Prerequisites

• Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

Some environment configurations (i.e. Debian buster docker container running on host Ubuntu 18.04 with Apparmor enabled) fail to restarting JMX exporter service because cgroups conflicts. The issue is solved setting PrivateTmp=no

Versions

All.

As done with other roles, TravisCI and Ansible Galaxy integration will be included

Prerequisites

• Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

In the current role version, if the exporter scraping configuration provided is wrong, the role will execute everything and mark it as success, as the service doesn't fail but the scrape does. Adding a check after provision execution will help the role fail faster and mitigate this "silent misconfigurations".

Steps to Reproduce

1. Add wrong exporter configuration
2. Run the role in a playbook

Expected behavior: Fail, because configuration is wrong.

Actual behavior: Happily mark the execution as correct and let the exporter scrape wrongly.

Reproduces how often: Always

Versions

All

The role cannot be deployed to Ansible Galaxy because of missing parameters in the role metadata

When using this role within a playbook there could be different config files for different exporters in different hosts. In order to choose the proper config file, JMX_EXPORTER | Copy exporter config task will be modified and it will only copy the file if it is named after the host's group. This file must be placed in jmx_exporter_config_template_path directory.

Prerequisites

• Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

We need support for Debian 10 in this role.
We also must update requirements (Ansible, molecule, etc.).

Prerequisites

• Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

This role should install latest jmx_exporter version (0.2.0). There are some improvements, the most interesting IMHO is that dependence on Jetty has been dropped.

Prerequisites

• Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

Molecule 1.25 is fairly outdated and most of our roles are being tested with Molecule 2.x, so it would be a nice moment to migrate and use Goss as well.

Prerequisites

• Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

Sometimes systemctl restart jmx_exporter fails because the jmx process is not killed.

Steps to Reproduce

1. systemctl restart jmx_exporter

Expected behavior:
Service is restarted.

Actual behavior:
Restart fails with exit 1, because it kills the opened shell not the java process.

Reproduces how often:
Sometimes. Often in container environments.

Versions

All.

This line (https://github.com/idealista/prometheus_jmx_exporter-role/blob/master/templates/logrotate.j2#L12)

 [ ! -f /run/jmx_exporter.pid ] || systemctl restart jmx_exporter.service

should check /run/jmx_exporter/jmx_exporter.pid

The way configuration templates are provided at the moment is a bit tricky and not intuitive:

https://github.com/idealista/prometheus_jmx_exporter-role/blob/master/tasks/config.yml#L11

Using the way other roles provide configuration would be better, IMHO.

Prerequisites

[x] Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
[x] Checked that there is not already provided the described functionality

Description

Service doesn't go up.

Steps to Reproduce

1. deploy the role
2. systemctl status jmx_exporter:

$ sudo systemctl status jmx_exporter
● jmx_exporter.service - prometheus jmx exporter
   Loaded: loaded (/etc/systemd/system/jmx_exporter.service; enabled; vendor preset: enabled)
   Active: failed (Result: exit-code) since Mon 2019-07-15 12:01:26 UTC; 2min 34s ago
  Process: 13059 ExecStartPost=/bin/sh -c echo $MAINPID > /run/jmx_exporter/jmx_exporter.pid (code=exited, status=0/SUCCESS)
  Process: 13058 ExecStart=/bin/sh -c /usr/bin/java $JAVA_OPTS $JAVA_MEM                        -cp /opt/jmx_exporter/jmx_exporter.jar io.prometheus.jmx.WebServ
 Main PID: 13058 (code=exited, status=127)

Jul 15 12:01:26 mymachine systemd[1]: Starting prometheus jmx exporter...
Jul 15 12:01:26 mymachine systemd[1]: jmx_exporter.service: Main process exited, code=exited, status=127/n/a
Jul 15 12:01:26 mymachine systemd[1]: jmx_exporter.service: PID 13058 read from file /run/jmx_exporter/jmx_exporter.pid does not exist or...
Jul 15 12:01:26 mymachine systemd[1]: Failed to start prometheus jmx exporter.
Jul 15 12:01:26 mymachine systemd[1]: jmx_exporter.service: Unit entered failed state.
Jul 15 12:01:26 mymachine systemd[1]: jmx_exporter.service: Failed with result 'exit-code'.

Expected behavior:

Service is up

Actual behavior:

Fails with error read from file /run/jmx_exporter/jmx_exporter.pid does not exist

Versions

master: 5d0cdf7

As seen in http://blog.mague.com/?p=696, if we want to use Jconsole to debug the exporter in a Vagrant machine, we need to add -Dcom.sun.management.jmxremote.rmi.port={{ jmx_exporter_internal_port }} \ to the jmx_exporter.service.j2

Please take into account that port forwarding is needed as well.

Description

Versions > 0.17.0 has other artifact name and this role fails to install it.

Steps to Reproduce

1. [First Step]
2. [Second Step]
3. [and so on...]

Expected behavior:
[What you expect to happen]

Actual behavior:
[What actually happens]

Reproduces how often:
[What percentage of the time does it reproduce?]

Environment

• The release version/s you are using:
• OS:
• Others:

Additional Information

[Any additional information, configuration or data that might be necessary to reproduce the issue.]

Prerequisites

• [X ] Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

It would be nice to provide the jmx exporter as javaagent instead of as a service. Maybe give the choice between one of them or both

Prerequisites

• Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

As is described here is necessary to add an external library called jboss-cli-client.jar to connect to JMX subsystem exposed by WildFly.

Actually is impossible to add an external library because the service is using "java -jar" instead of "java -cp" to launch the scrapper.

Versions

<= 1.4.0

Additional Information

• https://developer.jboss.org/thread/269919

JMX has the port 9404 defined as default in the "Default port allocations" wiki.

Should prometheus_jmx_exporter_role use the default JMX Exporter Port (9404 instead of 9110)?

@jmonterrubio, @jnogol, @sorobon

Description

jmx_exporter_url variable is in vars directory and its impossible to change it.

Why is this needed?

If i want to use a repo manager like nexus or whathever.

Prerequisites

• [ X ] Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

Skeleton paths group are using user variable name instead of group variable name

Expected behavior: Should use group name variable

Actual behavior: Uses user name variable

Reproduces how often: 100%

Versions

All

Additional Information

https://github.com/idealista/prometheus_jmx_exporter-role/blob/master/tasks/install.yml#L21

Prerequisites

• [ X] Put an X between the brackets on this line if you have done all of the following:
  • Checked that your issue isn't already filled: https://github.com/issues?utf8=✓&q=is%3Aissue+user%3Aidealista
  • Checked that there is not already provided the described functionality

Description

Logging can be provided in the java service using the flag java.util.logging.config.file so it would be nice to support it. Also adding a default config file.