[Bug - Mac Release on M1 and Windows] Distributable not able to run on main (9/11/2023) #354
Comments
|
I wouldn't proceed with many more dev features (including interfaces) until this is resolved since it makes the initial setup process harder for alpha testers |
|
On M1 Mac, on main, creating a clean M1 Mac environment does not even solve for me using conda or mamba. I'm not sure how that came to be (surely it worked at one point??), and why @CodyCBakerPhD could create an env but not me. In any case, if I create a clean M1 Mac environment using the |
|
I'm on the move today but I will try building on dandihub (linux) and with a non-m1 mac configuration tonight. |
|
I confirmed that I was able to get the M1 dist working on @rly's branch ( Retrying DANDI Hub with that branch/environment update now |
|
Now that I think about it, did we ever get the full app to work on the Hub? Last time we hit the wall was on the Flask not even building, so does anyone remember being able to run it there successfully? I still get the same errors, but they are related to GPU aspects (even on the GPU instance) so IDK. Manually navigating to the local host shows the front-end displaying something, but it says the backend isn't running |
However, when I use the CI to successfully deploy from the If I download the then, once I bypass the security things due to the lingering signing issue #309, the app immediately crashes upon launch - I get a brief glimpse of the splash screen, which is more than I get on Windows I also get the following crash report, if it is of any use
-------------------------------------
Translated Report (Full Report Below)
-------------------------------------
Process: NWB GUIDE [9523] Date/Time: 2023-09-12 14:58:01.3701 -0400 Sleep/Wake UUID: A247CA82-3F30-49F4-BB42-DB7434BE75C3 Time Awake Since Boot: 9100 seconds System Integrity Protection: enabled Crashed Thread: 0 CrBrowserMain Dispatch queue: com.apple.main-thread Exception Type: EXC_BAD_ACCESS (SIGBUS) Termination Reason: Namespace SIGNAL, Code 10 Bus error: 10 VM Region Info: 0x742074636c is in 0x7400a40000-0x7500000000; bytes after start: 533750636 bytes before end: 3750468755 Thread 0 Crashed:: CrBrowserMain Dispatch queue: com.apple.main-thread Thread 1:: com.apple.rosetta.exceptionserver Thread 2: Thread 3: Thread 4: Thread 5: Thread 6:: ThreadPoolServiceThread Thread 7:: ThreadPoolForegroundWorker Thread 8:: ThreadPoolBackgroundWorker Thread 9:: ThreadPoolForegroundWorker Thread 10:: Chrome_IOThread Thread 11:: MemoryInfra Thread 12: Thread 13: Thread 14: Thread 15: Thread 16: Thread 17:: NetworkConfigWatcher Thread 18:: CrShutdownDetector Thread 19:: NetworkConfigWatcher Thread 20:: ThreadPoolForegroundWorker Thread 21:: ThreadPoolForegroundWorker Thread 22:: ThreadPoolForegroundWorker Thread 23:: ThreadPoolSingleThreadForegroundBlocking0 Thread 24:: CompositorTileWorker1 Thread 25:: ThreadPoolSingleThreadSharedForeground1 Thread 26: Thread 27:: NetworkConfigWatcher Thread 28:: CacheThread_BlockFile Thread 29: Thread 30: Thread 31: Thread 32: Thread 33: Thread 34: Thread 35: Thread 36: Thread 37:: com.apple.NSEventThread Thread 38:: ThreadPoolBackgroundWorker Thread 39:: ThreadPoolSingleThreadSharedBackgroundBlocking2 Thread 0 crashed with X86 Thread State (64-bit): Binary Images: External Modification Summary: VM Region Summary: REGION TYPE SIZE COUNT (non-coalesced) Full Report{"app_name":"NWB GUIDE","timestamp":"2023-09-12 14:58:42.00 -0400","app_version":"0.0.4","slice_uuid":"4c4c44a6-5555-3144-a12e-12d3e30796b2","build_version":"0.0.4","platform":1,"bundleID":"com.catalystneuro.nwbguide","share_with_app_devs":1,"is_first_party":0,"bug_type":"309","os_version":"macOS 13.3.1 (22E261)","roots_installed":0,"name":"NWB GUIDE","incident_id":"C2EC6554-AA87-43D2-BC7C-8CD3784F8B92"} ], Model: MacBookPro18,2, BootROM 8422.100.650, proc 10:8:2 processors, 64 GB, SMC This goes back to the question of if we will need manual deployments for M1 specific distributions. @rly When you get a chance, and while you're solving the #309 problem, could you try manually signing and deploying a M1 release from your machine using those debugged dev branches? |
CodyCBakerPhD
changed the title
|
After a fresh wipe of everything on Windows, the recursion depth issue is not present, at least when running on Electron issue ticket electron/electron#21503 is exactly what I see; tried all solutions i) cleared various AppData folders None of them have worked on Windows so far. Still fails to even display the splash screen; processes remain running in the background but nothing gets rendered/launched |
Original comment chain starting at #337 (comment)
Just reproduced it on the DANDI Hub, the one shared computing environment we all have access to; I suggest we start there and see if fixing that fixes the other platforms
I also wasn't able to run dev mode either on the Hub - I'm pretty sure both used to work, up to those previous backend distribution issues, right?
To be clear, workflow is
nwb-guide2a)
npm startto ensure dev mode is working2b) might need to do
npm ciifnpm startdoesn't work due to 'electron-vite' not being foundnpm run build:<platform>4a) Windows: run setup
.exe, follow instructions, then try to run application from the installed location4b) Mac:
.dmgto App folder, try to run from there4c) Linux: Configure
.AppImageto run as executable, then try to run it (on the Hub, do so from the remote desktop)Might be related to issues seen by @alessandratrapani on ubuntu?
The text was updated successfully, but these errors were encountered: