Closed icedevil2001 closed 5 years ago
Hi @icedevil2001 -- are you converting a particularly large number of files? Did you see the same error when you ran multi_to_single_fast5
a second time?
Hi @fbrennen, there is probably only 10 multi Fast5 files batched into 8000 reads each - I've tried this on my local machine which appears to work correctly however on the main server we still get the above error
Great, thanks. Did you see the issue when running the same command a second time? I'm looking at that unable to close file
bit and wondering if it was a filesystem issue that might have gone away later.
I ran it several times with different output directories and had the same issue. However, downgrading h5py from 2.9 to 2.7 resolved my issue. Thank you
Ooo, really! Thanks for that -- we'll go see if h5py version makes a difference.
To follow up, I tried using h5py 2.9.0 and it worked fine, so I'm still at a loss to explain why this happened. :(
Hello We are unable to convert multi fast5s to single fast5 and vice versa. Are there any other checks I could do to debug this issue?
Many thanks!
Error
ERROR:ont_fast5_api.conversion_tools.multi_to_single_fast5:Can't decrement id ref count (file write failed: time = Tue Feb 26 15:04:09 2019 ,filename = 'single2multi_f5_test/1/0033ec25-2ecc-4e76-b040-86fbad852d78.fast5', file descriptor = 6, errno = 5, error message = 'Input/output error', buf = 0x27daf18, total write size = 3040, bytes this sub-write = 3040, bytes actually written = 18446744073709551615, offset = 6336) Failed to copy read '0033ec25-2ecc-4e76-b040-86fbad852d78' from data/FAST5/multi/subset/FAKxxxx_2cc046c36949ccb5fde1bb37b6ed9f35f5d3116a_95.fast5 ERROR:ont_fast5_api.conversion_tools.multi_to_single_fast5:Can't decrement id ref count (unable to close file, errno = 5, error message = 'Input/output error')
Conda env: