Configure Homebridge as a Service on Debian

Homebridge is a fantastic mechanism for getting non-HomeKit-certified smart home tech talking to Apple. It’s nerd tech written for nerds, so it’s not the easiest thing in the world to get running. Here are some instructions to make things work on Debian and friends (including Raspberry Pi and Ubuntu):

Note that these instructions are going off old memory, so they may be slightly wrong – I’ll retest and fix anything that needs fixing soon!

First thing’s first – make sure everything is up to date:

sudo apt-get update
sudo apt-get upgrade

Install Node.js

Next, install Node.js from the NodeSource GitHub (don’t use the repositories of your distro, as they’re probably outdated). After installation, update to the latest release and install build tools:

curl -sL | sudo -E bash -
sudo apt-get install -y nodejs

sudo npm cache clean -f
sudo npm install -g n
sudo n stable

sudo apt-get install -y build-essential

I like to reboot at this point, but that’s in no way essential.

Install Homebridge

The following installs Homebridge (note the --unsafe-perm flag) and the Homebridge Dummy Switches plugin. The plugin isn’t critical, but I seem to remember having difficulties getting things working without at least one plugin. Besides, the Dummy Switches plugin can be fairly handy for certain types of automation..:

sudo apt-get install libavahi-compat-libdnssd-dev
sudo npm install -g --unsafe-perm homebridge
sudo npm install -g homebridge-dummy

Next, create a very basic config at ~/.homebridge/config.json:

  "bridge": {
    "name": "Homebridge",
    "username": "CC:22:3D:E3:CE:30",
    "port": 51826,
    "pin": "031-45-154"
  "description": "Homebridge Server",
  "platforms": [
  "accessories": [
      "accessory": "DummySwitch",
      "name": "My First Switch"

Run Homebridge directly at the command line to test:


An ascii-art QR code will appear; you can use this code to add Homebridge to HomeKit using the Home app. Ignore warnings about the accessory not being certified.

Create systemd service for Homebridge

The instructions below assume you used the instructions above to install Node.js and Homebridge – paths may be different if you didn’t.


The following commands create the service user and directories, then moves the existing configuration into the appropriate locations for starting Homebridge as a service. Finally, set the relevant permissions..:

sudo useradd -M --system homebridge
sudo mkdir /var/lib/homebridge

sudo cp ~/.homebridge/config.json /var/lib/homebridge/
sudo cp -r ~/.homebridge/persist /var/lib/homebridge

sudo chmod -R 0777 /var/lib/homebridge

Systemd config files

Create /etc/default/homebridge:

# Defaults / Configuration options for homebridge
# The following settings tells homebridge where to find
# the config.json file and where to persist the data
# (i.e. pairing and others)
HOMEBRIDGE_OPTS=-U /var/lib/homebridge

# If you uncomment the following line, homebridge will log more
# You can display this via systemd's journalctl:
# journalctl -f -u homebridge

Create /etc/systemd/system/homebridge.service:

Description=Node.js HomeKit Server

ExecStart=/usr/local/bin/homebridge $HOMEBRIDGE_OPTS


Finally, enable and start the service:

sudo systemctl daemon-reload
sudo systemctl enable homebridge
sudo systemctl start homebridge

Links and Howtos

Here’s what I used to get myself up and running:

SABnzbd+ and Sick Beard on Ubuntu Linux

There are heaps of guides out there on the Internets for setting up SABnzbd+ and Sick Beard for automated download goodness – but most of them assume you’re setting up SAB and Sick Beard on a computer that’s being used as a regular PC, rather than a media server. As a result, the instructions out there could use some tweaking..

Recipe, pre-requisites and assumptions:

  • You’re using Ubuntu Linux
  • You have a user account to use on the media server (your own account, another account .. just not the root account)
  • You already have details for a news server

Step 1: Install SABnzbd+

Add the SABnzbd+ repository, update apt:

sudo add-apt-repository ppa:jcfp/ppa
sudo aptitude update

Install SAB and all its’ dependencies:

sudo aptitude install sabnzbdplus

Configure SAB as a service – set USER= to the user account you’ve chosen and set HOST= to – everything else is fine to leave as is:

sudo vim /etc/default/sabnzbdplus

Start SAB:

sudo service sabnzbdplus start

Run the configuration wizard by going to http://servername:8080/sabnzbd/ and following the prompts. You’ll need to enter the news server details; also make sure that ‘I want SABnzbd to be viewable by any pc on my network’ is selected and set a username and password (optional).

How to install SABnzbd on Ubuntu?
SABnzbd Manual – Ubuntu PPA Repository

Step 2: Install Sick Beard

Sick Beard installation needs git – if you don’t have it already:

sudo aptitude install git

Create the directory for installation:

sudo mkdir /var/lib/sickbeard

Pull down a copy of Sick Beard via git, then change ownership of the whole lot to the chosen user account:

sudo git clone git:// /var/lib/sickbeard
sudo chown -R username.groupname /var/lib/sickbeard/

Change to the Sick Beard directory, copy the init file out to where it belongs:

cd /var/lib/sickbeard
sudo cp init.ubuntu /etc/init.d/sickbeard
sudo update-rc.d sickbeard defaults

Edit the defaults file and add the following lines (changing out the username and home directory):

sudo vim /etc/default/sickbeard

Start the Sick Beard service:

sudo service sickbeard start

Sick Beard should now be available at http://servername:8081/

Installing Sick Beard on Ubuntu Server

Step 3: Configure SABnzbd

Wherein the download directories are selected and Sick Beard configuration magic is made:

Load the SAB configuration section: http://servername:8080/sabnzbd/config/

In Folders, configure the Temporary Download Folder and Completed Download Folder items. Don’t choose directories that are the final destination for the TV – and make a note of the Completed Download Folder for later. Next, configure the Post-Processing Scripts Folder item – it should be /var/lib/sickbeard/autoProcessTV, but it depends on where Sick Beard is installed.

Scroll to the bottom, Save Changes and then Restart SABnzbd.

In General, make a note of the API Key for later.

In Categories, the magic to notify Sick Beard that the download has completed is configured – this is where the Post-Processing Scripts Folder comes in to play. Configure a new line on the table as follows:

Folder/PathThe Completed Download Folder from above
Groups / Indexer tagstv

Save – SABnzbd+ is now configured.

Step 4: Configure Sick Beard

Wherein the newznab services are configured and Sick Beard learns how to play with SAB . A few assumptions:

  • You have an account with a newznab service
  • The directory structure for your TV shows looks a little like this:

Load the Sick Beard configuration section: http://servername:8081/config/

In Search Settings, check Search NZBs, select SABnzbd from the dropdown and enter details for the SAB installation..:

SABnzbd URLhttp://servername:8080/
SABnzbd UsernameUsername, if any
SABnzbd PasswordPassword, if any
SABnzbd API KeyAPI Key from above
SABnzbd Categorytv

Test SABnzbd – if everything works, Save Changes.

In Search Providers, configure a newznab service according to the documentation from the provider. None of the built-in providers seem to work particularly well; configure a Custom provider for best results.

In Post Processing, the downloaded file gets picked up and put in to the final directory structure. The base directory for the TV show is configured elsewhere; this is just the season and file name configuration. The important things to change are..:

TV Download DirFolder from above (SAB’s Completed Download Folder)
Keep Original FilesUnchecked
Move Associated FilesUnchecked
Rename EpisodesChecked
Scan and ProcessUnchecked
Name PatternSeason 02/Show.Name.S02E03.HD.TV-RLSGROUP (from dropdown)
Multi-Episode StyleRepeat

Metadata is probably important if using XBMC, but it can be configured later.

Step 5: Test!

Wherein a new TV show (that hasn’t previously been downloaded) is added and an episode is downloaded to make sure the automation works:

Load the Sick Beard home page: http://servername:8081/home/

Select Add Shows then Add New Show. Start with a show that hasn’t been downloaded to test; add the others later. In Step 1, search for and select the show to download. In Step 2, select the directory that contains subdirectories with TV shows (/drive/tv/ from the above directory structure). Set this as the default. In Step 3, select the download quality, Save Defaults and Add Show. Wait a minute while Sick Beard gets the show details from the Internets..

The show ought to have been added with all episodes set to ‘Skipped’ – change one episode to ‘Wanted’ and within a minute it should begin downloading in SAB (and be transferred in to the appropriate directory once the download is complete). 🙂

If everything works, go back to Add Shows and then Add Existing Show. Based on the defaults set when adding a new show, Sick Beard should be able to see any other show downloaded; they can all be added to Sick Beard by stepping through the wizard to make sure the show name and directory have been correctly guessed. Enjoy! 😀

Software RAID on Ubuntu Linux


I recently purchased myself a HP ProLiant MicroServer with a stack of Western Digital Red hard drives with the intention of building a media server. After stumbling through a few tutorials that didn’t work very well, a colleague took pity on me and gave me some really good advice on how to get things running – so I figured a post on how to build a software-based RAID using Ubuntu Linux might be a reasonably useful thing. 🙂


  • Server hardware and dedicated hard drives for the RAID array
    (shouldn’t really matter what specific hardware gets used)
  • Ubuntu Server 13.04 or later
  • Patience


First up, install the server edition of Ubuntu Linux. All the default install options are probably fine; there are enough guides on the Internets to cover setting this up if you get lost.

Once Linux is up and running and all of the hard drives have been detected appropriately (use lsblk to confirm device names), use parted to configure the partition table and create a partition on each disk..:

sudo parted -a optimal
select /dev/sda
mklabel gpt
  -> raid_d1
  -> ext4
  -> 1MiB
  -> 3TiB

Rinse and repeat for each drive – you’ll need to change the device being selected each time (use the output from lsblk as a guide).

For the mkpart section, the question ‘Partition name?’ is mostly irrelevant (I went with “raid_d1”, “raid_d2” etc) .. and for the question ‘End?’, use the advertised size of the drive (in my case, 3 Terabytes).

In hindsight, because I’m using the full drive for the array (vs. some for the array and some for other partitions), I’m not even sure that creating partitions using parted > mkpart is entirely necessary. It doesn’t take long though, so I guess it doesn’t hurt. 😉

Time to build the array (this is where having patience comes in)..:

sudo mdadm --create /dev/md0 --chunk=256 --level=5 --raid-devices=4 /dev/sda /dev/sdb /dev/sdc /dev/sdd

My colleague suggested a chunk size of 256k (vs. the default of 512k) for better performance. Other than that, the command should make plenty of sense: you’re creating a RAID 5 array at /dev/md0 with 4 devices (as listed at the end of the command).

In case it complains mdadm isn’t installed..:

sudo aptitude install mdadm

This will take ages .. ~18 hours for me. Monitor the progress with..:

watch -n5 cat /proc/mdstat

Once the array has completed building, you need to..:

  1. Update /etc/mdadm/mdadm.conf to list the hard drives that form part of the array (DON’T list the array device itself)
  2. Update initrd
  3. Reboot

I customised the following two lines in mdadm.conf (no need to change anything else from defaults)..:

DEVICE /dev/sda /dev/sdb /dev/sdc /dev/sdd
MAILADDR some@email.address


sudo update-initramfs -u
sudo reboot

After the reboot, your RAID array should turn up at /dev/md0. Confirm with the following..:

sudo mdadm --detail /dev/md0

If it’s not there, try..:

sudo mdadm --detail /dev/md127

If the array turns up at /dev/md127, it could mean that you have problems with mdadm.conf – see this Ubuntu Forums thread for further reading.

Now the array is configured, time to format it. Choice of filesystem is up to you, but I chose XFS (it plays nice with RAID and large file sizes). First, install XFS as it’s not a part of the standard Ubuntu Server image, then create the filesystem..:

sudo aptitude install xfsprogs
sudo mkfs.xfs -L data /dev/md0

Finally, get the UUID of the file system and load that in to /etc/fstab. Use blkid to find the UUID that belongs to /dev/md0..:

sudo blkid
/dev/md0: LABEL="data" UUID="3d3cf1c1-6015-4b5d-ac08-e38832fa29d6" TYPE="xfs"

Now, add that to /etc/fstab (I’m using /data as my mount point, but whatever works best for you)..:

# RAID array
UUID=3d3cf1c1-6015-4b5d-ac08-e38832fa29d6 /data xfs defaults 0 0

Reboot once more for good measure. Hopefully you get something like this as your df -h output..:

Filesystem Size Used Avail Use% Mounted on
/dev/sde1 28G 1.4G 25G 6% /
none 4.0K 0 4.0K 0% /sys/fs/cgroup
udev 929M 8.0K 929M 1% /dev
tmpfs 188M 284K 188M 1% /run
none 5.0M 0 5.0M 0% /run/lock
none 939M 4.0K 939M 1% /run/shm
none 100M 0 100M 0% /run/user
/dev/md0 8.2T 15G 8.2T 1% /data