Comment 0 for bug 988881

# /etc/init.d/clvm status ; echo $?
 * Cluster LVM Daemon is not running
0

/etc/init.d/clvm is showing, that Cluster LVM Daemon is not running, but it always exits with 0.
So processes that controll CLVM using "/etc/init.d/clvm status", that rely on correct exit codes, are going to fail.

In consequence it is not possible to use "/etc/init.d/clvm" as a script to a resource in "cluster.conf".

You cannot do sth. like:

  <resources>
   <script file="/etc/init.d/clvm" name="clvm"/>
  </resources>

  <service autostart="1" domain="only1" exclusive="0" name="clvm1" recovery="restart">
   <script ref="clvm"/>
  </service>

Is it possible to change this behavior like that:
# /etc/init.d/clvm status ; echo $?
 * Cluster LVM Daemon is not running
1

E.g. like proposed in my patch?

Best regards
Christian