I have had 2 recent instances where the finicky data producer down problem kept me from seeing events. There is nothing in the logs to indicate why, and I know this sometimes happens.
What I don’t understand is why such a mature product does not inform the owner when it has gone offline, either locally or from the server back-end. This seems like such an easy thing to implement. Why has this not been done, or did I miss a setting?
Thanks.
Well, you do get such a message if your machine has been down for 3 or more days. Here is the message you get:
*Hello Shaker,
*We are committed to helping keep every Raspberry Shake up and running. We have not seen your device AM.R2DDD connected to the network in 3 or more days.*We would love to have you back!
*You may be getting this email because:
*Your Raspberry Shake has not shared data to our Servers for multiple days
- If your Raspberry Shake device is intentionally disconnected, then please ignore this message.
- If your Raspberry Shake device is connected and you think it should be transmitting data, please unplug the Raspberry Shake and plug it back in.
- If this does not resolve the situation, please sign into ShakeNet and report the issue to our " Tech Support".
I’m OK with 3 days because my location has flaky internet and power supply, so the Shake often goes down for hours at a time. I don’t want to be flooded with emails telling me something I can do nothing about.
Thanks for the information TideMan. I didn’t realize that I’d get a message after three days, but that seems very long for me. I also believe my shake has been in a non-reporting state for longer than that without any sort of notification. I also haven’t lost internet service for any length of time in at least the past 5 years, and my house and the shake are on backup power (with a UPS to tide the shake over until the generator is up), so it never goes down.
I’d like to be able to determine the value I would prefer for notification, and to be able to set it from a few hours to days if desired. My other Raspberry Pi community projects (Satnogs, Flightaware, Birdnet) all allow that choice.
Regards.
Yes, being able to set the length of downtime before a message is sent would be a good feature.
Are you listening Stormchaser?
I surely am K2PI, TideMan!
To summarize (also for other readers): yes, we do send automated emails when a Shake is offline for 3 days, with a further reply after 30 days of inactivity. Some of these emails often end up in the Spam or Undesired Mail folder with some providers (probably because they are seen as unwanted e-mails similar to the commercial ones usually sent out in bulk), so it’s worth checking those if you haven’t already.
After various tests and considering feedback from our users, we set the two timeframes above as the most optimized/well-received between getting too many emails (such as TideMan’s case) because of local internet/power reasons and waiting too long before a notification is sent on the opposite side.
However, K2PI’s suggestion is sound, and having the possibility of autonomously setting an alert interval is an interesting addition to the Shake system. I don’t know how difficult this would be to implement, but I’ll report the suggestion to our team as usual, so that they can see what can be done on the matter.
Thank you both for this!
