missing and report that accordingly.

Kind regards
Reiner

On Wed, 04 Aug 2021 11:02:41 +0000
Simon Eismann <sim...@un...> wrote:

> Hi Andre,
> 
> as I'm not sure which version we used as I can't find the branch  
> anymore, we recompiled it with the current master branch, but  
> unfortunately, this did not solve the issue. I've attached a  
> screenshot of the error message we are currently getting.
> 
> Best regards,
> Simon
> 
> Quoting Andre van Hoorn <and...@is...>:
> 
> > Hi Simon,
> >
> > thanks for reaching out to us.
> >
> > Which version did you use? In the past weeks, we have conducted a
> > bunch of fixes (also related to AspectJ and Java 11) that can be
> > found in the master branch. The current master branch basically
> > reflects the next Kieker release to be released very soon.
> >
> > I am CC'ing Reiner and David who worked on several issues in the
> > last weeks.
> >
> > Best, André
> >
> > Am 29.07.21 um 20:52 schrieb Simon Eismann:  
> >> Dear kieker-users,
> >>
> >> we have been working on migrating the teastore project  
> >> (https://github.com/DescartesResearch/TeaStore) to Java 11 and we  
> >> are running into some issues with Kieker and Java 11.
> >>
> >> We compiled Kieker from a pull request for an AspectJ/Java11 fix  
> >> (at the time of writing I can not find this pull request anymore,
> >> I guess it?) and are using the AMQP writer to collect the
> >> messages. After the migration to Java 11, the kieker client does
> >> no longer connect to the AMQP server at all. The service logs show
> >> a NullPointerException at ChunkWriterTask.java:111 in  
> >> kieker-monitoring, which seems to originate from its writer being  
> >> null.
> >>
> >> Does anybody know what we might be doing wrong here?
> >>
> >> I've also included instructions on how to reproduce the issue
> >> below.
> >>
> >> Best regards,
> >> Simon Eismann
> >>
> >> ## How to reproduce
> >>
> >> Create docker-compose_kieker.yaml:
> >>
> >> ```
> >> version: '3'
> >> services:
> >>   rabbitmq:
> >>     image: descartesresearch/teastore-kieker-rabbitmq
> >>     expose:
> >>       - "5672"
> >>     ports:
> >>       - "8081:8080"
> >>       - "5672:5672"
> >>       - "15672:15672"
> >>   registry:
> >>     image: simont1105/teastore-registry
> >>     expose:
> >>       - "8080"
> >>   db:
> >>     image: descartesresearch/teastore-db
> >>     expose:
> >>       - "3306"
> >>     ports:
> >>       - "3306:3306"
> >>   persistence:
> >>     image: simont1105/teastore-persistence
> >>     expose:
> >>       - "8080"
> >>     environment:
> >>       HOST_NAME: "persistence"
> >>       REGISTRY_HOST: "registry"
> >>       DB_HOST: "db"
> >>       DB_PORT: "3306"
> >>       RABBITMQ_HOST: "rabbitmq"
> >>   auth:
> >>     image: simont1105/teastore-auth
> >>     expose:
> >>       - "8080"
> >>     environment:
> >>       HOST_NAME: "auth"
> >>       REGISTRY_HOST: "registry"
> >>       RABBITMQ_HOST: "rabbitmq"
> >>   image:
> >>     image: simont1105/teastore-image
> >>     expose:
> >>       - "8080"
> >>     environment:
> >>       HOST_NAME: "image"
> >>       REGISTRY_HOST: "registry"
> >>       RABBITMQ_HOST: "rabbitmq"
> >>   recommender:
> >>     image: simont1105/teastore-recommender
> >>     expose:
> >>       - "8080"
> >>     environment:
> >>       HOST_NAME: "recommender"
> >>       REGISTRY_HOST: "registry"
> >>       RABBITMQ_HOST: "rabbitmq"
> >>   webui:
> >>     image: simont1105/teastore-webui
> >>     environment:
> >>       HOST_NAME: "webui"
> >>       REGISTRY_HOST: "registry"
> >>       RABBITMQ_HOST: "rabbitmq"
> >>     ports:
> >>       - "8080:8080"
> >> ```
> >>
> >> Then execute:
> >>
> >> ```shell
> >> docker-compose -f docker-compose_kieker.yaml up -d
> >> ```
> >>
> >> After startup, open http://localhost:8080 (TeaStore),  
> >> http://localhost:15672 (RabbitMQ management, credentials:  
> >> admin:nimda) and http://localhost:8081 (RabbitMQ data) in browser.
> >>
> >> ## Building the TeaStore
> >>
> >> Requirements: openjdk11 and maven >3.6, docker and docker-compose.
> >>
> >> ```shell
> >> mvn clean install -Dmaven.test.skip=true
> >> ./examples/docker_build.sh -p -r simont1105 -v java11
> >> # IMPORTANT: change simont1105 to your your docker repo in this
> >> and the docker-compose file!
> >> ```
> >>
> >> This would allow to try other kieker packages, just replace all  
> >> kieker 1.15 jars with new ones of the same kind.
> >>
> >>
> >> ------------------------------------------------------------------------------------------
> >> M.Sc. Simon Eismann
> >> Doctoral Researcher
> >>
> >> Chair of Software Engineering
> >> Department of Computer Science
> >> University of Würzburg
> >>
> >> Am Hubland, Informatikgebäude (M2), Room 107/108
> >> 97074 Würzburg, Germany
> >>
> >> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
> >> http://go.uni-wuerzburg.de/eismann/
> >> ------------------------------------------------------------------------------------------
> >> _______________________________________________
> >> Kieker-users mailing list
> >> Kie...@li...
> >> https://lists.sourceforge.net/lists/listinfo/kieker-users  
> >
> > -- 
> > Dr.-Ing. André van Hoorn
> >
> > University of Stuttgart, Inst. of Software Engineering
> > Software Quality and Architecture Group
> > Universitätsstraße 38, D-70569 Stuttgart, Germany
> >
> > Room: 1.332
> > Phone:  +49 (0)711 685-88-252, Fax: -472
> > E-Mail: van...@in...
> >
> > https://www.iste.uni-stuttgart.de/institute/team/van-Hoorn/
> > http://kieker-monitoring.net/, http://research.spec.org/  
> 
> 
> 
> ------------------------------------------------------------------------------------------
> M.Sc. Simon Eismann
> Doctoral Researcher
> 
> Chair of Software Engineering
> Department of Computer Science
> University of Würzburg
> 
> Am Hubland, Informatikgebäude (M2), Room 107/108
> 97074 Würzburg, Germany
> 
> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
> http://go.uni-wuerzburg.de/eismann/
> ------------------------------------------------------------------------------------------



-- 
Dr.-Ing. Reiner Jung, Software Engineering Group
Project OceanDSL
Dept. Computer Science, Kiel University, D-24118 Kiel, Germany
Tel: +49 (0)431 880-2776, Fax: -7617
Email: rei...@em...
http://se.informatik.uni-kiel.de/

Re: [Kieker-users] Using Kieker with Java 11

[Andre v. H. <and...@is...>]

Hi Simon,

thanks for reaching out to us.

Which version did you use? In the past weeks, we have conducted a bunch
of fixes (also related to AspectJ and Java 11) that can be found in the
master branch. The current master branch basically reflects the next
Kieker release to be released very soon.

I am CC'ing Reiner and David who worked on several issues in the last
weeks.

Best, André

Am 29.07.21 um 20:52 schrieb Simon Eismann:
> Dear kieker-users,
> 
> we have been working on migrating the teastore project 
> (https://github.com/DescartesResearch/TeaStore) to Java 11 and we are 
> running into some issues with Kieker and Java 11.
> 
> We compiled Kieker from a pull request for an AspectJ/Java11 fix (at the 
> time of writing I can not find this pull request anymore, I guess it?) 
> and are using the AMQP writer to collect the messages. After the 
> migration to Java 11, the kieker client does no longer connect to the 
> AMQP server at all. The service logs show a NullPointerException at 
> ChunkWriterTask.java:111 in kieker-monitoring, which seems to originate 
> from its writer being null.
> 
> Does anybody know what we might be doing wrong here?
> 
> I've also included instructions on how to reproduce the issue below.
> 
> Best regards,
> Simon Eismann
> 
> ## How to reproduce
> 
> Create docker-compose_kieker.yaml:
> 
> ```
> version: '3'
> services:
>    rabbitmq:
>      image: descartesresearch/teastore-kieker-rabbitmq
>      expose:
>        - "5672"
>      ports:
>        - "8081:8080"
>        - "5672:5672"
>        - "15672:15672"
>    registry:
>      image: simont1105/teastore-registry
>      expose:
>        - "8080"
>    db:
>      image: descartesresearch/teastore-db
>      expose:
>        - "3306"
>      ports:
>        - "3306:3306"
>    persistence:
>      image: simont1105/teastore-persistence
>      expose:
>        - "8080"
>      environment:
>        HOST_NAME: "persistence"
>        REGISTRY_HOST: "registry"
>        DB_HOST: "db"
>        DB_PORT: "3306"
>        RABBITMQ_HOST: "rabbitmq"
>    auth:
>      image: simont1105/teastore-auth
>      expose:
>        - "8080"
>      environment:
>        HOST_NAME: "auth"
>        REGISTRY_HOST: "registry"
>        RABBITMQ_HOST: "rabbitmq"
>    image:
>      image: simont1105/teastore-image
>      expose:
>        - "8080"
>      environment:
>        HOST_NAME: "image"
>        REGISTRY_HOST: "registry"
>        RABBITMQ_HOST: "rabbitmq"
>    recommender:
>      image: simont1105/teastore-recommender
>      expose:
>        - "8080"
>      environment:
>        HOST_NAME: "recommender"
>        REGISTRY_HOST: "registry"
>        RABBITMQ_HOST: "rabbitmq"
>    webui:
>      image: simont1105/teastore-webui
>      environment:
>        HOST_NAME: "webui"
>        REGISTRY_HOST: "registry"
>        RABBITMQ_HOST: "rabbitmq"
>      ports:
>        - "8080:8080"
> ```
> 
> Then execute:
> 
> ```shell
> docker-compose -f docker-compose_kieker.yaml up -d
> ```
> 
> After startup, open http://localhost:8080 (TeaStore), 
> http://localhost:15672 (RabbitMQ management, credentials: admin:nimda) 
> and http://localhost:8081 (RabbitMQ data) in browser.
> 
> ## Building the TeaStore
> 
> Requirements: openjdk11 and maven >3.6, docker and docker-compose.
> 
> ```shell
> mvn clean install -Dmaven.test.skip=true
> ./examples/docker_build.sh -p -r simont1105 -v java11
> # IMPORTANT: change simont1105 to your your docker repo in this and the 
> docker-compose file!
> ```
> 
> This would allow to try other kieker packages, just replace all kieker 
> 1.15 jars with new ones of the same kind.
> 
> 
> ------------------------------------------------------------------------------------------ 
> 
> M.Sc. Simon Eismann
> Doctoral Researcher
> 
> Chair of Software Engineering
> Department of Computer Science
> University of Würzburg
> 
> Am Hubland, Informatikgebäude (M2), Room 107/108
> 97074 Würzburg, Germany
> 
> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
> http://go.uni-wuerzburg.de/eismann/
> ------------------------------------------------------------------------------------------ 
> 
> 
> 
> _______________________________________________
> Kieker-users mailing list
> Kie...@li...
> https://lists.sourceforge.net/lists/listinfo/kieker-users

-- 
Dr.-Ing. André van Hoorn

University of Stuttgart, Inst. of Software Engineering
Software Quality and Architecture Group
Universitätsstraße 38, D-70569 Stuttgart, Germany

Room: 1.332
Phone:  +49 (0)711 685-88-252, Fax: -472
E-Mail: van...@in...

https://www.iste.uni-stuttgart.de/institute/team/van-Hoorn/
http://kieker-monitoring.net/, http://research.spec.org/

[Kieker-users] Using Kieker with Java 11

[Simon E. <sim...@un...>]

Dear kieker-users,

we have been working on migrating the teastore project  
(https://github.com/DescartesResearch/TeaStore) to Java 11 and we are  
running into some issues with Kieker and Java 11.

We compiled Kieker from a pull request for an AspectJ/Java11 fix (at  
the time of writing I can not find this pull request anymore, I guess  
it?) and are using the AMQP writer to collect the messages. After the  
migration to Java 11, the kieker client does no longer connect to the  
AMQP server at all. The service logs show a NullPointerException at  
ChunkWriterTask.java:111 in kieker-monitoring, which seems to  
originate from its writer being null.

Does anybody know what we might be doing wrong here?

I've also included instructions on how to reproduce the issue below.

Best regards,
Simon Eismann

## How to reproduce

Create docker-compose_kieker.yaml:

```
version: '3'
services:
   rabbitmq:
     image: descartesresearch/teastore-kieker-rabbitmq
     expose:
       - "5672"
     ports:
       - "8081:8080"
       - "5672:5672"
       - "15672:15672"
   registry:
     image: simont1105/teastore-registry
     expose:
       - "8080"
   db:
     image: descartesresearch/teastore-db
     expose:
       - "3306"
     ports:
       - "3306:3306"
   persistence:
     image: simont1105/teastore-persistence
     expose:
       - "8080"
     environment:
       HOST_NAME: "persistence"
       REGISTRY_HOST: "registry"
       DB_HOST: "db"
       DB_PORT: "3306"
       RABBITMQ_HOST: "rabbitmq"
   auth:
     image: simont1105/teastore-auth
     expose:
       - "8080"
     environment:
       HOST_NAME: "auth"
       REGISTRY_HOST: "registry"
       RABBITMQ_HOST: "rabbitmq"
   image:
     image: simont1105/teastore-image
     expose:
       - "8080"
     environment:
       HOST_NAME: "image"
       REGISTRY_HOST: "registry"
       RABBITMQ_HOST: "rabbitmq"
   recommender:
     image: simont1105/teastore-recommender
     expose:
       - "8080"
     environment:
       HOST_NAME: "recommender"
       REGISTRY_HOST: "registry"
       RABBITMQ_HOST: "rabbitmq"
   webui:
     image: simont1105/teastore-webui
     environment:
       HOST_NAME: "webui"
       REGISTRY_HOST: "registry"
       RABBITMQ_HOST: "rabbitmq"
     ports:
       - "8080:8080"
```

Then execute:

```shell
docker-compose -f docker-compose_kieker.yaml up -d
```

After startup, open http://localhost:8080 (TeaStore),  
http://localhost:15672 (RabbitMQ management, credentials: admin:nimda)  
and http://localhost:8081 (RabbitMQ data) in browser.

## Building the TeaStore

Requirements: openjdk11 and maven >3.6, docker and docker-compose.

```shell
mvn clean install -Dmaven.test.skip=true
./examples/docker_build.sh -p -r simont1105 -v java11
# IMPORTANT: change simont1105 to your your docker repo in this and  
the docker-compose file!
```

This would allow to try other kieker packages, just replace all kieker  
1.15 jars with new ones of the same kind.


------------------------------------------------------------------------------------------
M.Sc. Simon Eismann
Doctoral Researcher

Chair of Software Engineering
Department of Computer Science
University of Würzburg

Am Hubland, Informatikgebäude (M2), Room 107/108
97074 Würzburg, Germany

Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
http://go.uni-wuerzburg.de/eismann/
------------------------------------------------------------------------------------------

[Kieker-users] Representative traces & Loop instrumentation

[Bernardo A. <ber...@gm...>]

Hello again. I previously sent you an email which you kindly answered back
and consequently I've been able to manage all the issues since then so I'm
very grateful for your help.
This time, I'm not at a dead-end but I'm searching for possible ways to
optimize/decrease the amount of collected data through dynamic analysis as
a post-processing phase.

To contextualize, imagine that my AspectJ instrumentation is catching all
the functions that are executed for a given user request. Now, imagine that
some of those functions are called in the body of a loop which will iterate
N times.
Therefore, in most of those iterations the logs will be the same (if the
same path is traversed) leading to an excessive amount of duplicated
information.
A well-suited use case for this problem can be the presence of extensive
lists on websites especially if they don't use pagination (the whole list
is sent at once).

Given that, I have 2 questions that are related to 2 different ways of
reducing the amount of data that I'm currently thinking of:
*1) Does your TraceAnalysis tool already provide different ways of getting
representative traces or allow the customization of how the representative
traces can be gathered?*
As far as I know, the representative traces present on the
*traceAssemblyEquivClasses.txt* file are only the ones that have a
different number of executions and, if they have the same number of
executions, then they should have at least a log line that is different
from the other. Otherwise, they are considered the same and the cardinality
is raised.
What I would really like to achieve is a different concept of
representative traces where a trace is considered representative if it is
not a subsequence of others. E.g: Let the following letters be function
names - A, B, C, D - and it is known that C belongs to the body of a loop.
Given 3 different user requests it was obtained the following logs/traces
for the same endpoint:
a) A B C C C D
b) A B C D
c) B C
With this in mind, the trace a) is a representative of both b) and c) since
these two are a subset of it and the order between the function calls is
kept.
As you may already be guessing, with this new concept both traces b) and c)
would be discarded in a post processing phase and consequently less data
would be produced. (sometimes i call it a "trim the fat" process)

*2)* *Do you know a way to log the beginning and the end of a loop through
AspectJ or other technology that Kieker is/can be integrated with?*
I already searched for possible *AspectJ* pointcuts to achieve *loop
logging *but I guess there is no native pointcut dedicated to loops
(correct me if i'm wrong). There is, however, this effort
https://www.researchgate.net/publication/221014879_A_join_point_for_loops_in_AspectJ
that
I found out but given that the official AspectJ documentation does not
refer anything about my requirement I guess my time will be wasted with a
lot of headaches in the middle.

The foundation behind this question is about knowing where to start
searching for duplicated sequences of logs ( trace *a)* can serve as an
example where C appears 3 times in a row). In a post-processing phase one
could directly start "compacting" a trace by searching for the beginning
and the end of loops. The final result would be that trace *a)* is,
instead, A B (C)x3 D which takes less space than the original and no data
is lost. Right now, without knowing the location of loops, one has to
traverse the whole trace from the beginning and search for these sort of
duplicated sequences which may not even belong to a loop and the algorithm
(let's call it sequence matching) by itself takes a lot longer since it's
"blind". As I already witnessed, working with traces that ascend to more
than 1 million of log lines (easily achieved when lists have more than 1000
elements) makes the processing phase almost unfeasible when searching for
these loop sequences (blindly).

Thank you for your attention and hope to hear from you soon.

Kind regards,
Bernardo Andrade

[Kieker-users] AccessControlException in AspectJ-based instrumentation

[<mr...@in...>]

Hello,

 

I want to use Kieker to monitor Elasticsearch.

 

Therefor I modified the elasticsearch start script to use the
kieker-aspectj.jar as javaagent.

 

But when I start the script it throws following exception:

 

Exception in thread "main" java.lang.ExceptionInInitializerError

        at
kieker.monitoring.probe.aspectj.AspectJLoader.<clinit>(AspectJLoader.java:40
)

        at
java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native
Method)

        at
java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodA
ccessorImpl.java:62)

        at
java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(Delegatin
gMethodAccessorImpl.java:43)

        at java.base/java.lang.reflect.Method.invoke(Method.java:564)

        at
java.instrument/sun.instrument.InstrumentationImpl.loadClassAndStartAgent(In
strumentationImpl.java:513)

        at
java.instrument/sun.instrument.InstrumentationImpl.loadClassAndCallPremain(I
nstrumentationImpl.java:525)

Caused by: java.security.AccessControlException: access denied
("java.util.PropertyPermission" "kieker.common.logging.Log" "read")

        at
java.base/java.security.AccessControlContext.checkPermission(AccessControlCo
ntext.java:472)

        at
java.base/java.security.AccessController.checkPermission(AccessController.ja
va:1036)

        at
java.base/java.lang.SecurityManager.checkPermission(SecurityManager.java:408
)

        at
java.base/java.lang.SecurityManager.checkPropertyAccess(SecurityManager.java
:1152)

        at java.base/java.lang.System.getProperty(System.java:847)

        at kieker.common.logging.LogFactory.<clinit>(LogFactory.java:41)

        ... 7 more

*** java.lang.instrument ASSERTION FAILED ***: "result" with message agent
load/premain call failed at
./open/src/java.instrument/share/native/libinstrument/JPLISAgent.c line: 422

FATAL ERROR in native method: processing of -javaagent failed,
processJavaStart failed

 

I tried to add a concrete security file allowing the program to access that
property, but that did not changed anything.

 

So I am really stuck now. 

Could anyone help me out?

 

Greetings 

Marius

Re: [Kieker-users] Issues & Trace Analysis Tool enhancement

[Andre v. H. <van...@in...>]

Dear Bernardo,

thanks for your request.

On 17.05.20 00:40, Bernardo Andrade wrote:
> Hello, I'm currently using /Kieker-1.13/ to generate execution traces
> and then applying your trace analysis tool to infer some aspects on my
> project. My intention with this mail is to seek some guidance from you
> with the goal of solving 2 issues or at least to discover an easy
> workaround for them.
> 
> (1) One thing I noticed is that Kieker is not being able to infer
> parameterized function return types (e.g, one line of the log contains
> only /java.util.Set/ but in reality, the function returns something like
> /java.util.Set<x.y.z>/). I would really love to know the
> parameterized type as I'm trying to infer things from a dynamic
> perspective. So.. can you give me any idea of how to properly get it?

the AspectJ-based instrumentation fetches the type information at
runtime. As the type information for generic types is removed after
the compilation, it seems like you won't be able to do this by extending
the current probes (that would be the approach apart from this problem).

See also
https://stackoverflow.com/questions/47264428/aspectj-pointcut-to-match-a-method-that-has-generic-parameters

> (2) Is there a way to attach/concatenate function return types to the
> output of the trace analysis (/executionTraces/ file)? Because somehow,
> after its execution, the return type of each function disappears and I
> would really appreciate to have an output very similar to the traces
> file (/.dat /file). If I'm not mistaken (correct me if i'm wrong), on
> the /.dat/ file you display the whole /operationSignature/ (return type
> + function path/name) of the /OperationExecutionRecord /and on
> the /executionTraces/ one you are cutting out the return type. Btw, I'm
> currently executing it through the /trace-analysis.sh/ script you provide. 

I understand that you would like to have the return type included in a
file like [1].

[1]
https://github.com/kieker-monitoring/kieker/blob/master/kieker-examples/userguide/ch5--trace-monitoring-aspectj/testdata/kieker-20100830-082225522-UTC-example-plots/executionTraces.txt
[2]
https://github.com/kieker-monitoring/kieker/blob/master/kieker-examples/userguide/ch5--trace-monitoring-aspectj/testdata/kieker-20100830-082225522-UTC

The responsible line of code for printing the lines into [1] can be
found in
https://github.com/kieker-monitoring/kieker/blob/master/kieker-tools/src/kieker/tools/trace/analysis/filter/traceWriter/ExecutionTraceWriterFilter.java#L103

It calls the ExecutionExtrace's toString method [3] that again calls
the toString method of each Execution [4]. Indeed, the Execution [4]
does not output the return type. However, via the Operation->Signature,
the returnType is available.

Hence, you could create a patch by changing the toString method [4].

[3]
https://github.com/kieker-monitoring/kieker/blob/master/kieker-tools/src/kieker/tools/trace/analysis/systemModel/ExecutionTrace.java#L251

[4]
https://github.com/kieker-monitoring/kieker/blob/master/kieker-tools/src/kieker/tools/trace/analysis/systemModel/Execution.java#L196


> P.S: Congrats for your recent /1.14/ release👏

thanks ;-).

Best, André

> 
> Regards,
> Bernardo Andrade
> 
> 
> _______________________________________________
> Kieker-users mailing list
> Kie...@li...
> https://lists.sourceforge.net/lists/listinfo/kieker-users
> 

-- 
Dr.-Ing. André van Hoorn

University of Stuttgart, Inst. of Software Technology
Reliable Software Systems Group
Universitätsstraße 38, D-70569 Stuttgart, Germany

Room: 1.332
Phone:  +49 (0)711 685-88-252, Fax: -472
E-Mail: van...@in...

https://www.iste.uni-stuttgart.de/institute/team/van-Hoorn/
http://kieker-monitoring.net/, http://research.spec.org/

[Kieker-users] Issues & Trace Analysis Tool enhancement

[Bernardo A. <ber...@gm...>]

Hello, I'm currently using *Kieker-1.13* to generate execution traces and
then applying your trace analysis tool to infer some aspects on my project.
My intention with this mail is to seek some guidance from you with the goal
of solving 2 issues or at least to discover an easy workaround for them.

(1) One thing I noticed is that Kieker is not being able to infer
parameterized function return types (e.g, one line of the log contains only
*java.util.Set* but in reality, the function returns something like
*java.util.Set<x.y.z>*). I would really love to know the parameterized type
as I'm trying to infer things from a dynamic perspective. So.. can you give
me any idea of how to properly get it?

(2) Is there a way to attach/concatenate function return types to the
output of the trace analysis (*executionTraces* file)? Because somehow,
after its execution, the return type of each function disappears and I
would really appreciate to have an output very similar to the traces
file (*.dat
*file). If I'm not mistaken (correct me if i'm wrong), on the *.dat* file
you display the whole *operationSignature* (return type + function
path/name) of the *OperationExecutionRecord *and on the *executionTraces*
one you are cutting out the return type. Btw, I'm currently executing it
through the *trace-analysis.sh* script you provide.

Thanks in advance!

P.S: Congrats for your recent *1.14* release👏

Regards,
Bernardo Andrade

[Kieker-users] Kieker 1.14 released

[Andre v. H. <van...@in...>]

Dear all,

On May 12, 2020, we released version 1.14 of our Kieker framework for
application performance monitoring and dynamic software analysis. As
usual, the release is available for download at
http://kieker-monitoring.net/download/.

Release Notes: http://kieker-monitoring.net/release-notes/

Many thanks to all contributors!

Best, André

-- 
Dr.-Ing. André van Hoorn

University of Stuttgart, Inst. of Software Technology
Reliable Software Systems Group
Universitätsstraße 38, D-70569 Stuttgart, Germany

Room: 1.332
Phone:  +49 (0)711 685-88-252, Fax: -472
E-Mail: van...@in...

https://www.iste.uni-stuttgart.de/institute/team/van-Hoorn/
http://kieker-monitoring.net/, http://research.spec.org/

[Kieker-users] Using Kieker in a JBoss Environment

[<stu...@ma...>]

Hello,

 

I have a question about running Kieker in a JBoss environment. I followed
this guide
(https://kieker-monitoring.atlassian.net/wiki/spaces/DOC/pages/24215574/Usin
g+Kieker+in+Different+Java+EE+Environments), even though I am not sure
whether this is still up to date, since I have heard that the AspectJ Weaver
is not a standalone module anymore. I am very helpful for any help and hints
on what the problem could be.
I am using kieker-1.14-SNAPSHOT-aspectj.jar and jboss-eap-6.3. This is my
stack trace:

 

[AppClassLoader@18b4aac2] info AspectJ Weaver Version 1.8.2 built on
Thursday Aug 14, 2014 at 21:45:02 GMT

[AppClassLoader@18b4aac2] info register classloader
sun.misc.Launcher$AppClassLoader@18b4aac2

[AppClassLoader@18b4aac2] info register aspect
kieker.monitoring.probe.aspectj.flow.operationExecution.FullInstrumentationN
oGetterAndSetter

[AppClassLoader@18b4aac2] info register aspect
kieker.monitoring.probe.aspectj.database.Fullinstrumentation

[AppClassLoader@18b4aac2] info processing reweavable type
kieker.monitoring.probe.aspectj.flow.operationExecution.AbstractAspect:
kieker\monitoring\probe\aspectj\flow\operationExecution\AbstractAspect.java

[AppClassLoader@18b4aac2] info successfully verified type
kieker.monitoring.probe.aspectj.flow.operationExecution.AbstractAspect
exists.  Originates from
kieker\monitoring\probe\aspectj\flow\operationExecution\AbstractAspect.java

[AppClassLoader@18b4aac2] info processing reweavable type
kieker.monitoring.probe.aspectj.AbstractAspectJProbe:
kieker\monitoring\probe\aspectj\AbstractAspectJProbe.java

[AppClassLoader@18b4aac2] info successfully verified type
kieker.monitoring.probe.aspectj.AbstractAspectJProbe exists.  Originates
from kieker\monitoring\probe\aspectj\AbstractAspectJProbe.java

[AppClassLoader@18b4aac2] info processing reweavable type
kieker.monitoring.probe.aspectj.flow.operationExecution.FullInstrumentationN
oGetterAndSetter:
kieker\monitoring\probe\aspectj\flow\operationExecution\FullInstrumentationN
oGetterAndSetter.java

[AppClassLoader@18b4aac2] info successfully verified type
kieker.monitoring.probe.aspectj.flow.operationExecution.FullInstrumentationN
oGetterAndSetter exists.  Originates from
kieker\monitoring\probe\aspectj\flow\operationExecution\FullInstrumentationN
oGetterAndSetter.java

Exception in thread "main" java.lang.ExceptionInInitializerError

        at
kieker.monitoring.core.controller.MonitoringController.getInstance(Monitorin
gController.java:360)

        at
kieker.monitoring.probe.aspectj.flow.operationExecution.AbstractAspect.<clin
it>(AbstractAspect.java:42)

        at org.jboss.modules.StreamUtil.safeClose(StreamUtil.java:1)

        at org.jboss.modules.Main.<clinit>(Main.java:512)

Caused by: java.lang.IllegalStateException:
java.lang.reflect.InvocationTargetException

        at
kieker.monitoring.core.controller.WriterController.newQueue(WriterController
.java:202)

        at
kieker.monitoring.core.controller.WriterController.<init>(WriterController.j
ava:92)

        at
kieker.monitoring.core.controller.MonitoringController.<init>(MonitoringCont
roller.java:63)

        at
kieker.monitoring.core.controller.MonitoringController.createInstance(Monito
ringController.java:82)

        at
kieker.monitoring.core.controller.MonitoringController$LazyHolder.<clinit>(M
onitoringController.java:367)

        ... 4 more

Caused by: java.lang.reflect.InvocationTargetException

        at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native
Method)

        at
sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAcces
sorImpl.java:62)

        at
sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstruc
torAccessorImpl.java:45)

        at java.lang.reflect.Constructor.newInstance(Constructor.java:423)

        at
kieker.monitoring.core.controller.WriterController.newQueue(WriterController
.java:190)

        ... 8 more

Caused by: org.aspectj.lang.NoAspectBoundException:
kieker.monitoring.probe.aspectj.flow.operationExecution.FullInstrumentationN
oGetterAndSetter

        at
kieker.monitoring.probe.aspectj.flow.operationExecution.FullInstrumentationN
oGetterAndSetter.aspectOf(FullInstrumentationNoGetterAndSetter.java:1)

        at org.jctools.util.Pow2.roundToPowerOfTwo(Pow2.java:1)

        at
org.jctools.queues.ConcurrentCircularArrayQueue.<init>(ConcurrentCircularArr
ayQueue.java:43)

        at
org.jctools.queues.MpscArrayQueueL1Pad.<init>(MpscArrayQueue.java:28)

        at
org.jctools.queues.MpscArrayQueueProducerIndexField.<init>(MpscArrayQueue.ja
va:54)

        at
org.jctools.queues.MpscArrayQueueMidPad.<init>(MpscArrayQueue.java:76)

        at
org.jctools.queues.MpscArrayQueueProducerLimitField.<init>(MpscArrayQueue.ja
va:103)

        at
org.jctools.queues.MpscArrayQueueL2Pad.<init>(MpscArrayQueue.java:125)

        at
org.jctools.queues.MpscArrayQueueConsumerIndexField.<init>(MpscArrayQueue.ja
va:151)

        at
org.jctools.queues.MpscArrayQueueL3Pad.<init>(MpscArrayQueue.java:178)

        at org.jctools.queues.MpscArrayQueue.<init>(MpscArrayQueue.java:199)

        ... 13 more

 

 

 

Thanks,

Stephan L.

[Kieker-users] Utilization measurements

[Simon E. <sim...@un...>]

Hello Kieker-users,

I am currently using Kieker to measure the utilization of an  
application using the following setup:

    <context-param>
     	<param-name>CPUMemUsageServletContextListener.samplingIntervalSeconds
	</param-name>
	<param-value>1</param-value>
    </context-param>
    <context-param>
	<param-name>CPUMemUsageServletContextListener.initialSamplingDelaySeconds
	</param-name>
	<param-value>0</param-value>
    </context-param>
    <listener>
	<listener-class>kieker.monitoring.probe.servlet.CPUMemUsageServletContextListener
	</listener-class>
    </listener>

However, the utilization in the logs is lower than the utilization  
reported by the system. Calling

System.out.println(record.getNice() + record.getIdle() +  
record.getIrq() + record.getSystem() + record.getUser() +  
record.getWait());

on the CPUUtilizationRecords leads to the following output:

0.78
0.6494845360824743
0.72
1.0
0.9591836734693877
0.61
0.7745098039215687
0.6701030927835052
0.9801980198019802
0.9587628865979382
0.5686274509803921
0.7474747474747474
0.6666666666666667
1.0

Should this not always result in a total of 1 or am I missing  
something here? The application is deployed in a docker container in a  
Debian 9 VM managed by KVM on a centos 7 machine. Is maybe the heavy  
virtualization causing some problems for the utilization measurements?  
However docker stats/cAdvisior do report the correct utilizations.

Kind Regards,
Simon Eismann

------------------------------------------------------------------------------------------
M.Sc. Simon Eismann
Doctoral Researcher

Chair of Software Engineering
Department of Computer Science
University of Würzburg

Am Hubland, Informatikgebäude (M2), Room 107/108
97074 Würzburg, Germany

Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
http://go.uni-wuerzburg.de/eismann/
------------------------------------------------------------------------------------------

[Kieker-users] Parameter logging

[Johannes G. <joh...@un...>]

Hi everybody,

we have instrumented our TeaStore (https://github.com/DescartesResearch/TeaStore) application using Kieker.  
We now want to include optional logging of parameter values for individual method calls. 
I am quite sure that Kieker supports this (as I think this was done in the iObserve (https://www.iobserve-devops.net/) project?), but could not find the right ressources on how to actually do this.

Could you point me to some documentation or example on how to log parameters on single methods calls?

Thank you very much!


All the best, 

Johannes


--
----------------------------------------------------------------------
M.Sc. Johannes Grohmann
Doctoral Researcher

Chair of Computer Science II - Software Engineering

University of Würzburg
Am Hubland, Informatikgebäude, Room A110
97074 Würzburg, Germany

Phone: +49 (931) 31 86875
https://go.uniwue.de/grohmann
----------------------------------------------------------------------

Re: [Kieker-users] Tracking Http Calls

[Simon E. <sim...@un...>]

Hello Andre,

We are using Servlets and basic HTTP calls. If I create an assembled  
call path on the log, requests from one Servlet to another are shown  
as coming from 'Entry', i.e. the start node of the tree. Sorry, that  
was a bit confusing.

Regards,
Simon

Quoting Andre van Hoorn <van...@in...>:

> Hi Simon,
>
> what does `Entry` mean? What technology stack are you using?
> (Servlets, ...)
>
> Best, André
>
> On 02.02.2018 14:35, Simon Eismann wrote:
>> Hello Kieker-users,
>>
>> we are currently trying to monitor a distributed application, which uses
>> calls to Rest-Interfaces to communicate between the individual services.
>> Our current monitoring configuration does not "understand" the http
>> calls and assumes that all traffic comes from 'Entry'. Is there any way
>> to trace http calls using kieker?
>>
>> Regards,
>> Simon Eismann
>>
>>
>> ------------------------------------------------------------------------------------------
>>
>> M.Sc. Simon Eismann
>> Doctoral Researcher
>>
>> Chair of Software Engineering
>> Department of Computer Science
>> University of Würzburg
>>
>> Am Hubland, Informatikgebäude (M2), Room 107/108
>> 97074 Würzburg, Germany
>>
>> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
>> http://go.uni-wuerzburg.de/eismann/
>> ------------------------------------------------------------------------------------------
>>
>>
>> ------------------------------------------------------------------------------
>>
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> _______________________________________________
>> Kieker-users mailing list
>> Kie...@li...
>> https://lists.sourceforge.net/lists/listinfo/kieker-users
>
> --
> Dr.-Ing. André van Hoorn
>
> University of Stuttgart, Inst. of Software Technology
> Reliable Software Systems Group
> Universitätsstraße 38, D-70569 Stuttgart, Germany
>
> Room: 1.332
> Phone:  +49 (0)711 685-88-252, Fax: -472
> E-Mail: van...@in...
>
> http://www.iste.uni-stuttgart.de/rss/people/vanhoorn/
> http://kieker-monitoring.net/, http://research.spec.org/



------------------------------------------------------------------------------------------
M.Sc. Simon Eismann
Doctoral Researcher

Chair of Software Engineering
Department of Computer Science
University of Würzburg

Am Hubland, Informatikgebäude (M2), Room 107/108
97074 Würzburg, Germany

Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
http://go.uni-wuerzburg.de/eismann/
------------------------------------------------------------------------------------------

Re: [Kieker-users] Tracking Http Calls

[Andre v. H. <van...@in...>]

Hi Simon,

what does `Entry` mean? What technology stack are you using?
(Servlets, ...)

Best, André

On 02.02.2018 14:35, Simon Eismann wrote:
> Hello Kieker-users,
> 
> we are currently trying to monitor a distributed application, which uses
> calls to Rest-Interfaces to communicate between the individual services.
> Our current monitoring configuration does not "understand" the http
> calls and assumes that all traffic comes from 'Entry'. Is there any way
> to trace http calls using kieker?
> 
> Regards,
> Simon Eismann
> 
> 
> ------------------------------------------------------------------------------------------
> 
> M.Sc. Simon Eismann
> Doctoral Researcher
> 
> Chair of Software Engineering
> Department of Computer Science
> University of Würzburg
> 
> Am Hubland, Informatikgebäude (M2), Room 107/108
> 97074 Würzburg, Germany
> 
> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
> http://go.uni-wuerzburg.de/eismann/
> ------------------------------------------------------------------------------------------
> 
> 
> ------------------------------------------------------------------------------
> 
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> Kieker-users mailing list
> Kie...@li...
> https://lists.sourceforge.net/lists/listinfo/kieker-users

-- 
Dr.-Ing. André van Hoorn

University of Stuttgart, Inst. of Software Technology
Reliable Software Systems Group
Universitätsstraße 38, D-70569 Stuttgart, Germany

Room: 1.332
Phone:  +49 (0)711 685-88-252, Fax: -472
E-Mail: van...@in...

http://www.iste.uni-stuttgart.de/rss/people/vanhoorn/
http://kieker-monitoring.net/, http://research.spec.org/

[Kieker-users] Tracking Http Calls

[Simon E. <sim...@un...>]

Hello Kieker-users,

we are currently trying to monitor a distributed application, which  
uses calls to Rest-Interfaces to communicate between the individual  
services. Our current monitoring configuration does not "understand"  
the http calls and assumes that all traffic comes from 'Entry'. Is  
there any way to trace http calls using kieker?

Regards,
Simon Eismann


------------------------------------------------------------------------------------------
M.Sc. Simon Eismann
Doctoral Researcher

Chair of Software Engineering
Department of Computer Science
University of Würzburg

Am Hubland, Informatikgebäude (M2), Room 107/108
97074 Würzburg, Germany

Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
http://go.uni-wuerzburg.de/eismann/
------------------------------------------------------------------------------------------

Re: [Kieker-users] AMQP Writer Exception

[Holger K. <hk...@in...>]

Hi Simon,

which Kieker version are you using? At a first glance, it looks like you 
stumbled on a bug I fixed some time ago, but the fix will only be part 
of the upcoming version 1.14. If you are running Kieker version 1.13, 
please build Kieker off the current stable branch. If you have already 
done so, please give me a pointer so I can look further into the issue.

Best regards,
Holger

> Hi Holger,
> 
> can you help Simon?
> 
> Thanks, André
> 
> On 11.12.2017 16:31, Simon Eismann wrote:
>> Hello Kieker-users,
>> I am currently trying to write my Kieker logs to a RabbitMQ-server 
>> using
>> the AMQP-writer. Unfortunately, i am getting the following error 
>> message:
>> 
>> store_1        | [localhost-startStop-1] INFO
>> kieker.monitoring.core.controller.StateController - Enabling
>> monitoring store_1        | [localhost-startStop-1] INFO
>> kieker.monitoring.core.controller.MonitoringController - Current
>> State of kieker.monitoring (1.13-SNAPSHOT) Status: 'enabled'
>> store_1        |        Name: 'KIEKER'; Hostname: 'store';
>> experimentID: '1' store_1        | JMXController: JMX disabled
>> store_1        | TimeSource: 'kieker.monitoring.timer.SystemNanoTimer'
>> store_1        |        Time in nanoseconds (with nanoseconds
>> precision) since Thu Jan 01 00:00:00 UTC 1970'
>> store_1        | ProbeController: disabled
>> store_1        | WriterController:
>> store_1        |        Queue capacity: 10000
>> store_1        |        Insert behavior (a.k.a. QueueFullBehavior):
>> numBlocked: 0
>> store_1        | Writer: 'kieker.monitoring.writer.amqp.AmqpWriter'
>> store_1        |        Configuration:
>> store_1        |
>> kieker.monitoring.writer.amqp.AmqpWriter.queuename='kieker'
>> store_1        |
>> kieker.monitoring.writer.amqp.AmqpWriter.uri='amqp://admin:nimda@10.1.3.44'
>> store_1        |
>> store_1        |        Automatic assignment of logging timestamps:
>> 'true' store_1        | Sampling Controller: Periodic Sensor
>> available: Poolsize: '0'; Scheduled Tasks: '0'
>> store_1        | Exception in thread "Thread-7"
>> java.nio.BufferOverflowException
>> store_1        |        at
>> java.nio.HeapByteBuffer.put(HeapByteBuffer.java:189)
>> store_1        |        at
>> java.nio.ByteBuffer.put(ByteBuffer.java:859) store_1        |
>> at
>> kieker.monitoring.writer.amqp.AmqpWriter.onNewRegistryEntry(AmqpWriter.java:187)
>> 
>> store_1        |        at
>> kieker.monitoring.writer.amqp.AmqpWriter.onNewRegistryEntry(AmqpWriter.java:51)
>> 
>> store_1        |        at
>> kieker.monitoring.registry.WriterRegistry.register(WriterRegistry.java:59)
>> store_1        |        at
>> kieker.monitoring.registry.WriterRegistry.register(WriterRegistry.java:28)
>> store_1        |        at
>> kieker.monitoring.registry.RegisterAdapter.get(RegisterAdapter.java:41)
>> store_1        |        at
>> kieker.common.record.misc.KiekerMetadataRecord.registerStrings(KiekerMetadataRecord.java:215)
>> 
>> store_1        |        at
>> kieker.monitoring.writer.amqp.AmqpWriter.writeMonitoringRecord(AmqpWriter.java:146)
>> 
>> store_1        |        at
>> kieker.monitoring.writer.MonitoringWriterThread.run(MonitoringWriterThread.java:66)
>> 
>> 
>> I couldn't find any documentation on the AQMP writer aside from a 
>> short
>> section in the user guide, so I am assuming I configured something
>> badly. Is there any more in-depth documentation on the AQMP writer or 
>> an
>> example that uses it?
>> Otherwise did anybody encounter this error before and knows how to 
>> solve
>> it?
>> 
>> Viele Grüße,
>> Simon Eismann
>> 
>> PS: This is my first time posting to this mailing list if I did 
>> anything
>> wrong, let me know
>> 
>> ------------------------------------------------------------------------------------------
>> 
>> M.Sc. Simon Eismann
>> Doctoral Researcher
>> 
>> Chair of Software Engineering
>> Department of Computer Science
>> University of Würzburg
>> 
>> Am Hubland, Informatikgebäude (M2), Room 107/108
>> 97074 Würzburg, Germany
>> 
>> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
>> http://go.uni-wuerzburg.de/eismann/
>> ------------------------------------------------------------------------------------------
>> 
>> 
>> ------------------------------------------------------------------------------
>> 
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> _______________________________________________
>> Kieker-users mailing list
>> Kie...@li...
>> https://lists.sourceforge.net/lists/listinfo/kieker-users
> 
> --
> Dr.-Ing. André van Hoorn
> 
> University of Stuttgart, Inst. of Software Technology
> Reliable Software Systems Group
> Universitätsstraße 38, D-70569 Stuttgart, Germany
> 
> Room: 1.332
> Phone:  +49 (0)711 685-88-252, Fax: -472
> E-Mail: van...@in...
> 
> http://www.iste.uni-stuttgart.de/rss/people/vanhoorn/
> http://kieker-monitoring.net/, http://research.spec.org/
> 
> ==
> Calls for Contributions:
> 
> *QUDOS* 2018: 4th International Workshop on Quality-Aware DevOps
> (Co-located with ACM/SPEC ICPE 2018 in Berlin, Germany)
> Deadline for full and tool papers: Dec 20, 2017
> http://2018.qudos-workshop.org/
> 
> *ICPE* 2018: 9th ACM/SPEC Int. Conference on Performance Engineering
> Deadline for work-in-progress and vision papers: Jan 10, 2018
> https://icpe2018.spec.org/
> ==

Re: [Kieker-users] AMQP Writer Exception

[Andre v. H. <van...@in...>]

Hi Holger,

can you help Simon?

Thanks, André

On 11.12.2017 16:31, Simon Eismann wrote:
> Hello Kieker-users,
> I am currently trying to write my Kieker logs to a RabbitMQ-server using
> the AMQP-writer. Unfortunately, i am getting the following error message:
> 
> store_1        | [localhost-startStop-1] INFO
> kieker.monitoring.core.controller.StateController - Enabling
> monitoring store_1        | [localhost-startStop-1] INFO
> kieker.monitoring.core.controller.MonitoringController - Current
> State of kieker.monitoring (1.13-SNAPSHOT) Status: 'enabled'
> store_1        |        Name: 'KIEKER'; Hostname: 'store';
> experimentID: '1' store_1        | JMXController: JMX disabled
> store_1        | TimeSource: 'kieker.monitoring.timer.SystemNanoTimer'
> store_1        |        Time in nanoseconds (with nanoseconds
> precision) since Thu Jan 01 00:00:00 UTC 1970'
> store_1        | ProbeController: disabled
> store_1        | WriterController:
> store_1        |        Queue capacity: 10000
> store_1        |        Insert behavior (a.k.a. QueueFullBehavior):
> numBlocked: 0
> store_1        | Writer: 'kieker.monitoring.writer.amqp.AmqpWriter'
> store_1        |        Configuration:
> store_1        |
> kieker.monitoring.writer.amqp.AmqpWriter.queuename='kieker'
> store_1        |
> kieker.monitoring.writer.amqp.AmqpWriter.uri='amqp://admin:nimda@10.1.3.44'
> store_1        |
> store_1        |        Automatic assignment of logging timestamps:
> 'true' store_1        | Sampling Controller: Periodic Sensor
> available: Poolsize: '0'; Scheduled Tasks: '0'
> store_1        | Exception in thread "Thread-7"
> java.nio.BufferOverflowException
> store_1        |        at
> java.nio.HeapByteBuffer.put(HeapByteBuffer.java:189)
> store_1        |        at
> java.nio.ByteBuffer.put(ByteBuffer.java:859) store_1        |
> at
> kieker.monitoring.writer.amqp.AmqpWriter.onNewRegistryEntry(AmqpWriter.java:187)
> 
> store_1        |        at
> kieker.monitoring.writer.amqp.AmqpWriter.onNewRegistryEntry(AmqpWriter.java:51)
> 
> store_1        |        at
> kieker.monitoring.registry.WriterRegistry.register(WriterRegistry.java:59)
> store_1        |        at
> kieker.monitoring.registry.WriterRegistry.register(WriterRegistry.java:28)
> store_1        |        at
> kieker.monitoring.registry.RegisterAdapter.get(RegisterAdapter.java:41)
> store_1        |        at
> kieker.common.record.misc.KiekerMetadataRecord.registerStrings(KiekerMetadataRecord.java:215)
> 
> store_1        |        at
> kieker.monitoring.writer.amqp.AmqpWriter.writeMonitoringRecord(AmqpWriter.java:146)
> 
> store_1        |        at
> kieker.monitoring.writer.MonitoringWriterThread.run(MonitoringWriterThread.java:66)
> 
> 
> I couldn't find any documentation on the AQMP writer aside from a short
> section in the user guide, so I am assuming I configured something
> badly. Is there any more in-depth documentation on the AQMP writer or an
> example that uses it?
> Otherwise did anybody encounter this error before and knows how to solve
> it?
> 
> Viele Grüße,
> Simon Eismann
> 
> PS: This is my first time posting to this mailing list if I did anything
> wrong, let me know
> 
> ------------------------------------------------------------------------------------------
> 
> M.Sc. Simon Eismann
> Doctoral Researcher
> 
> Chair of Software Engineering
> Department of Computer Science
> University of Würzburg
> 
> Am Hubland, Informatikgebäude (M2), Room 107/108
> 97074 Würzburg, Germany
> 
> Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
> http://go.uni-wuerzburg.de/eismann/
> ------------------------------------------------------------------------------------------
> 
> 
> ------------------------------------------------------------------------------
> 
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> Kieker-users mailing list
> Kie...@li...
> https://lists.sourceforge.net/lists/listinfo/kieker-users

-- 
Dr.-Ing. André van Hoorn

University of Stuttgart, Inst. of Software Technology
Reliable Software Systems Group
Universitätsstraße 38, D-70569 Stuttgart, Germany

Room: 1.332
Phone:  +49 (0)711 685-88-252, Fax: -472
E-Mail: van...@in...

http://www.iste.uni-stuttgart.de/rss/people/vanhoorn/
http://kieker-monitoring.net/, http://research.spec.org/

==
Calls for Contributions:

*QUDOS* 2018: 4th International Workshop on Quality-Aware DevOps
(Co-located with ACM/SPEC ICPE 2018 in Berlin, Germany)
Deadline for full and tool papers: Dec 20, 2017
http://2018.qudos-workshop.org/

*ICPE* 2018: 9th ACM/SPEC Int. Conference on Performance Engineering
Deadline for work-in-progress and vision papers: Jan 10, 2018
https://icpe2018.spec.org/
==

[Kieker-users] AMQP Writer Exception

[Simon E. <sim...@un...>]

Hello Kieker-users,
I am currently trying to write my Kieker logs to a RabbitMQ-server  
using the AMQP-writer. Unfortunately, i am getting the following error  
message:

store_1        | [localhost-startStop-1] INFO
kieker.monitoring.core.controller.StateController - Enabling
monitoring store_1        | [localhost-startStop-1] INFO
kieker.monitoring.core.controller.MonitoringController - Current
State of kieker.monitoring (1.13-SNAPSHOT) Status: 'enabled'
store_1        |        Name: 'KIEKER'; Hostname: 'store';
experimentID: '1' store_1        | JMXController: JMX disabled
store_1        | TimeSource: 'kieker.monitoring.timer.SystemNanoTimer'
store_1        |        Time in nanoseconds (with nanoseconds
precision) since Thu Jan 01 00:00:00 UTC 1970'
store_1        | ProbeController: disabled
store_1        | WriterController:
store_1        |        Queue capacity: 10000
store_1        |        Insert behavior (a.k.a. QueueFullBehavior):
numBlocked: 0
store_1        | Writer: 'kieker.monitoring.writer.amqp.AmqpWriter'
store_1        |        Configuration:
store_1        |
kieker.monitoring.writer.amqp.AmqpWriter.queuename='kieker'
store_1        |
kieker.monitoring.writer.amqp.AmqpWriter.uri='amqp://admin:nimda@10.1.3.44'
store_1        |
store_1        |        Automatic assignment of logging timestamps:
'true' store_1        | Sampling Controller: Periodic Sensor
available: Poolsize: '0'; Scheduled Tasks: '0'
store_1        | Exception in thread "Thread-7"
java.nio.BufferOverflowException
store_1        |        at
java.nio.HeapByteBuffer.put(HeapByteBuffer.java:189)
store_1        |        at
java.nio.ByteBuffer.put(ByteBuffer.java:859) store_1        |
at
kieker.monitoring.writer.amqp.AmqpWriter.onNewRegistryEntry(AmqpWriter.java:187)
store_1        |        at
kieker.monitoring.writer.amqp.AmqpWriter.onNewRegistryEntry(AmqpWriter.java:51)
store_1        |        at
kieker.monitoring.registry.WriterRegistry.register(WriterRegistry.java:59)
store_1        |        at
kieker.monitoring.registry.WriterRegistry.register(WriterRegistry.java:28)
store_1        |        at
kieker.monitoring.registry.RegisterAdapter.get(RegisterAdapter.java:41)
store_1        |        at
kieker.common.record.misc.KiekerMetadataRecord.registerStrings(KiekerMetadataRecord.java:215)
store_1        |        at
kieker.monitoring.writer.amqp.AmqpWriter.writeMonitoringRecord(AmqpWriter.java:146)
store_1        |        at
kieker.monitoring.writer.MonitoringWriterThread.run(MonitoringWriterThread.java:66)

I couldn't find any documentation on the AQMP writer aside from a  
short section in the user guide, so I am assuming I configured  
something badly. Is there any more in-depth documentation on the AQMP  
writer or an example that uses it?
Otherwise did anybody encounter this error before and knows how to solve it?

Viele Grüße,
Simon Eismann

PS: This is my first time posting to this mailing list if I did  
anything wrong, let me know

------------------------------------------------------------------------------------------
M.Sc. Simon Eismann
Doctoral Researcher

Chair of Software Engineering
Department of Computer Science
University of Würzburg

Am Hubland, Informatikgebäude (M2), Room 107/108
97074 Würzburg, Germany

Phone: +49 (931) 31 89655, Fax: +49 (931) 31-86603
http://go.uni-wuerzburg.de/eismann/
------------------------------------------------------------------------------------------

Re: [Kieker-users] Distribute Kieker Components

[Reiner J. <rei...@em...>]

Hi Christoph,

indeed the finalization of Kieker 1.13 has been delayed. This is due
to a wide range of changes we made in Kieker and especially in its
build infrastructure and its documentation. The project is actively
developed and you may find further information in our new ticketing
system [1] and wiki [2]. We are still in the process of moving
information from our old trac instance.

For the analysis, we are switching from our old pipe-and-filter
framework to the new Teetime framework, which allows to create new
filters much more easily. Projects, like iObserve, are already
contributing and using this new infrastructure. They use the 1.13
snapshot releases, as they are stable enough for their purposes.

We also added a DSL to specify event types and we are working towards a
DSL supporting instrumenting software systems based on various
technologies. 

For the rest of your questions see below.

[1] https://kieker-monitoring.atlassian.net/secure/Dashboard.jspa
[2] https://kieker-monitoring.atlassian.net/wiki/discover/all-updates

On Thu, 19 Jan 2017 17:39:36 +0100
"Bigchristoph" <big...@gm...> wrote:

> Good evening to all of you,
> 
>  
> 
> I have 2 questions and I hope somebody can answer them ;)
> 
>  
> 
> 1)      Does anywhere exist a tutorial or somekind like that
> describing the distributation of the components of kieker?

Maybe I du not understand the question, but we have a user guid, which
shows how to apply Kieker. In case if you want to know what to do to
use on multiple hosts, you apply Kieker to the software on the host
(this depends on the technology). In case you are using image, you
apply this to the image and distribute that onto different hosts.
Kieker is mainly configured by a kieker.monitoring.properties file
which can be customized for different hosts.

> So that
> one host is collecting the data and another one is doing the
> analysis? 

This is what we usually do. In that case the monitoring is applied to
some software on some host and it is configured to send the data to
another host. This can be done via JMS and TCP (and others). The
analysis can then run on another host and receive the data. 

> 2)      How active is this project?

See above. Very active.

> Does a roadmap exist for new releases in the future?

We will release a new Kieker version. For the roadmap and release
schedule, I will relay this question to @andré.

Kind regards
Reiner



-- 
Dr.-Ing. Reiner Jung, Software Engineering Group
Dept. Computer Science, Kiel University, D-24118 Kiel, Germany
Tel: +49 (0)431 880-2776, Fax: -7617
Email: rei...@em...
http://se.informatik.uni-kiel.de/

[Kieker-users] Distribute Kieker Components

[Bigchristoph <big...@gm...>]

Good evening to all of you,

 

I have 2 questions and I hope somebody can answer them ;)

 

1)      Does anywhere exist a tutorial or somekind like that describing the
distributation of the components of kieker? So that one host is collecting
the data and another one is doing the analysis? 

2)      How active is this project? Does a roadmap exist for new releases in
the future? Because I saw the last release was in the year 2015 and the new
version (1.13) was announced for April 2016 (now it is nearly a year later).

 

Thank you for your attention and wish you a nice evening,

 

Christoph