alfredkrohmer

–> systemd-nspawn -bD ubuntu1404_test
Spawning container ubuntu1404_test on /frzclient/images/ubuntu1404_test.
Press ^] three times within 1s to kill container.
init: Unable to create device: /dev/kmsg
 * Stopping Send an event to indicate plymouth is up                                                                                                                        [ OK ]
 * Starting Clean /tmp directory                                                                                                                                            [ OK ]
 * Stopping Clean /tmp directory                                                                                                                                            [ OK ]
 * Stopping Populate /dev filesystem                                                                                                                                        [ OK ]
 * Starting Signal sysvinit that virtual filesystems are mounted                                          

alfredkrohmer

# synchronizes the given block, but only when called with the same argument
def synchronize_with_unique(arg, mutex, map, &block)
  # get the lock for the supplied argument while increasing the refcount
  lock_info = mutex.synchronize do
     info = (map[arg] ||= { lock: Mutex.new, count: 0 })
     info[:count] += 1
     info
  end
  
  # do the actual locking

alfredkrohmer

#!/bin/sh

# This script decides if we are part of the process of bootstrapping a new etcd cluster
# or if we are joining an already existing cluster.

set -eux -o pipefail

# own IP address
own_ip=$(curl -f -s http://169.254.169.254/latest/meta-data/local-ipv4)

alfredkrohmer

require 'sinatra'
require 'sinatra-websocket'
require 'docker-api'
require 'pty'

set :server, 'thin'

def new_state
  {
    lock: Mutex.new,

alfredkrohmer

Nov 16 17:56:39 desktop kernel: rtusb init mt7610u --->
Nov 16 17:56:39 desktop kernel: 
                                
                                === pAd = ffffc90003e34000, size = 1371608 ===
Nov 16 17:56:39 desktop kernel: <-- RTMPAllocTxRxRingMemory, Status=0
Nov 16 17:56:39 desktop kernel: <-- RTMPAllocAdapterBlock, Status=0
Nov 16 17:56:39 desktop kernel: ==>rlt_wlan_chip_onoff(): OnOff:1, Reset= 1, pAd->WlanFunCtrl:0x0, Reg-WlanFunCtrl=0x20a
Nov 16 17:56:40 desktop kernel: RtmpChipOpsEepromHook::e2p_type=0, inf_Type=2
Nov 16 17:56:40 desktop kernel: RtmpEepromGetDefault::e2p_dafault=1
Nov 16 17:56:40 desktop kernel: NVM is EFUSE mode

alfredkrohmer

Physical layer

The dongle itself is sending out data using 802.11a (5 GHz WiFi) with OFDM and 6 Mbit/s data rate:

Radiotap Header v0, Length 38
    Header revision: 0
    Header pad: 0
    Header length: 38
    Present flags

alfredkrohmer

#!/bin/sh

FLASH_TIME=$(opkg info busybox | grep '^Installed-Time: ')

for i in $(opkg list-installed | cut -d' ' -f1)
do
    if [ "$(opkg info $i | grep '^Installed-Time: ')" != "$FLASH_TIME" ]
    then
        echo $i
    fi

alfredkrohmer

Proposal: Deduplicated storage and transfer of container images

Docker uses layers to build containers and images on top of other images. This saves storage and transfer of layers already existing on the system. Rkt uses no layers and needs to fully download any new image.

This document proposes to use deduplication to store and transfer container images. This has multiple advantages over layering:

alfredkrohmer

#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <fcntl.h>

int main(int argc, char **argv) {
    if(argc < 4 || argc % 2) {
        printf(
            "Usage: %s <path to GPIO value file> <interval 1> <interval 2> ... <interval n>\n\n"
            "Provide the path to the 'value' file of the GPIO in sysfs and an even number of intervals\n", argv[0]);

alfredkrohmer

#!/bin/bash

set -ex

# ----------------------------------------------------------------------------
# Configuration:

case "$CONNECTION_ID"
in
    "Some VPN name") # enter name of VPN connection in NetworkManager here