-
Notifications
You must be signed in to change notification settings - Fork 0
Installation
Same guide as on README.md.
To install, you will need at least one slave and one master. All slaves will connect to the master (make sure firewall allows SSH connections from all slaves)
Installing nafsdm can be done using the guide below.
Before installing, make sure you do not have any existing installation of nafsdm (the installer will not upgrade current installations)
Connect to your master server and initiate the installation script.
bash <(curl -s https://raw.githubusercontent.com/MrKaKisen/nafsdm/master/setupMaster.sh)
Follow the steps in the installation.
If the above command fails, you can also manually download the script and execute it (using for example wget).
Once it's finished, run the master once, as it will generate keys (the installer also reminds you of this).
nafsdm-master
Now, copy the SSH key contents somewhere to your computer as it will be needed on the slaves later on.
cat /home/master-nafsdm/.ssh/nafsdm_rsa (this will print the key)
The master installation is now finished and should be ready to use. You can use nafsdmctl to add/remove domains or check the status of connected daemons.
You will have to perform these steps on every slave you would like to install this on. Connect to your slave and initiate the installation script.
bash <(curl -s https://raw.githubusercontent.com/MrKaKisen/nafsdm/master/setupSlave.sh)
Follow the steps in the installation.
Once it's finished, you can open up the config with your editor of choice (example below uses nano) to set it up correctly.
nano /home/slave-nafsdm/config.conf
Here is an explanation of what every option is.
host = hostname / IP for the master node we confgiured earlier
user = the username where nafsdm on the master installed. By default, it's master-nafsdm.
update_interval = how often the slave will update it's configuration (seconds).
type = type of the slave system. Is it ubuntu, debian or centos?
bindPath = path of configuration file nafsdm will write to (bind configuration)
nodeName = name of this node
The development section isn't something you should edit unless you're absolutely sure of what you're doing.
Once you've updated the config, you can paste over the SSH key you saved earlier to the slave. Paste it in to the file mentioned below.
nano /home/slave-nafsdm/.ssh/master_key
The key also needs to have correct permissions.
chmod 600 /home/slave-nafsdm/.ssh/master_key (nafsdm will NOT work if the correct permissions are not used)
You're done! You should now be able to start the slave (if everything is correctly configured).
nafscli start
As of version 1.2.5, you can control most of the basic operations on the slave using nafscli. Run nafscli for a list of available commands.
Replace start with stop or restart if you would like to do that later on (or status to check if it's running)
If anything fails, you can check the log.
nafscli log
When running the slave the first time, you will probably have to accept the "fingerprint". To accept it, stop the daemon.
nafscli stop
And run the slave in debug mode, which will let you accept the fingerprint.
python /home/slave-nafsdm/pythondaemon/__main__.py
When it asks you, just type "yes" and hit enter. When it's done, hit Control+C to stop the daemon and then boot the daemon back up again using service (as usual). It should look something like this:
The authenticity of host 'example.example (0.0.0.0)' can't be established.
ECDSA key fingerprint is SHA256:xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.
Are you sure you want to continue connecting (yes/no)?