It took me some time to figure this out, maybe it can be useful to someone else...
I kept getting "no ping" error for unknown reasons. It turned out that I had enabled DumpPreUserCmd and UserCmdCheckStatus for that specific host to check for an encrypted partition being mounted.
Due to some changes on the host after an upgrade, the script triggered by DumpPreUserCmd was returning an error... And for some reason this showed as a ping error in the web interface.
I eventually figured it out by checking the last lines of the LOG.** file in the pc/[host] directory, which looked like this:
2018-05-08 19:15:45 Output from DumpPreUserCmd: /[partition] is not mounted 2018-05-08 19:15:45 DumpPreUserCmd returned error status 256... exiting
If it still fails after you've set "root_url" to the correct value in grafana.ini, you might want to check whether you can run phantomjs from the command line.
If you get "QXcbConnection: Could not connect to display / PhantomJS has crashed", then the explanation is here: Debian Bug #817277.
To fix it, I installed xvfb (apt-get install xvfb), and edited /usr/bin/phantomjs so that the last line now looks like this:
Apache config extract (you will need to enable mod_proxy, mod_proxy_http and mod_headers for this to work):
<VirtualHost *:80> ProxyPreserveHost On ProxyRequests Off ProxyPass / http://localhost:3000/ ProxyPassReverse / http://localhost:3000/ RequestHeader set "X-WEBAUTH-USER" "admin" </VirtualHost>
On a separate Apache instance exposed to more networks I did the following:
Configured Apache as reverse proxy to the internal instance
Restricted access from specific IP addresses
Setup a rule to redirect requests to the root of the website (and only those) to a specific dashboard
This is how the Apache config looks like (requires mod_proxy, mod_proxy_http and mod_alias; IP addresses, host names etc. changed)
<VirtualHost *:80> ServerName sub.example.org ServerAlias www.sub.example.org <Location /> Require ip 192.0.2.0/24 Require ip 203.0.113.0/24 Require ip 2001:0db8:85a4::/64 Require ip 2001:0db8:85a5::/64 RedirectMatch ^/$ /dashboard/db/mydashboard </Location> ProxyPreserveHost On ProxyRequests Off ProxyPass / http://[2001:0db8:85a3::aaaa:8a2e:0370:7334]/ ProxyPassReverse / http://[2001:0db8:85a3::aaaa:8a2e:0370:7334]/ </VirtualHost>
Using a public IPv6 address on the internal host allows the whole thing to work with just a few firewall rules, without the need to mess with NAT or a VPN.
On Raspbian, you don't need to install the 8192cu driver manually, as it comes with the distribution
To avoid disconnection issues, you need however to disable power management. This needs to be done in two places. First, this is what I have in /etc/modprobe.d/8192cu.conf:
options 8192cu rtw_power_mgnt=0 rtw_enusbss=0
This alone did not prevent the issue. I also had to add this entry to /etc/network/interfaces:
auto wlan0 allow-hotplug wlan0 iface wlan0 inet dhcp wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf wireless-power off iface default inet dhcp
Somehow, most of the time my Arduino-based EggBot clone outputs one byte of undetermined garbage before saying "Ready". This caused gcode-sender.rb to fail with "invalid byte sequence in UTF-8 (ArgumentError)".
Eventually I found the solution in a post by Wayne Brissette here: https://www.ruby-forum.com/topic/4409620. Basically you need to treat the string as binary instead of utf-8. This small change in the code that waits for the robot to be ready fixed the issue for me:
--- gcode-sender.rb.orig +++ gcode-sender.rb @@ -30,7 +30,7 @@ sp.read_timeout = 0 # Necessary for Windows. while line = sp.gets print line - break if line.match(/^Ready/) + break if line.force_encoding("BINARY").match(/^.?Ready/) end File.foreach(filename) do |line| line.chomp!
