Tessie won't start

[Doug_G]

Well this is a new one on me. Went to drive the car and it was dead. The car's firmware appears to have bricked itself somehow.

ID:1535 VMS: Linux kernel version doesn't match car-wide release
ID:887 Can't start Version number mismatch.

 

[Johan]

This happened with the car just having sat there? Or was there some kind of upgrade just before?

 

[Doug_G]

Just sittng there. Roadsters can't do over-the-air updates.

 

[Johan]

Is there some hard reset procedure for the Roadster?

 

[Doug_G]

Yes, you open and close the charge port door repeatedly. Tried that, nada.

 

[Johan]

What about physically disconnecting the 12V battery for 10 minutes? Wouldn't that make the computers do a proper reboot? Or do they do that every time the car is started on the Roadster???

 

[Doug_G]

Catch-22: the right front wheel well is up against the wall of the garage, and I can't move the car.

Also it's a bit of an involved procedure just to try something that may or may not work. I changed the battery last month and it's a bit of a pain in the butt.

 

[Johan]

What do I know anyway? Just trying to practice this "reasoning from first principles" strategy that an idol of mine has perfected.

 

[Jason S]

It prolly can't communicate with the other modules. But I've no idea where the connections are.
Look for chewed wires imho.

 

[wiztecy]

This happened to a person 2 years ago, kernel / version mismatch. Most likely an RFS / filesystem corruption... I work on Unix, grew up testing server based (working on SCO Unix) and now do embedded which I also been doing quite some time in Silicon Valley. I know it pretty well since its my profession and was a hobby of mine. This stuff is known to happen and I've seen it happen, the only way to get around it is to have a fully redundant filesystem running that can recover when things go south, but then your performance goes down the drain. We looked into doing that on the Amazon Kindle when I worked there but ditched it due to the enormous overhead.

The fix which resolved the issue I recalled was to have Tesla upload the correct VMS firmware image. Roadster worked fine after the update. I'll try to find the old thread, but I don't know how lucky I'll be.

Also dump your logs, you should be able to see the system level error messages.

 

[wiztecy]

Found it:
Help! Stranded w Software Error 0x1533, 0x400

Thread #23:
"Just to update, the car went to Mt. Kisco, they've never seen this fault before. Apparently, re-flashing the firmware fixed it (as I'd suspected). I'm quite pleased by the way the folks at the Mt. Kisco center handled the issue. That said, I'll be very curious to see if it reoccurs, as they don't know why it happened."

 

[Johan]

Man, what a great online community.

 

[Doug_G]

Ah, that's why my search didn't find it - 1533 versus 1535. I wonder what the difference is?

Seems like re-flashing may restore the car to proper operation. Hopefully Tesla can send someone to do that. I'd rather not flatbed another car to Montreal!!!

 

[AEdennis]

Ah, that's why my search didn't find it - 1533 versus 1535. I wonder what the difference is?

Seems like re-flashing may restore the car to proper operation. Hopefully Tesla can send someone to do that. I'd rather not flatbed another car to Montreal!!!

Will they charge the Ranger fee to get someone out there to flash your car?

 

[wiztecy]

Ah, that's why my search didn't find it - 1533 versus 1535. I wonder what the difference is?

Panicopticon had a 1.5 Roadster, you have a 2.0. They're running different firmware images so it may throw a similar but slightly different error code.

 

[Panicopticon]

I have a 2.0 actually. My issue was a rootfs version mismatch rather than the kernel. That likely explains the different error code and number. The fix is to take it in and have them reflash the VMS.

I suspect you can get the car started and moving by disconnecting the two bulk connectors on the VMS and leaving them off. This will cause some other faults, but the VMS shouldn't actually required for the car to run. Keep in mind that if you do this the power door locks, trunk release and a few other similar things won't work.

 

[Doug_G]

Will they charge the Ranger fee to get someone out there to flash your car?

Most likely.

 

[Panicopticon]

Out of curiosity, were you charging when it happened? Do you have an OVMS, and if so did it throw a bunch of errors beforehand? You could also dump and check the logs...

 

[bonnie]

So sorry to read this thread, Doug_G . Hoping Tessie will be back on the road shortly.

 

[Doug_G]

I don't have an OVMS installed. Don't know when it actually happened, car was sitting in the garage at the time. Log just dumps a couple of errors; I think they are the same ones that appear on the screen.

I expect I'll hear back from Montreal on Monday.