Open fonnesbeck opened 7 years ago
Is there any particular action which will reproduce this?
There's nothing obviously wrong in the console - those .js.map
files should not affect what it's doing.
It seems to occur shortly after a cell has finished running. Python is not using CPU during the hang; only the browser becomes unresponsive.
OK, this has just occurred again, and it happened without any cell being executed in the notebook. I had just started jupyter in a project and opened one notebook. As I was scrolling down, the interface locked up. Looking at the activity monitor, I noticed that the process was quickly gobbling up memory, until it hit just over 14GB. The notebook then became responsive again, and the memory consumption dropped down to just over 10GB. However, this was still without any cells being executed. Here is the current memory consumption info from this process:
http://localhost:8890, … 10.74 GB 6.42 GB 7 196 18075 fonnescj 39.9 2:20.86 1
I can confirm that it occurs every time I load the following notebook and scroll around on it:
I noticed that there is a MathJax loading message in the status bar when it happens, so perhaps it is related?
I can also confirm that this behavior does not occur on the current 5.0.0 release from conda-forge.
I can confirm that this still occurs in master quite frequently. Same "failed to load resource" messages in the JS console each time.
is there a way to kill all processes for default workspace? safari is locked up and when restarted after killing terminal keeps loading old workspace, using mac mojave
I have noticed that notebooks will lock up during use on Safari making the browser unresponsive for several minutes. The notebook becomes responsive again after waiting several minutes.
Looking at the console afterwards shows the following:
Running current master under Safari 10.1.2 and Python 3.6.