Keep in mind that you might not experience the issue immediately, but need to let the program run for a few seconds before you set the breakpoint. When letting the program run for a little while and then add the breakpoint to line 28, you can reproduce a similar problem, where the program stops at first, but upon resuming execution it enters a deadlock behavior and freezes. Similarly, a breakpoint in line 28 only stops once per iteration, even though there are multiple print statements happening on different workers. It stops at first but continuing does not work anymore. When setting it to 0, everything works fine, but as soon as you go to 1 or above, the breakpoint behavior gets really weird. Try to set a breakpoint in line 68 and the play with the num_workers variable. ![]() is probably excessive for debugging this, but it lists all the libraries that I have in my virtual environment. It requires a few libraries, but you should be able to install them easily. I tried to distill my code down to the bare minimum to demonstrate the issue and put it into this Github Gist: The linked issue is different from the problem that I am facing, unfortunately. I guess, I will downgrade to an old P圜harm version until this issue is resolved. Libc++abi.dylib: terminating with uncaught exception of type std::_1::system_error: Broken pipe I also get a deadlock-like situation sometimes with breakpoints, with this stacktrace when I terminate the program:įile "/Applications/P圜harm.app/Contents/plugins/python/helpers/pydev/_pydevd_bundle/pydevd_comm.py", line 1769, in do_itįile "/Applications/Xcode.app/Contents/Developer/Library/Frameworks/amework/Versions/3.7/lib/python3.7/threading.py", line 852, in startįile "/Applications/Xcode.app/Contents/Developer/Library/Frameworks/amework/Versions/3.7/lib/python3.7/threading.py", line 552, in waitįile "/Applications/Xcode.app/Contents/Developer/Library/Frameworks/amework/Versions/3.7/lib/python3.7/threading.py", line 296, in wait ![]() Breakpoints from the main thread are usually respected, but coming from a background worker they often aren't. My best guess is that it is related to multi-threading. Even the same breakpoint is sometimes ignored. I can't find any clear pattern why breakpoints are disregarded, but I know that I have massive problems developing, when my breakpoints don't break. ![]() Some breakpoints are being respected, others aren't. Same problem here with P圜harm 2020.1.2 on macOS Catalina 10.15.5 with Python 3.7 (Anaconda installation).
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |