The bootloader menu has to be able to list snapshots created with tagged attribute by snapper or other tools. These attributes control how snapshots are displayed and organized in the boot menu. All the changes are reflected in real time, that is when a snapshot is created it's ready to boot without the hassle to call grub2-mkconfig to refresh menu. The operation and navigating to the snapshot is intuitive and out of box and booting is even more easy to hit the button. The hackweek is based on current implementation to improve the code structure and testing, like code clean up and make it in better shape get upstreamed, improving the menu to use extended attribute to organize the boot menu with defined attributes which can be tagged by snapshot tools like snapper etc.

Rescue system has a lot of options , but most of them can only be set at boot time, first idea is too provide a way to configure them while the system is running.

some ideas

I already have a [python script][0] processing mails received in the Calendar folder to get the ical event and push it to [radicale][1]. It has several drawbacks: * It doesn't detect appointment changes (pretty easy to fix as each event has a unique ID)

Setup and operate a Mode-S Beast that receives the ADS-B broadcast signals sent out by airplanes.

Building on top of Rouslan Dimitrov's Microcontroller Class D Amplifier, I developed the idea to feed the power amplifier with PCM data directly. Rouslan does some dithering and noise shaping, but I'm convinced that proper interpolation would sound better. The ATtiny is too slow to do even cubic interpolation so I'll move that into the host.

Build Service needs an openid. Imagine following case:

Examining vmcore with crash (very) often means to look at multiple outputs at once, which in console enviroment simply is not easily achievable. Since last hackweek, there is a

• jruby
• ruboto homepage

Currently, opensuse-quickstart sets up a one-machine Cloud based on OpenStack, either with KVM or virtualisation. In order for bootstrapping further hosts, it would be much easier to set up only a bare-metal cloud on one machine, and PXE boot a 2nd machine via OpenStack Nova/Ironic.

Customers may face problems with their systems we weren't aware of, if we focus on testing single components. We simply can't copy each and every customer setup, but we want to have test results that are more relevant.

i want to make a prototype of a watch for waking up me in the night in order to take care of my daughter. this watch includes arm based MCU with time display on panel(OLED or LED), vibrate, beeper. using 3D printer for the box of this watch.

Description

One of the PuppetLabs guys is trying to enforce best practices on administering Puppet modules, by using base modules and build services and roles on top of them. That way we can re-use external / third-party modules from forge.puppetlabs.com, ideally without touching them at all. He names that "the Lego approach". Full article.

Hamsta currently lacks a standardized network interface that would allow other systems accessing its capabilities. There is a custom network interface with text-parsing based protocol that is poorly documented and needs to be implemented on the client side to be usable. This interface does not use any standardized way of communication or library. This project suggests to add a standardized web service interface (let us call it API) to Hamsta. Expected attributes are following.

Wicked is a network configuration infrastructure incorporating a number of existing frameworks into a unified architecture, providing a DBUS interface to network configuration. https://github.com/openSUSE/wicked