nut-server packaging incorrectly assumes host will be client, too

Bug #1418175 reported by Nic Bernstein on 2015-02-04
This bug affects 1 person
Affects Status Importance Assigned to Milestone
nut (Ubuntu)

Bug Description

The packaging for nut-server assumes that host will also be a client, which is an invalid assumption. As is stated <a href="">on the nut website</a>:
<quote>Finally, there’s a special case. Type A systems are connected to a UPS’s serial port, but don’t depend on it for power. This usually happens when a UPS is physically close to a box and can reach the serial port, but the wiring is such that it doesn’t actually feed it.</quote>

There is no good reason for nut=server to depend upon nut-client.

Furthermore, /etc/init.d/nut, in the start-stop-client() function should not include "netserver" in the case $MODE... statement, or else a different mode should be introduced which starts only the driver and upsd components, as is described in the documentation.

It is a potentially dangerous and critical assumption that a nut host will always be powered by, or only be powered by, a ups which it is also monitoring. The software supports much greater freedom than this, and the packaging should reflect this.

This is specifically in reference to the Precise (12.04.2 LTS) packaging of nut-2.6.3-1ubuntu1.1, but the same dependency is observed to exist in nut-2.7.1-1ubuntu1 for Trusty (14.04.1 LTS).

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers