EVmote App

[rpo]

I have not had any trips recorded since the morning of December 1st. The site appears to be working OK otherwise. Just wanted to let you know in case something was going on you weren't aware of!

Same here. I also have never been able to get the "Share" my trip to work either. The recipient never receives the email.

 

[Andyw2100]

I have not had any trips recorded since the morning of December 1st. The site appears to be working OK otherwise. Just wanted to let you know in case something was going on you weren't aware of!

Same here.

I also have not had trips recorded since December 1.

 

[rodb]

I have not had any trips recorded since the morning of December 1st. The site appears to be working OK otherwise. Just wanted to let you know in case something was going on you weren't aware of!

+1...Nothing since 12/1

 

[Wojski]

+1...Nothing since 12/1

Just checked mine, same as above.

 

[redevries]

Is this still actively being developed?

 

[Andyw2100]

Is this still actively being developed?

Ryan (cryptyk) hasn't been responding to the reports of the trip logging errors for about a week. Perhaps he has had some sort of personal emergency or something that has taken precedence over this work for a short time. With all the time and effort he has put into the project thus far, I can't imagine that he would just "stop development", and I definitely can't imagine him doing that without letting us know what was going on.

I expect we'll hear from Ryan some time soon, when he is able. In the mean time I think we all just need to be patient, and not jump the gun with posts like the above, that might seem to significantly exaggerate the severity of the issue.

 

[cryptyk]

Hi all,
Sorry for the absence. As Andy alluded to, I've been dealing with "real life" stuff for the last two weeks.

I'll figure out why trip tracking isn't working today. New feature development may be slow for the next few weeks, however.

Thanks for your patience,
Ryan

 

[Andyw2100]

Hi all,
Sorry for the absence. As Andy alluded to, I've been dealing with "real life" stuff for the last two weeks.

I'll figure out why trip tracking isn't working today. New feature development may be slow for the next few weeks, however.

Thanks for your patience,
Ryan

Thanks for taking the time to update us, and to do what you can.

Good luck with whatever the "real life stuff" is.

 

[cryptyk]

I've fixed the GPS tracking issue. Sorry about that!
It will continue to be more and more reliable as these things pop up and I fix them

Thanks,
Ryan

 

[mikeash]

Thanks for the fix. My driving from today showed up, so I can confirm it's working for me again.

No worries on the delay. You don't owe us anything! (At least not until you let us pay for something.)

 

[Andyw2100]

I've fixed the GPS tracking issue. Sorry about that!
It will continue to be more and more reliable as these things pop up and I fix them

Thanks,
Ryan

Unfortunately I think it may not be working again. My trip today was not recorded.

 

[cryptyk]

Confirmed. I'm going to restart it now and try to figure out why it's dropping some people off.

 

[cryptyk]

Should be fixed. When the database experienced a hiccup, it would throw some errors and knock those particular users offline.
Shouldn't do that anymore

 

[Andyw2100]

Should be fixed. When the database experienced a hiccup, it would throw some errors and knock those particular users offline.
Shouldn't do that anymore

Thanks for the quick fix!

 

[MikeBur]

Could this be because we're swamping the TM servers with requests from multiple requesters, eg yesterday I ended up running VT in parallel for my insane launch testing to get the energy use. Both VT and this failed logging huge chunks of time.

Ideally, I'd like to be able to stop evmote recording / polling whilst I use VT, or other data logging tools, as I want to be a good client

If memory serves, and hopefully helps your debugging, this was ~9:15am (PST) yesterday

 

[cryptyk]

I've never seen any issues arise from overloading the Tesla APIs in evMote. I manage the traffic pretty well. That's another way of saying that every issue with evMote so far has been 100% because I screwed something up

There is a lot of instability with the Tesla API though. It doesn't feel like it's a priority for them. Here's a dashboard I use to keep an eye on evMote:

At the top right, you can see the constant errors that come back from the Tesla API. I usually just auto-retry those requests and they work. Not ideal.
Also in the middle right, you can see errors I get back from the GPS tracking API. I usually just drop that particular GPS coordinate. You wouldn't notice because the tracks come in pretty quickly and one point isn't noticeable.
Also, note the dip in the bottom right graph. That's the graph for number of trips that were stopped. You can see that when the GPS tracker crashed, it stopped logging trips. There's now an alert when that happens

Thanks,
Ryan

 

[wayner]

Great job on this - I just found out about it and started using it a few days ago. It is awesome!

A couple of (minor) issues on the trip log. On one trip to work it didn't know that my trip was over and it has my trip duration as 9h 32m. But that only happened once.

I am in Canada and use metric on my car which normally shows up correctly on EVMote. However earlier today I hit the web site on my iPad and it was showing the units as miles. But now it is back to metric.

I love the "you forgot to plug in your car" email.

 

[MikeBur]

I've never seen any issues arise from overloading the Tesla APIs in evMote. I manage the traffic pretty well. That's another way of saying that every issue with evMote so far has been 100% because I screwed something up

There is a lot of instability with the Tesla API though. It doesn't feel like it's a priority for them. Here's a dashboard I use to keep an eye on evMote:

At the top right, you can see the constant errors that come back from the Tesla API. I usually just auto-retry those requests and they work. Not ideal.
Also in the middle right, you can see errors I get back from the GPS tracking API. I usually just drop that particular GPS coordinate. You wouldn't notice because the tracks come in pretty quickly and one point isn't noticeable.
Also, note the dip in the bottom right graph. That's the graph for number of trips that were stopped. You can see that when the GPS tracker crashed, it stopped logging trips. There's now an alert when that happens

Thanks,
Ryan

thanks Ryan, though honestly wasn't implying either system was abusing the api, rather that if each app assumes the "safe" behavior of the same qty/freq of requests as phone app then running two simultaneously (or heck 3 - insert remote S here ;-)) would be "less safe" ;-)

Like the graphs, though can't determine if these are good numbers or bad numbers, presuming this is server load for unknown number of clients? Trying to correlate these by eye I do wonder if there's some weirdness on the web api vs tesla Api error ratio?

again, if it's not clear this is great! Want to not bring down anyone else's experience if I'm data logging

 

[cryptyk]

thanks Ryan, though honestly wasn't implying either system was abusing the api, rather that if each app assumes the "safe" behavior of the same qty/freq of requests as phone app then running two simultaneously (or heck 3 - insert remote S here ;-)) would be "less safe" ;-)

Like the graphs, though can't determine if these are good numbers or bad numbers, presuming this is server load for unknown number of clients? Trying to correlate these by eye I do wonder if there's some weirdness on the web api vs tesla Api error ratio?

again, if it's not clear this is great! Want to not bring down anyone else's experience if I'm data logging

I see your point. I'll definitely know if it becomes a problem

Quick graph explanation, left-to-right, top-to-bottom:
1) Number of times the web app exits, usually becuase of a deployment. Occasionally because of a crash.
2) Same thing for the GPS tracker app.
3) Errors seen because of Tesla API instability. These shoot way up sometimes in the middle of the night when they deploy new APIs, etc.
4) Crashes from the web app
5) Crashes from the GPS tracker app
6) Instability in the Tesla GPS tracker APIs. They regularly send bad data back.
7) The number of IFTTT calls ya'll are making.
8) The number of trips that were recorded stopped. It's fun to see how this changes over a 24h period.

Thanks,
Ryan

 

[MikeBur]

Great explanation, thanks Ryan. Next time in SD, I'll definitely buy you a beer at the Field? ;-)