-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
strider
Active Member
I can now confirm that this works on 2.5s. Walked into the garage this afternoon and noticed the coolant pump running. I then recalled it had been running the last 3 days straight. I found that odd as the termperature has been quite mild the last week or so. I also figured I should check the charge state. I reached in through the open window and pushed the VDS. Communication Error. Service Required. And all the readouts were dashes. WTF? Had never seen that before. I grabbed my key and turned the car on. The PRND buttons lit up and the headlights would come on but that was it. Also checked the charge port and no white ring.
After some searching here on TMC I found this thread. Unplugged and Plugged the VMS and all is well. So thank you TMC! Without this site I would be a lot poorer
After some searching here on TMC I found this thread. Unplugged and Plugged the VMS and all is well. So thank you TMC! Without this site I would be a lot poorer
rudholm
Member
After removing the USB stick from my 2.0 I got the "Communication Error. Service Required." message and all values were dashes. My heart sank when I couldn't even turn the car on. I got out of the car, took a few deep breaths, then noticed that it looked like the VMS rebooted itself or something since it went blank for a bit then came back up and my car worked again without me having to do anything. Whew.
Yeah, VMS lockups / watchdog reboots are not uncommon when reading/writing from USB sticks. It seems the software in the VMS is 'picky' about what sticks it wants to work with. I suggest trying another brand of stick.
rudholm
Member
I used a 1GB Tesla branded USB stick. Last time I did a log dump (two years ago) I used a different stick and had no issues. I need to figure out which one I used before.
S-2000 Roadster
#1244
So far, this thread has indicated that there are no special precautions - that this is "just a reboot."
However, I'm concerned that this is like pulling the power on a Linux system without requesting a shutdown first. Can anyone comment on whether there are risks to "rebooting" the VMS abruptly like this? In other words, I'm inclined to avoid a reboot unless it's the only option.
However, I'm concerned that this is like pulling the power on a Linux system without requesting a shutdown first. Can anyone comment on whether there are risks to "rebooting" the VMS abruptly like this? In other words, I'm inclined to avoid a reboot unless it's the only option.
Linux will do an FSCK check of the filesystem and automatically fix things if needed. Only bad thing of pulling the plug on an OS is that you can't guarantee your data will be written and that same data will be there after the reboot sync memory wasn't flushed and synced to disk.
It really depends on so many things internal to the way tesla designed their VMS. Even the best designed systems can fail in unexpected ways when power is cut during disk operations.
Best advise I can give is to only abruptly kill the power if you have no other choice. Try to shutdown/reboot cleanly if at all possible.
FYI, I work with embedded Linux systems testing them and looking for failures. My expertise of 20 years is with Unix embedded systems. At my current company our test units have no power button and we do as part of our daily tests abruptly pull power on a running system, so no clean shutdown and disks / memory actively reading and writing as power is pulled. Out of 4 years of doing these tests no devices had any catastrophic failures due to an unclean shutdown or pulling of power. All systems have a shell console connected to monitor their bootup sequences and console output. Upon the following boot, u-boot fires up which then loads the kernel which then verifies if the filesystem is good to mount, if it detected an unclean shutdown or possible error case, it'll run a file system check like FSCK, if an issue is found it will try to correct the issue and then remount the filesystem and boot up as it normally does.
The only time I've seen where you can't do a pull of power is during a true firmware update of a chip. Then its typically unrecoverable by the user and needs to be JTAG'ed to get it back up and running again.
Yes, if you have an option to do a clean reboot / shutdown then indeed do so. That's a proper shutdown. However we don't have that ability unless we tie in a console, gain root access to the terminal shell and issue the reboot or shutdown command locally. We don't have that access nor privilege. And nor does the service centers, they abruptly pull power just as we do. Hence why I created the thread so other's don't need to go into the service center when the VMS is hung or needs a cold boot. But by all means go to the service center if you like, they'll do the same thing as I said before
The only time I've seen where you can't do a pull of power is during a true firmware update of a chip. Then its typically unrecoverable by the user and needs to be JTAG'ed to get it back up and running again.
Yes, if you have an option to do a clean reboot / shutdown then indeed do so. That's a proper shutdown. However we don't have that ability unless we tie in a console, gain root access to the terminal shell and issue the reboot or shutdown command locally. We don't have that access nor privilege. And nor does the service centers, they abruptly pull power just as we do. Hence why I created the thread so other's don't need to go into the service center when the VMS is hung or needs a cold boot. But by all means go to the service center if you like, they'll do the same thing as I said before
Last edited:
strider
Active Member
Yeah, this is the first time this has happened to me in almost 7 years of ownership. I tried everything I could think of to "wake" the VMS but it appeared to be hard-locked. From what I could tell a power cycle was my only choice. I guess I could have let it sit for awhile to see if a watchdog or something triggered a reboot but it sure didn't feel like it was going to come back on its own.
Maybe there's a way to trigger a soft reboot via Tattler or OVMS via the diagnostic cable?
Maybe there's a way to trigger a soft reboot via Tattler or OVMS via the diagnostic cable?
strider
Active Member
Thanks. That what I thought. I opened and closed that thing like 15 times in quick succession (really wears on the tip of your finger to unlatch it that many times) but no dice.
backmost
Member
I'm trying to reset my VMS and attempting to pull the blue and black plug but finding it impossible even after lifting the little tab
Rotarypower
Member
So, I am having issues with my HVAC (no battery cooling) and will post that question separately, however, while in the service menu trouble shooting lack of cooling, I think I fat fingered an input and after exiting the service menu I received error messages ID: 1554 and 2096. Since then the car appears to have lost communication between the VMS and VDS and I can not get back into the service menu (password does not work). Charging still works and I was able to initiate and end charging cycles using OVMS (THANK YOU @markwj )!! I have pulled the 30 amp 12V fuse in the front but after reading through this thread appears I would also need to pull the battery disconnect which I do not want to do. So, my next course is to pull the 2 larger connectors from the VMS (black and blue) hoping that will reboot the unit and fix this issue. These seem pretty straight forward with regard to unlatching and latching (see video below) but any other thoughts on additional course of action? Obviously I will heed @markwj ‘s warning about the connectors, as being on your back in the footwell of the passenger area makes a simple task difficult.
Ampseal connectors mating and unmating @ 2:04
Ampseal connectors mating and unmating @ 2:04
Attachments
Rotarypower
Member
Ok well pulling the plugs is more painful than you would think. Anyway, doing that did not fix my issue. I still get both errors. I even pulled the 30 amp fuse in the front area. So I am now needing to resort to pulling the main ESS disconnect? Does anyone know specifically what else needs to be powered down? Is it the VDS? I would rather find a way to power down the specific component (VDS?) than risk breaking the main disconnect. I have never pulled it so I am sure it is stuck firmly and will be a hastle. Is there a way to easily power down the VDS? It continued to stay on even after pulling the fuse and disconnecting both mains to the VMS. Thoughts?
Rotarypower
Member
Well seems disconnecting the two large plugs from the VMS had made things worse. I now get no charging or range data from the VDS, and OVMS isn’t getting charge data or receiving or responding to messages. Well this is making my stress level go pretty high.
If the VDS is giving VMS comms error messages, but OVMS is still able to communicate with the VMS, that implies the issue is more like with your VDS, or the cabling. Note that all three (VMS, VDS, OVMS) are on the same instrumentation CAN bus.
Similar threads
