Hi all, couldn’t find a thread that got into this, but if it has been covered elsewhere I’d appreciate a redirect.
So I built the openflexure microscope over the course of the last two months - first time working with electronics in general, but I managed to get everything assembled, the micro SD loaded image for the raspberry pi, and whatnot.
Then I found out the openflexure connect doesn’t have a file download option for Mac or Manjaro, the two options at my disposal right now.
I haven’t tried Boot camp on mac, but if it is anything like winebottler I probably will have trouble with it. Has either worked for anyone?
As for manjaro, which I am running on a Pinebook Pro, I tried “sudo snap install openflexure-connect” in the konsole but the error states that the snap install only words for amd64, and I have arm64.
Tried installing something called AppImageLauncher on manjaro, but I couldn’t figure that out either. Admittedly, I am totally new to anything console-related or manual at all.
Is there any way to operate the openflexure microscope on either of these? Thank you in advance. Guidance here will be appreciated.
In the Openflexure Connect downloads there is a version for ARM Linux, did you try that one?
@WilliamW
Thanks for the suggestion, but I don’t seem to have anything to run .appimage files on Pinebook/Manjaro. And I had trouble trying to install AppImageLauncher. I’m currently trying to get windows 10 iso onto a DVD or something for boot camp.
You can also run connect on the same Pi that is running the microscope server. That would not need anything installed on a different computer. Connect just gives an interface to the microscope commands, you could send the commands direct from a program in Python instead.
So the interface provided by Connect isn’t needed to get the image from the Pi camera?
And without Connect, I could just be send commands with the Mac’s console program? Via ethernet?
The simplest way (apart from Connect) is the Python client.
The documentation for the server API should be visible on your microscope server at http://{your microscope IP address}/api/v2/docs/swagger-ui
I’ll look into the python client if the trouble persists, thank you. I tried that URL for the microscope server, and it just gave me an AT&T page with a list of devices connected the network.
But I managed to get windows 10 running through boot camp, uploaded the Arduino nano sketch, got openflexure connect running, but now I’m having trouble connecting to the microscope.
I’m not sure how to connect the Pi to wifi, so I just have it plugged into the router (?) via ethernet, and the Pi now registers on the AT&T list of devices connected. But I still can’t connect it to Openflexure Connect. Might have something to do with the port number, which I have no idea how to find for the Pi.
Tried installing something called AppImageLauncher on manjaro, but I couldn’t figure that out either. Admittedly, I am totally new to anything console-related or manual at all.
You should be able to just run the AppImage without installing anything else. Make the .appimage file executable, and just run it directly and it should load just fine.
I’m not sure how to connect the Pi to wifi, so I just have it plugged into the router (?) via ethernet, and the Pi now registers on the AT&T list of devices connected.
OpenFlexure Connect should discover the microscope if it’s on the same network. Otherwise, you can enter the IP address and it should auto-fill the port to 5000.
Might have something to do with the port number, which I have no idea how to find for the Pi.
Yeah if you’re accessing the interface just via a browser you’ll need to enter port 5000.
I’ve now documented this better on the website: Install the Software
Thanks for all the suggestions - I’m stilling trying some possibilities you’ve laid out, but nothing has been working so far.
I fear the image may not have been uploaded to the micro SD. I’m able to get the IP address of the raspberry pi when I enter “ping raspberry.pi” into the command prompt, but when I enter “ping microscope.local” nothing comes up. Does that mean the image wasn’t uploaded properly?
And as for Openflexure Connect, which I am now running on Windows within a Mac boot camp, every possibility I try yields (“connect remotely” or “connect locally”) the “Network Error” response.
I assume that because I’m able to ping the raspberry.pi, and get its IP address, that it is connected to the local network, no? Could it be that the raspberry pi is connected, but the microscope functionality has been installed improperly?
That might also explain why I’m unable to access microscope.local via internet browser.
Thank you both again - if nothing else, I’m learning a great deal throughout all this.
We have now tested running OpenFlexure Connect on a Mac.
The instructions for installing the application on a Mac have now been written in the handbook.
Currently it is not as simple as with Windows or Linux and so at the moment I would only advise you to do so if you are comfortable using command line tools. We aim to make it easier to do so in future, so let us know if you have any feedback.
Sorry if I’ve missed the suggestion earlier, but if you’re on a platform where OpenFlexure Connect isn’t easy, you should just be able to type http://microscope.local:5000/ into a web browser (Chrome is best tested) and get the interface there - unless this has changed, @jc2450?
That does presume your platform understands mDNS for name resolution - but this should be the case on a mac. Failing that, you can use your microscope’s IP address instead.
1 Like
ah yes, didn’t think of the web interface! Thanks
Just a note, on my home network it shows up as http://microscope:5000 . Apparently this means something is mon-standard about my network, but as it is plain out-of-the-box home broadband router it is probably not unique. So if microscope.local does not work, it might be worth trying that alternative.
I can access mine as either http://microscope:5000/ or http://microscope.local:5000/. One of those two should always work (I think it’s .local) and the unqualified form is platform-dependent but it does work on Windows at least. When I was doing a lot of reading about mDNS I understood why, but I have now forgotten!
1 Like