Hi From your picture, it is running, click the run button should working

I did boot, but as you can see the system goes into lockout immediately. I assume it is an EasyEDA System Bug that does not make AutoRouter work properly, as it did until two days ago. Please, if possible, solve this big problem. Thank you so much 

Here is how you can see also with linux you have the same problem, we hope you solve it. 

Stop wasting your time with the autorouter and route the board yourself. You have 11 components. If you placed them nicely, so the ratsnest shows what needs to be close, that leaves you with a 10 minute job, instead of a week on forums. If you think autorouter will save you time, think again. You also have 220V input there, don't forget that if you didn't tell the autorouter the clearance and track thickness and other rules it should follow, the traces will be garbage and your project will be just something to learn mistakes from. Don't forget mounting holes too. Stay safe.

Yes, of course, doing freehand makes the tracks better, but yes, as I have other more complex PCBs to do, this functionality is convenient, just to fix it.

I'm seeing the same issue with the router saying "connection closed" after a couple of traces, while the editor says its still running.  I suspect they have rounding errors somewhere -- sometimes tweaking the design rules very slightly helps, sometimes switching from mm to mils, but it's pretty random.  And even when it does finish, the editor says "cannot start again before previous activity is stopped". The cloud router tends to give up at the same point, so there's probably plenty of information in the server logs... (and this is definitely a regression -- the router used to be a lot more robust) PS. Does the router have some debug flag or log that I could check, to see if actually says something before giving up?)

Auto router is not good enough, suggest routing manually, recommed "Routing Confilct - TrackArroud" at the right-hand panel. Sometimes, if you can't get it done, try the tips below. \- Make sure the net of PCB doesn't contain the special charaters\, such as  ; \~ \ / \[ \] =  etc\. the chrarter \- and \_ are supported\. \- Make sure the board oultine is closed\, doesn't has board oultine overlap situation\. \- Make sure there are no DRC cleance errors \(short circuit issue\)\, such as two different network pads overlapping\, or different net pads in the same location within the package\. \- Make sure no footprint outside the board outline\. \- Use local auto router rather than cloud server\. \- Make sure PCB rule doesn't have 3 decimal places\, EasyEDA only support 2 decimal places\. \- Skip the GND nets\, add copper area to GND net\. \- Use small tracks and small clearance\, but make sure the value is more than 6mil\. \- Route some key tracks manually before auto routing and ignore them when auto routing\. \- Add more layers\, 4 layers or 6 layers\. but that will make more cost\. \- Change the components layout\, make them have more space between each other\. \- Don't make any via/pad overlap the different objects which can be set the net\. \- Stop the auto\-router\, close the script\, and re\-open the script to run local auto\-router server again\. \- Sometimes\, The auto\-router won't work at the first time\, please modify the layout and try serveral times\. if it still fail, suggest routing manually, it is not good enough, we will wirte a new one, but now we need to develop some feature fisrt, there is no plan yet.

Thanks. Please note though that this is a regression -- I'm seeing this for layouts that the local router (0.8.10) used to handle just fine, but some recent change to the editor broke things, causing the router (same version that worked before) to abort without any message, and the editor getting confused (leaving the "running" box open, and causing "cannot start again" messages also after a successful run). Updating to 0.8.11 didn't change anything. I'd look for changes to the code that talks to the local router, or code that converts the editor model to the router model. My comment about cloud was that assuming you log router crashes on your side of things, odds are that you have detailed information there. In the meantime, it seems routing with the canvas set to mm is more fragile, so people who run into this might want to experiment with doing the placement with the canvas set to mils, or at least using mils when running the router and making sure all trace widths are an integer number of mils.