Jenkins : Hudson Speaks! Plugin

Plugin Information

Distribution of this plugin has been suspended due to unresolved security vulnerabilities, see below.

The current version of this plugin may not be safe to use. Please review the following warnings before use:

This plugin gives Jenkins a voice using FreeTTS.

Prerequisites

This plugin requires that:

  • Hudson's server machine has a working sound card and speaker.
  • Hudson's server machine's speaker is not in a remote server room where there's no one around to hear it. Preferably in audible range of the development team Hudson is working for.

IMPORTANT: Please use the 'Test speech' button in the global preferences before using Hudson Speaks! as a build notifier. If the test does not complete, or no sound is heard, you should not use Hudson Speaks! as a build notifier, and check your audio hardware.

During testing it has been found that if a linux machine does not have a correctly configured audio device, the FreeTTS library call can hang trying to speak. This will cause your build executor to hang too, and result in Hudson failing.

Configuration

Hudson Speaks! is configured using a Jelly XML Script (just like a lot of the internals of Hudson itself).

The context that the Jelly script runs in is pre-configured with these variables:

  • ${build} - the current build
  • ${duration} - the duration of the build formatted for speech.
Example script (the default):
<j:choose>
    <j:when test="${build.result!='SUCCESS' || build.project.lastBuild.result!='SUCCESS'}">
        Your attention please. Project ${build.project.name}, build number ${build.number}: ${build.result} in ${duration}.
        <j:if test="${build.result!='SUCCESS'}"> Get fixing those bugs team!</j:if>
    </j:when>
    <j:otherwise><!-- Say nothing --></j:otherwise>
</j:choose>

This means an announcement will only be made if the current build, or the previous build was not a success. In other words the project was broken by this build, the project is still broken, or the project was fixed by this build.

The script can be specified at the global level, and also overridden at the project level.

Troubleshooting

This Plugin should work out-of-the-box on Windows, but often under Linux the sound device is not accessible.

If Hudson refuses to talk on Linux, but the 'Test speech' says success, check that the sound device (often /dev/dsp) is writable by your Hudson user:

$ ls -l /dev/dsp
crw-rw-rw- 1 root root 14, 3 Oct 22 16:35 /dev/dsp

If not, get someone with sufficient system privileges to change it for you:

$ chmod +rw /dev/dsp

Alternatives

An alternative to this is to use the CCTray app that comes with CruiseControl.NET and have it monitor your Hudson server. You can set this up on a shared machine or on individual developer machines. To have CCTray monitor Hudson, set it up to monitor a custom URL that looks like http://hudsonserver:hudsonport/hudsonpath/cc.xml. See Monitoring Jenkins for more details.

Basically, appending cc.xml to almost any Hudson URL (project, view, etc.) will return an XML document that CCTray can parse. You can then use CCTray to play .wav files, speak, or even control X10 devices.

CCTray is available at http://sourceforge.net/projects/ccnet/files/

Changelog

  • Version 0.1.1 (2nd December 2009)
    • Repackaged with Hudson 1.324 (no functional changes)
  • Version 0.1 (29th November 2009)
    • Initial release