Workaround: Can't stop GUI without stopping full node, harvester, etc

This is a problem I’ve had: I like the GUI, so I start it after starting up my full node. I would like to shut down the GUI after I’ve looked over some stats, partially because the GUI eats a significant amount of CPU time even when the GUI window is minimized. If I try to shut down the GUI, it will shut down all the other services, and there does not appear to be any developer intention to fix it, as this has been reported and closed.

I found a simple workaround on Linux - you can just send a STOP signal to the single “renderer” process when you want it to stop updating, then send a CONT signal to that process when you want the GUI to update again. For example, on Ubuntu I can find the renderer process like this:

chiacoin@bf1:~$ ps ux | grep renderer
chiacoin   11162  0.0  0.1 3890091572 245232 ?   Tl   Jan07   0:52 /opt/chia/chia-blockchain --type=renderer --enable-crash-reporter=88f24a2a-95df-4fe4-89c2-8366568a01f0,no_channel --user-data-dir=/home/chiacoin/.chia/mainnet/gui/Chia Blockchain --app-path=/opt/chia/resources/app.asar --no-sandbox --no-zygote --first-renderer-process --disable-gpu-compositing --lang=en-US --num-raster-threads=4 --enable-main-frame-before-activation --renderer-client-id=4 --time-ticks-at-unix-epoch=-1704684692626225 --launch-time-ticks=1144812129 --shared-files=v8_context_snapshot_data:100 --field-trial-handle=0,i,17974037567885796888,10811020955294331078,262144 --disable-features=SpareRendererForSitePerProcess

Then to stop the process:
chiacoin@bf1:~$ kill -STOP 11162

And to start it:
chiacoin@bf1:~$ kill -CONT 11162

Hope this helps others who don’t want the GUI burning CPU time!

1 Like

It’s possible to do similar in Windows, but I can’t say I’m aware of it using excessive resources.


Depends on your view of “excessive” I suppose. In my case, the GUI uses about 40% of a Threadripper core CPU continuously. In my experience, it’s unusual for a server application to have a GUI running continuously, burning power like this. Given the lack of a way to fix it, and since others have reported it - this workaround is a decent way to manage it away!

1 Like

I seemed to have touched a nerve there, no idea why, I simply stated I hadn’t seen it.

At various times I tried other approaches (jumping over to CLI wallet… FlexFarmer when it was a thing…). Thanks for the tip, good to know in case I jump back to a SBC/Celeron farming rig again at some point (my last board was pretty wheezy and would hose on full GUI sometimes).

1 Like

Don’t worry he replied to me earlier in a thread when I hadn’t spoken to him and wanted to inform me on how much money he spent on systems to farm Chia. Only problem is I don’t fucking care.
And 40% of one core on a threadripper with 1292 cores! oh the horror ! Guess he can’t be bothered to use the command line .


I have a TIP for GUI:
Each time when you leave your computer just use Logout button in right and then minimize GUI and lock your computer.


Benefit: all services are running in background with less usage of HW resources :wink:

1 Like

I’m away from the GUI for now (driving), but doesn’t that stop the services? If not, then I missed it as did the one who reported the issue to the devs…

====== EDIT
I see you’ve said it keeps the svcs live - I’ll have to try this put when I get to a VNC session - thanks for pointing it out - DUH!

I verified this - it parks the GUI more or less and consumes about 1% CPU which is fine!
====== /EDIT