You could use the free cloud InfluxDB and Grfana setups, then my pre-made dashboard. You can view all the hub info on one dashboard using that setup. If you dont mind sending some data to a cloud service.
2 Likes
I not too worries about the cloud for this type of data... Especially for the remote hubs... I hardly ever have any issues with them, but it would be nice to be able to check their stats quickly without having to remote int...
Where do I start with your suggestion?
This is a very detailed write up for Influx and Grafana. It is pretty easy especially if using the cloud service. Using HE with with InfluxDB, Grafana, with InfluxDB Logger
Then here are some pre-made dashboards you can use once setup: Ready-to-Use Grafana Dashboards! - Beta Test / Feedback
3 Likes
Thank you!
1 Like
Over the last few platform beta updates, I started graphing dbSize, and I have noticed that the dbSize parameter seems to stop updating for varying periods while every other parameter tracks keeps updating. In the attached, you can see two examples. They are the straight horizontal lines over an extended period. I have looked at the stored data and the device, and this is a Hub info thing. There are gaps in the data stored, and the field in the device has not been updated, no matter what I do.
Have I hit some kind of limit at 180? I know before the new Database engine the excessive db size warning was set to 160. @thebearmay , might there be related limits on the range that Hub Info can report? Any other suggestions?
LJ
EDIT2: In looking at the events list on the device it shows that the driver has not been receiving updated dbsizes for over 24 hours, and the the states listing of the device never changes from 177.
I just create a dashboard in each remote hub and use legacy cloud dashboard links to access snd check status.
Was this for me? It doesn't seem related, but no one else has posted here in 3 weeks. 
LJ
No for superdupe asking how to handle remote hub monitoring.
1 Like
If the database doesn’t change its reported size you won’t see an dbSize attribute change event.
1 Like
So just to be clear, are you saying that the only way Hub Info would not record a change in dbSize is if the HE firmware stopped updating dbSize? How do I validate that this the case? If that is the case, do you know if anyone else has seen this freeze up in dbSize reporting or any other state? Should I report this to Victor and others in the beta groups?
LJ
It would not record a change if the value has not changed. When you push an event to the hub it filters out duplicated events with the same value by default unless you force it as a state change. So the event is still pushed but it is hidden from the event log.
Just check the endpoint manually: /hub/advanced/databaseSize
You can see if that is matching up with your chart.
Possibly the hub has lost track of the db size for some reason and just keep reporting the same value. A restart will probably clear it up if thats the case.
1 Like
Iirc the driver polls that information according to the frequency you configure, so it’s not going to reflect every change in db size in real time.
1 Like
Thanks, so the HE is reporting 177 consistently. Thus not an Hub Info problem. I will reboot, but I was holding out on that to see how high dbSize will go on 2.3.9 beta. Since I can't seem to find that anymore it is hardly worth continuing with that effort. I will run this by Victor and Copeland in the beta forums.
Thanks guys,
LJ
Yes, but it has been almost 48 Hours without change.
LJ
That may be accurate if it has reached a point where the additions and subtractions to the database are equal. Mine very rarely differ.
If the hub isn't reporting a change in DB size, then there likely isn't a change in DB size...
Looking at all my history, it seems unlikely that the hub would not show some variability, and never show any compression, even at the nightly db cleanup, and stay at exactly 177 for 48 hours.
LJ
Do you ever see exact flatline on dbsize for 48 hours with no indication of database compression? I record every five minutes.
You guys are far more knowledgeable than I, so if you guys find this behavior normal I am inclined to let it run for a while.
LJ
@thebearmay, can you tell me how to extract a piece of data from the timeZone string using RM? I tried using the token function but I can't get anything extracted. Maybe it can't be done?
Never liked trying to use RM to do string manipulation, but can be done and you're on the right track. It will take you two token functions one to break on the commas and another to break that value at the colon.
Best example I could find is from the RM master himself - Extracting a token from a String Variable
1 Like
