Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Job is in D state for a long time #50

Open
kaltinel opened this issue Jul 29, 2021 · 10 comments
Open

Job is in D state for a long time #50

kaltinel opened this issue Jul 29, 2021 · 10 comments
Assignees
Labels
bug Something isn't working

Comments

@kaltinel
Copy link

kaltinel commented Jul 29, 2021

Hi,
I am trying to plot an individual fast5 file, and therefore installed the requirements to run:
python SquigglePlot.py -i ~/data/test.fast5
(version python = 3.7)

However, after the script said 'fast5 file is being looked at', the process has stopped in D state (uninterrupted sleep).
I do not know what is expected me to do in this state, as D state is generally associated with I/O.

Can I have some help regarding this please?
Thank you.

@Psy-Fer Psy-Fer self-assigned this Jul 29, 2021
@Psy-Fer
Copy link
Owner

Psy-Fer commented Jul 30, 2021

Hello, that is strange.

Would you be willing to share 1 fast5 file with me to test it?
I have not seen this bug before, so will need to do some troubleshooting.

Thanks
James

@Psy-Fer Psy-Fer added the bug Something isn't working label Jul 30, 2021
@kaltinel
Copy link
Author

Thank you for your reply.
Sure, I was testing on the example fast5 files of ont_fast5_api : https://github.com/nanoporetech/ont_fast5_api/blob/master/test/data/single_reads/fe85b517-62ee-4a33-8767-41cab5d5ab39.fast5

It reads the file and gets stuck on the D state.
ppp

I look forward to hearing back from your side. Thanks

@Psy-Fer
Copy link
Owner

Psy-Fer commented Aug 2, 2021

Hello,

Could you do a git pull and then try again, but with the --single flag added?

SquiggleKit was originally designed around single fast5 files, and when multi files came, it created some complexity, because the fast5 scheme was confusing and had no 100% way of detecting the differences, so took a while for the community to build all the checks. My solution was to let the user define it, so multi is default, and single files need the --single flag.

Saying that, there was a bug on a check that was throwing a different error (not this bug), and I fixed that and tested it with the fast5 above.

Let me know if this fixes it for you.

James

@kaltinel
Copy link
Author

kaltinel commented Aug 2, 2021

Thank you very much for your detailed answer, James.

I tried the argument you suggested after a git pull:

python SquigglePlot.py --single -i fe85b517-62ee-4a33-8767-41cab5d5ab39.fast5

And the job still is in D state..
I am using Python 3.7.4, could it be the issue, somehow?

And, how long does it take for you to plot this fast5 file? Is your job always in the R state?

I appreciate your feedback, thanks

@Psy-Fer
Copy link
Owner

Psy-Fer commented Aug 2, 2021

Hmm,

Try it with python2 if you can (it was originally built for that and should work). I'll see if I can troubleshoot why it's failing with python 3.

James

@kaltinel
Copy link
Author

kaltinel commented Aug 2, 2021

I tried it with py2.7 now, and it directly goes into the D state as well..

PS: I do not get the terminal output of fast5 file is being looked at anymore.

I do not know what it could be..

Thank you for the help!

@Psy-Fer
Copy link
Owner

Psy-Fer commented Aug 2, 2021

Okay now that's interesting.

What happens if you just run it with -h as the only argument?

You should be getting errors on stderr if something is going wrong

@kaltinel
Copy link
Author

kaltinel commented Aug 2, 2021

I am getting a pretty healthy help section actually (in py2.7):

>> python SquigglePlot.py --help
usage: SquigglePlot.py [-h] [-p F5_PATH | -s SIGNAL | -i IND [IND ...]]
                       [-r READID] [--single] [--head] [--raw_signal] [-n NUM]
                       [--lim_hi LIM_HI] [--lim_low LIM_LOW]
                       [--plot_colour PLOT_COLOUR] [--save SAVE]
                       [--save_path SAVE_PATH] [--no_show] [--dpi DPI]

SquigglePlot - plotting the raw signal data after (optional) conversion to pA

optional arguments:
  -h, --help            show this help message and exit
  -p F5_PATH, --f5_path F5_PATH
                        Fast5 top dir
  -s SIGNAL, --signal SIGNAL
                        Extracted signal file from SquigglePull. Currently not
                        compatible with conversion
  -i IND [IND ...], --ind IND [IND ...]
                        Individual fast5 file/s
  -r READID, --readID READID
                        Individual readID to extract from a multifast5 file
  --single              single fast5 files.
  --head                Header present in signal or flat file
  --raw_signal          Plot raw signal instead of converting to pA
  -n NUM, --Num NUM     Section of signal to look at - -n 2000 or -n 100,1500
  --lim_hi LIM_HI       Upper limit for signal outliers
  --lim_low LIM_LOW     Lower limit for signal outliers
  --plot_colour PLOT_COLOUR
                        Colour of signal plot, takes any pyplot entry:
                        k,r,b,g,red,blue,etc...
  --save SAVE           Save file readname_saveArg.pdf --save saveArg.pdf, use
                        png, etc for other file types
  --save_path SAVE_PATH
                        Save filepath
  --no_show             Do not show plot (used for saving many)
  --dpi DPI             Change DPI for publication figs, eg: --dpi 300

@Psy-Fer
Copy link
Owner

Psy-Fer commented Aug 2, 2021

Okay, could you try the -i and --single commands like before, but with --save test.png --save_path ./ --no_show

This will check if the back end of matplotlib is working, without the interactive plots.

Let me know if it produces anything.

@kaltinel
Copy link
Author

kaltinel commented Aug 2, 2021

Unfortunately, it directly goes into D state again..

nff

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

2 participants