How to interpret Galaxy Error (OSError: [Errno 16] Device or resource busy)

Question: How to interpret Galaxy Error (OSError: [Errno 16] Device or resource busy)

3.4 years ago by

United States

I am trying to add a multi sequence alignment tool (PASTA) to my local instance of galaxy. Currently I have a python script that calls the bioinformatics alignment software and it works, and I have a working wrapper for my python script. However, when on galaxy I get the following strange error for a directory in galaxy and I do not understand how to read this error or what I should do to solve it.

Error:

OSError: [Errno 16] Device or resource busy: '/home/usr/galaxy/database/tmp/pymp-17OSe8/.nfs00000000a0004b85000002d4'

When I attempt to find this file, it does not exist. I have a hunch that galaxy is attempting to remove a file that doesn't exist, but I don't understand why this directory exists, and why it is trying to access a file that is no longer there. Can someone explain to me what galaxy is trying to here, or what this error means?

Thanks!

software error galaxy • 2.8k views

ADD COMMENT • link •

modified 3.4 years ago • written 3.4 years ago by david.tran.dt2 • 0

3.4 years ago by

Bjoern Gruening ♦ 5.1k

Germany

Bjoern Gruening ♦ 5.1k wrote:

My guess is that your Galaxy distribution is located on a NFS filesystem and the Galaxy temp directory is also located on this filesystem. Every tools gets it's own temporary directory, which will be deleted after tool execution. In your case it can not be deleted because of some NFS left-over. Try to put the Galaxy temp dir on a local (as fast as possible) file system. I think this will solve your problem.

Cheers,

Bjoern

ADD COMMENT • link written 3.4 years ago by Bjoern Gruening ♦ 5.1k

3.4 years ago by

david.tran.dt2 • 0

United States

david.tran.dt2 • 0 wrote:

So I tried moving the pymp-170Se8 and that still didn't work. Did you mean move my entire 'tmp' directory? Here is the full error report:

Traceback (most recent call last):
File "/nfs/sw/.conda/envs/galaxy/lib/python2.7/multiprocessing/util.py", line 274, in _run_finalizers
    finalizer()
File "/nfs/sw/.conda/envs/galaxy/lib/python2.7/multiprocessing/util.py", line 207, in __call__
    res = self._callback(*self._args, **self._kwargs)
File "/nfs/sw/.conda/envs/galaxy/lib/python2.7/shutil.py", line 252, in rmtree
    onerror(os.remove, fullname, sys.exc_info())
File "/nfs/sw/.conda/envs/galaxy/lib/python2.7/shutil.py", line 250, in rmtree
    os.remove(fullname)
OSError: [Errno 16] Device or resource busy: '/nfs/home/usr/galaxy/database/tmp/pymp-8fxaXt/.nfs00000000f0003232000002f0'

Thanks,

David

ADD COMMENT • link written 3.4 years ago by david.tran.dt2 • 0

Have a look at your config/galaxy.ini file and change the temp directory setting. If you are using object stores you need to specify it in your object store settings.

ADD REPLY • link written 3.4 years ago by Bjoern Gruening ♦ 5.1k

Just to be clear within my galaxy.ini I should modify the following line of code:

# Temporary files are stored in this directory.
#new_file_path = database/tmp

I just tried changing that to a different directory but I still get the same error.

ADD REPLY • link written 3.4 years ago by david.tran.dt2 • 0

You changed it to a different path that is not located on a NFS mount and you get the same error with the same old directory?

ADD REPLY • link written 3.4 years ago by Bjoern Gruening ♦ 5.1k

oh I see now. For some reason, I thought simply moving the output out of the galaxy/database/tmp directory would suffice. I moved it off the NFS mount and the error was gone! Thank you for your help and patience Bjoern!

ADD REPLY • link written 3.4 years ago by david.tran.dt2 • 0

Please log in to add an answer.

Similar posts • Search »