Pluto ignores stop signals #228
Labels
backend
Concerning the julia server and runtime
bug
Something isn't working
good first issue
Good for newcomers
help welcome
If you are experienced in this topic - let us know!
Comments
|
I've seen this too one time - it happens when the worker process gets out of sync somehow. It would be really useful if you could find a MWE to reproduce this error! If you can't reproduce it, but you find it again, please report it again! And it would help if you:
|
|
I will tell you if I found that again. It happened to me after the erroneous definition of one variable two times in different cells. Then I see this when trying to stop a series of Plots from finishing (they were probably precompiling) as they were possibly corrupted by the error that I have been attempting to fix while making another mistake (choosing an incorrect variable for the plot). I'm pretty sure that I clicked the stop button multiple times. |
fonsp
added
bug
|
It would be really helpful if people could post notebooks that don't behave well with the stopping function! |
|
Hello @fonsp, I am loving the reactive Pluto notebook environment and I would like to adopt it into my professional academic workflow! Here is an example how to reproduce the Warning: both Julia and NumPy are linked with MKL, which may cause conflicts and crashes (#433).
└ @ PyCall ~/.julia/packages/PyCall/BcTLp/src/numpy.jl:73
From worker 2:
From worker 2: signal (11): Segmentation fault
From worker 2: in expression starting at none:1
From worker 2: LSteps_Any_Loop4gas_1 at ~/.julia/artifacts/ebc509c4504f29e9138fc092addde563485a036d/lib/libmkl_mc3.so (unknown line)
From worker 2: dcopy_ at ~/.julia/artifacts/ebc509c4504f29e9138fc092addde563485a036d/lib/libmkl_intel_ilp64.so (unknown line)
From worker 2: unknown function (ip: 0x9fffffffd)
From worker 2: Allocations: 97692833 (Pool: 97653787; Big: 39046); GC: 110
Worker 2 terminated.Distributed
.ProcessExitedException(2)Distributed.ProcessExitedException(2)
[...]
[ Info: Tried to stop idle workspace - ignoring.
[ Info: Tried to stop idle workspace - ignoring.
[ Info: Tried to stop idle workspace - ignoring.
[ Info: Tried to stop idle workspace - ignoring.I am not sure if this might be helpful to isolate any |
|
Running into the same problem very frequently. In my case, I was just getting started and installing packages: begin
import Pkg
Pkg.activate(mktempdir())
packages = ["WGLMakie", "PlutoUI", "Plotly", "AbstractPlotting"]
Pkg.add(packages)
using Plots
using AbstractPlotting, Plotly, WGLMakie
using PlutoUI
endThe cell has been running for 10+ mins now, I'm restarting Pluto. I'm including error messages from previous (multiple) runs of the same notebook: |
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
|
Update on this issue (merged to original #848): Even then, the file picker works in Chrome but not in Firefox. The solution for now is to keep using Chrome. When it comes to processes (Pluto?) ignoring stop signals, it keeps happening but mostly with long-running ram-intensive tasks like loading a 500MB csv into a dataframe, or query the same data from a SQLite DB. For example, I'd take a CSV dataset (~600mb), load it into a SQLite DB file, and then have a forever-running cell when trying to query the DB and use a dataframe as a sink. The cell would run for several minutes with no output, and execution doesn't stop when requested. The DB file is 2.3GB, almost x4 the original data, and the raw data can be loaded into a dataframe in <1min from the CSV. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi!
First, thanks for Pluto! :)
I've just run into this problem using Julia 1.4.2 and Pluto v0.11.0 in Ubuntu and Chrome.
Thanks
The text was updated successfully, but these errors were encountered: