With the changes to sensor names for forecastio, I made a new version to use the updated sensor names. Interestingly the old ones still seemed to work but you can never be too careful - they may be deprecated and be removed at some time in the future.
Version 1.72
What’s this? Add a link?
Here you go -
I have asked @fabaff if we can list the various projects we have going in the forum sticky.
One of the reasons I requested the the third part category… now I can ask my stupid question and not look like a fool for it being ten posts above! LOL 
I had this working great until I did the last git pull update; now it won’t start as a service anymore. Runs fine in dev mode and starts right up but not starting on reboots like it used to. Not sure where to start looking as I changed nothing in the init scripts except the path.
What exactly is broken ? Does it run from the command line OK?
Yes, as I said I can run each and load them in dev mode - command line. The scripts you provided to start it up on reboot like a service - that were working fine just before the update - stopped working. I copied the example scripts to the directory as specified in the instructions and edited the path to reflect my install location. So those didn’t change as far as I can tell.
Here’s the dashing init (given the ext YAML so I could upload it):
dashing.yaml (1.8 KB)
And here’s the hapush (same deal, ext added to upload)
hapush.yaml (1.7 KB)
Nothing much changed that would have caused this - we need logfiles!
Does it say anything interesting in /logs/thin.log?
Also, what about /hapush/hapush.log?
Not going to bother with hapush, because it’s mostly complaining that it can’t find the dashing instance which is obvious. Give me a few seconds to look at the thin log before posting it.
And as always, thanks.
It looks like the connection to the API is being refused but I’m not sure. I’ve attached the last parts of the log.
thin.log.yaml (28.5 KB)
I concur - did your Home Assistant instance start up correctly after the upgrade?
Do you mean the dashboard pull or the recent HA upgrade?
Cause it would actually be yes on both sides. But I did have an all day CloudMQTT outage yesterday that was mucking up my HA setup. Didn’t know that if MQTT doesn’t init, I lose my ZWave switches as well. Still got the sensor info, just no on/off switches. Huh?!?
But everything is working today. Except starting dashing, of course.
What happens if you manually run the init script?
$ sudo /etc/init.d/dashing start
Sorry, stepped away for a moment.
Now this is weird, because the error seems like it’s already running, but when I go to the internal IP address at port 3030, the connection is refused.
Starting daemon: dashingbundler: failed to load command: thin (/usr/local/bin/thin)
Thin::PidFileExist: /home/pi/hadashboard/dashing.pid already exists, seems like it’s already running (process ID: 1065). Stop the process or delete /home/pi/hadashboard/dashing.pid.
/var/lib/gems/2.1.0/gems/thin-1.6.4/lib/thin/daemonizing.rb:171:in remove_stale_pid_file' /var/lib/gems/2.1.0/gems/thin-1.6.4/lib/thin/daemonizing.rb:43:in daemonize’
/var/lib/gems/2.1.0/gems/thin-1.6.4/lib/thin/controllers/controller.rb:62:in start' /var/lib/gems/2.1.0/gems/thin-1.6.4/lib/thin/runner.rb:200:in run_command’
/var/lib/gems/2.1.0/gems/thin-1.6.4/lib/thin/runner.rb:156:in run!' /var/lib/gems/2.1.0/gems/thin-1.6.4/bin/thin:6:in <top (required)>’
/usr/local/bin/thin:23:in load' /usr/local/bin/thin:23:in <top (required)>’
Should I delete the pid file and reboot?
EDIT #1: It looks like it was the stale PID that was causing the issue. I don’t remember having a bad shutdown, but it’s certainly possible I did and that was the result; a leftover pid file.
I’m going to reboot and see what happens…
EDIT #2: Confirmed. I’ll still need to check HAPush, but I’m guessing it will prove to be okay as well.
I never would have figured this out without you telling me to manually run the init script as I wouldn’t have seen it. Much, much thanks!
EDIT #3: I’m thinking now that I may have done this myself. I would suggest adding something to the docs about upgrading that suggests users stop both services before doing the git pull and then restart. I’m wondering if doing the upgrade with them both running caused the stale PID?
Glad you figured it out 
Is there something else I need to do to use the iFrame widget like @jbags81 did for the weather radar? All he mentioned in his post was that he incuded this in his dashboard:
<li data-row="3" data-col="2" data-sizex="2" data-sizey="2">
<div data-id="iframeId1" data-view="Iframe" data-src="https://radblast.wunderground.com/cgi-bin/radar/WUNIDS_map?station=DTX&brand=wui&num=40&delay=50&type=N0R&frame=0&scale=0.5&noclutter=1&showstorms=0&mapx=400&mapy=240¢erx=360.5¢ery=315.5&transx=-39.5&transy=75.5&showlabels=1&severe=0&rainsnow=1&lightning=0&smooth=0&rand=24399365&lat=0&lon=0&label=you"></div>
</li>
I’ve done this, but I just get a blank widget.
I’m almost done with this to the point I can mount my tablet and go, but this bit is holding me up… 
I even tried using the image widget but no luck.
Don’t know, I’ve never used it - I’ll take a look when I have some time.
I didn’t think you had; that’s why I tagged @jbags81 since his example had it running. But thanks for checking in.
Man, what a job keeping things straight! I had to comment everything so I could group together things in the same rows and then order them in column position so I knew what was going on. I’m going to have to move the page buttons on the side for my crappy little tablet as it’s rendering a bit of the bottom cut off.
But it’s great to finally get this done so I can check another thing off the list. An actual list, because I was getting a bit slack.
Sounds like a lot of work Did you know you can change the aspect ratio of the tiles so they feed flawlessly on your tablet?