mirror of
https://github.com/jetkvm/kvm.git
synced 2026-03-02 22:58:00 -05:00
Powering JetKVM through ATX or DC Power Extension does not work #216
Labels
No labels
component/keyboard-layout
component: cloud
component: device screen
component: extensions
component: hid/keyboard
component: hid/mouse
component: network
component: timesync
component: ui
component: updater
component: usb
component: usb/hid
component: usb/storage
component: video
component: webrtc
component: webserver
need-more-details
status: working-in-progress
wontfix
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
starred/kvm#216
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @equinox0815 on GitHub (Apr 27, 2025).
Originally assigned to: @adamshiervani on GitHub.
I got my ATX and DC Power Extension about 4 Weeks ago and up until now was not able to power the JetKVM trough the extensions. I reported this to contact@jetkvm.com 3 weeks ago but besides an automatic response that informed me that responses will take 4-5 business days it has been crickets ever since. Now 3 weeks are quite a bit longer than 4-5 business days and yes i know this is Kickstarter thing and not a long established company but i think this is not OK. I'm posting this here in the hopes that somebody else has or had the same issue and found a way to work around it. For me this issue and #37 make the device completely unusable. Below is the problem report that i sent 3 weeks ago:
I got my ATX Power as well as a DC Power extension this week and just did some testing with it. I must say that i had quite a lot of troubles getting them to "work".
The worst problem I'm having is that powering the JetKVM with either of the boards does not work. I tried a couple of different scenarios, with different power supplies and also tried swapping around the RJ-11 cables with different ones. None of which worked.
I then used a multi-meter to measure the voltage between Pin 1 and Pin 6 of the RJ-11 extension port of both extensions. The DC Power extension (when plugged into a 19V power supply) was only delivering 4.65 V. When i connect a USB-C power supply to the DC extension the Voltage stays about the same. I tested this with 3 different USB-C Power supplies - all of which do work with the JetKVM when plugged into it directly.
The ATX Extension board - when connected to the same USB-C power supplies - delivers 4.96 V on the RJ-11 pins - this looked more promising. Alas the JetKVM still won't power on. When i unplug/plug the RJ-11 connector on the JetKVM while the ATX Extension is powered over USB-C i can see a voltage drop from 4.96V to 4.92V so the JetKVM seems to at least try to draw current form the extension but apparently not enough for it to boot. I couldn't do the same test with the DC extension because i would need to open the case in order to have access to the RJ-11 pins while the cable is plugged in. But since the DC extension already has a low voltage without a JetKVM plugged in i don't think this would yield any new information anyway.
The powering-issue is already a deal-breaker for my use-case but i did some further testing with the DC-Power extension. With the JetKVM powered via USB-C, the DC extension connected to a 19V power supply and the DC output of the extension connected to the machine i want to control (Minisforum MS-01) i was able to load the extension in the Web interface and get voltage and current readings. Right after the extension is loaded i get a voltage reading of 1.8V and current reading of 0.0A. This is of course incorrect but after hitting the "Power On" button the voltage is finally shown correctly also the current and power readings seam plausible. So at least communication with the DC Power extension seam to work. Although it takes quite some time for the power of the controlled computer turns on - also i have to reload the page to get the proper readings. Not sure if this just the way it is at the moment or if this points to a communication problem as well. The log file at
/userdata/jetkvm/last.logdidn't show any errors or warnings during my tests but i also don't get any log messages if the RJ-11 cable is not connected to any extension. Only when unloading a module i get some log lines about the serial connection being closed. So the logs are not really anything to go by apparently.All the tests have been carried out using the latest firmware - as of this writing this is:
App: 0.3.8
System: 0.2.3
I will be very thankful for any help or hint anybody can give me.
@cmsj commented on GitHub (Apr 28, 2025):
I'm seeing something very similar: https://discord.com/channels/1270105230989594634/1366426264230957077
@equinox0815 commented on GitHub (Apr 28, 2025):
Sorry i have no discord account and to be honest i am not very keen on creating an account just to get basic customer support from a company.
@cmsj commented on GitHub (Apr 28, 2025):
FYI:
I'll update here if anything relevant happens on Discord.
@dcrdev commented on GitHub (May 3, 2025):
Received my ATX board today and am also seeing this
@rrman commented on GitHub (May 7, 2025):
Seems that I have the same issue as the OP. I have the DC Extension. I also have a Minisforum MS-01. I connected the JetKVM, DC Extension and my Minisforum MS-01. My JetKVM doesn't power on when relying on the power from the DC Extension. I have not yet configured the JetKVM. I need to go out and buy a USB-C power brick...
@equinox0815 commented on GitHub (May 9, 2025):
I have a major update...
Since i still haven't heard anything from support i decided to treat this thing as a total write-off. It is a really nice product that i would have loved to like but at the moment i don't have any trust in the product as well as the company.
The good news is: since i don't think i will ever send the product back in the hopes to get a working version i can now do whatever i want with it. I started by opening the DC power extension to inspect the board. I also did some reverse engineering to try to figure out which component may be responsible for the failure. Unfortunately i was not able to find any obvious culprits. So i decided to close the device up again. While trying to screw in the little metal sheet, on the side where the output DC jack as well as the RJ-11 connector are, i ran into the issue that the cut-outs for the connectors didn't want to line up. The reason for that was that the DC Jack is soldered on crooked and it needs quite a bit of force to make the case close-up. This got me thinking... what if the connectors are deformed to an extend where there is poor contact with the RJ-11 cable. So i left the case open, connected all the cables and the JetKVM booted up immediately!
I was able to log into the web interface and tried to turn of the controlled computer using the "Power On" button. The first click did nothing but the second click worked. My MS-01 was starting, however i didn't get any voltage, current or power readings.
So i decided to install the latest firmware. Here i ran into the next problem. At first it looked promising - after downloading the newest system and app image the device would reboot and i was redirected to the login mask. I logged back into the web interface where i was greeted by a popup informing me that the device is currently updating itself and will reboot soon. So i waited a little and sure enough the device was rebooting and i again was redirected to the login prompt. At this point i didn't think much of it but after logging back in, the firmware update popup came back - again. After about 10 reboots i unplugged the device from all cables and plugged it back in. Alas no change.. the device kept rebooting. Please keep in mind that at this point the device was getting power from the DC extension as well as the USB input since my MS-01 was powered on during all of this. So i don't think it had something to do with low power preventing the device from flashing the new firmware. After some time i was finally fast enough to open up the settings page before the update prompt could prevent me from doing so and i was able to uncheck the "Auto Update" check-box. This seemed to help and and the device stopped the relentless reboot cycle. To my surprise the version number shown in the general settings tab told me that device is now running a newer version. I ran the manual update check and sure enough:
Your system is running the latest version. No updates are currently available.Don't know if this is true or if this just another bug but i will take it.At this point i remembered why i wanted to update the firmware in the first place. I clicked to open the extension tab to see if i know have power readings but sadly this was not the case. The "Power On" button was active and "Power Off" button was grayed out but the MS-01 was definitely getting power. So apparently the communication with the extension seemed to be broken. I tried fiddling with the connectors to get it working. I also tried to reload the web-page, reset the device config and to re-install the extension but no change... so i decided a little break would be a good idea. About 10 minutes later i sat back down in front of my computer and suddenly voltage, current and power readings worked. Don't know what changed but at this point everything just worked. The DC control functionality is still quite shaky (i often have to click the button twice ore even three times) but basically everything seems to work now. I was even able to close the case of the extension by screwing the metal plate back on and it still worked. If only #361 would finally land in a released version of the firmware - all my problems would be fixed.
As for the ATX extension i also did give this another try and sure enough powering the JetKVM through this extension while it is plugged into a USB-C power supply works. This puts a dent into my it-was-the-crooked-dc-jack-in-the-dc-extension-theory - so it probably was a problem of the JetKVM device itself all along. A problem that has just magically gone away... somehow.
But of course i have to talk about the elephant in the room. I cannot trust this device what so ever. Every device that suddenly magically starts working will inevitably fail the same way. This is not acceptable - at least not in my current use-case. If i cannot trust the device to work as expected in a moment where i need it the most it is useless to me. So i will look for an alternative and most likely use this device for a use-case where i don't care if it breaks. The ATX extension has other problems as well - most notably this where apparently nobody checked if the ATX extension could actually be physically installed into a computer using the low profile bracket...
So for me that's it - i hope some of you can find a more satisfactory solution to the problem. and sorry for the really long text. It was my intention to document my experience with the product and the company. And maybe some of the Youtubers that seem to like the product very much (looking at you @geerlingguy) stop featuring the product as much without talking about the obvious problems the company and product has. Sorry for calling you out like this - this not really your fault - i'm just salty.
@rrman commented on GitHub (May 10, 2025):
My DC Extension works (i.e. it can power on and off my Minisforum through the barrel connection). The caveat is that the JetKVM has to be powered via USB-C.
I was almost ready to be tolerant of this limitation, but using the barrel extension has some quirks. There is no “memory” of the power state of the DC extension. For example, if the JetKVM’s USB-C is unplugged, power will also be taken away from the computer. I haven’t tested it much more, since I got tired of my computer ungracefully powering off, but I got the impression that if the JetKVM has to reboot (maybe as part of firmware update), it might take power away from the computer. The power state of the DC Extension must be saved somehow so whatever happens to the JetKVM, the power state (powered on or off) doesn’t change unless the user explicitly turns it off or on via the GUI.
I’m pleased with the JetKVM and I’m using it, after I accepted the fact that I’m not going to use the DC Extension, at least not until its problems can be addressed.
@geerlingguy commented on GitHub (May 11, 2025):
I haven't tested my DC extension yet, I've been testing with the WisdPi PoE adapter lately. Love that solution, want it built into the JetKVM though :)
@equinox0815 commented on GitHub (May 12, 2025):
Hi @geerlingguy, my main issue wasn't really that my extensions don't work properly. My issue is about how the company is dealing or rather is not dealing with the problem. I reported this issue to the company support address almost 6 weeks ago but haven't heard anything besides an automatic response - so i at least know they got my mail.
And yes of course backing anything on Kickstarter is always risky, but to be honest i was quite hopeful it would work out better. In no small part because your review, and also the reviews of a lot of other Youtubers, were very positive. All reviews that i have seen about the product emphasized how polished the device already is and how close to a finalized product this is. The way i see it, at least to some extend, the product is nothing but polish without real substance. Sure the hardware looks nice and and screen is very sexy. Also the UI looks sleek but all of this hides the fact that the software is missing a lot of features that i would call essential. For example none of the reviews mentioned that things like #37 are missing.
Again i don't think you are the bad guy here. I understand that you reviewed the device 6 months ago where it was a lot less clear where this project will go. As far as i can remember you even had a big disclaimer about it being a Kickstarter thing which is always risky. I don't even think your review was bad. I believe you presented the state of the product quite truthfully. The thing that is bugging me is now that this Github repo is filling up with bug reports - a lot of which don't get any response from somebody that works at the company. Now that a lot of the more problematic parts of the device shine through the cracks of the crazy amount of polish this product has received. Despite all this, the product is still featured a lot on Tech-Youtube - overwhelmingly in a positive light. It is this that i would like to see changed. And no i don't want you to stir up drama on social media by making a video that completely dunks on the company and how bad they are. I think just one sentence, that shines a little light on the problems this device still has, whenever you feature their product in one of your videos would go a long way. My hope is that the people behind the company wake up and realize that they kind of dropping the ball here.
I mean some of the issues here on Github feel like a group therapy session were people are looking for help but all they get is a little bit of peace in the knowledge that they are not the only one who has the problem. I think we should not forget that while this is an open source project, which is great, this is quite specialized piece of software that is only really useful in combination with hardware the company is selling. So i expect a little bit more engagement from people working at the company in the issue tracker.
So here we are - my text has been a little bit wordy again (sorry for that) but as i already said for me this is at least in part therapeutic. Also sorry that i solely lean on you Jeff. I'm not really interested in most of social media - i watch my fair share of videos on Tech-Youtube and i have a Github account but that's basically it. You are the only one i know that features the device in his videos who is also active on Github.
@simoncullenIE commented on GitHub (May 14, 2025):
If it helps, I tested the DC extension today on 2 different mini pcs.
Both worked as expected, the Jetkvm is being powered by the RJ11 connector and turns on before I connect the usb/hdmi,rj45.
The "Power On" and "Power Off" also work as expected, although sometimes the displayed Voltage values do not update and require a browser refresh.
#1
Minipc: Chatreey T9 N100
Power Supply: 12V 3A
#2
Minipc: Aoostar R1 NAS N100
Power Supply: 19V 4.74A
@tylamb1 commented on GitHub (May 22, 2025):
I successfully ran mine with only the board plugged into the JetKVM and power applied to the board with usb c.
@pauliusdotpro commented on GitHub (Jun 2, 2025):
I also ran into this issue. I cant power jetKVM through rj11 connected to atx extension board.
Edit: problem is that its not rj11 port, but rj12. rj12 cable works.
@lughino commented on GitHub (Jun 14, 2025):
Same issue here.
I have a Minisforum MS-A2 and the DC extension simply doesn't work. It seems it doesn't have enough power to start the machine. The JetKVM is turned on as soon as I put the DC barrel on the extension, proving that it's alimented properly from the DC extension.
But the DC extension is incapable of powering my mini PC.
Output 19V 12.63A
@equinox0815 commented on GitHub (Jun 14, 2025):
@lughino this is actually different than i was experiencing. In my case it was the JetKVM itself that was not powering up. The MS-01 always powered up through the DC extension. At least once i managed to turn it on through the JetKVMs Web-UI. Probably the communication between the JetKVM and your extension does not work properly and the DC extension does not receive the command to turn on your MS-A2?
@cmsj commented on GitHub (Sep 18, 2025):
I was just rummaging around in a drawer and came across the DC extension again, and tried a couple of experiments.
With a random 12V brick, the JetKVM is powered correctly through the RJ12 cable. Switching to the 19V supply from my Minisforum NAB6 though, still shows absolutely no power going to the JetKVM over RJ12.
I'm starting to suspect that the DC extension is not actually really able to work with 19V input, but it certainly doesn't help that it doesn't have the logo indicating barrel polarity.
JetKVM should really say something about this.
@onedr0p commented on GitHub (Nov 1, 2025):
I can confirm with my 3x ASUS NUC14 that the DC extension does not work, it is a 5.5x2.5mm 19V, 6.32A PSU. The NUC and JetKVM do not power on. All Jetkvms and DC extensions are on the latest firmware as of this post.
Unfortunately I was mislead buying the dc extension thinking that because it was in the 12-20V range it would work, now I have 3 dc extension devices I can't use and wasted nearly $90.
Will there ever be an official update from the Jetkvm team or are people going to continue buying this device on a chance it might not work?
@onedr0p commented on GitHub (Nov 5, 2025):
I have opened a PR to the docs with a reference to this issue, it would have been nice to see this before I bought three DC extensions. I probably would have only bought one to test instead of the three to ensure it worked.
https://github.com/jetkvm/website/pull/36
@xunholy commented on GitHub (Nov 5, 2025):
So I initially had the same issues, then I saw a post somewhere else that had some more information so I'll explain how I got it to work.
Once you've plugged in the DC adapter, you need to go back into the JetKVM interface, and do the following:
The device will now power on, if you don't the above and try and turn the power on with the physical device it won't work until you've done the above steps 👆
@equinox0815 commented on GitHub (Nov 8, 2025):
@xunholy you are describing a different thing. This issue describes problems powering the JetKVM device itself through the extensions. Those problems can of course not be fixed by changing anything in the JetKVM interface since the interface is of course not available if the JetKVM doesn't even power on.
@onedr0p commented on GitHub (Nov 14, 2025):
For me I needed to power on the jetkvm via an alternative method with the DC extension wired up to my computer while my computer was off. I then followed what @xunholy did. It seems like the DC extension should come from the factory in the "on" state to avoid this issue, if possible or documentation written to explain this to users.
@slagiewka commented on GitHub (Jan 2, 2026):
I'm not sure I have the same symptoms, but here's my experience:
ATX extension:
RJ-12 to JetKVM
USB-C power supply connected
JetKVM
RJ-12 in
RJ-45 in
HDMI in
USB-C directly connected to the server's USB-A OR with splitter connected to the server's USB without any power connected to the split.
App: 0.5.1
System: 0.2.5
In this scenario JetKVM gets power via RJ-12 when the server is off. However, the USB connection between JetKVM and the server doesn't work.
Power/reset functions work correctly with the server off and JetKVM on.
USB works when I disconnect USB-C power from the ATX board (but of course I then lose the ability to run JetKVM separately from the server).
@tgrushka commented on GitHub (Feb 19, 2026):
TL;DR: I'm reverting to my original idea of using a Shelly Plug for this purpose.
Oh my. I just received my DC power extension. Was so excited to have "fully integrated" management of my NUC server. Then, found this issue trying to figure out how to even set it up. It was nice that it came with the needed cables (DC and RJ-12). But...
In conclusion, I'm returning this DC Power extension, and opting instead for a Shelly Plug, which was my original plan. Yes, it's a separate UI, a separate IP/hostname on my interface (now I'll have
kvm.localandnuc-power.localor similar), but can be automated via script if I ever need to.On a small note, the self-signed certificate that the JetKVM generates is totally invalid )expires in the year 4096). iOS Safari won't accept it at all. Therefore, one must generate their own self-signed (properly formatted) cert and install it via Apple's monstrous nasty
.mobileconfig, which is another issue for another day that I intend to open here. But I can't really use it in iOS Safari anyway, after all that trouble.