Errors with SRA-toolkit: not enough space on the disk. (112)

Question

[JJDollar]

SOLVED (09.08.19): I was trying to save to hard drives with Fat32 file system, which has wont save files >4Gb. Reformatted the hard drive, and the files downloaded. Thank you for everyone' help!

I keep having issues when downloading datasets with SRA. This dataset I tried to download below is about 16Gb and it seems to fail after 3-4Gb downloads. Error says There is not enough space on the disk.(112), but I have it downloading on a 1Tb hard drive. Any suggestions?

C:\Users\USER\Downloads\sratoolkit.2.9.6-1-win64\bin>prefetch.exe -O "F:\ProjectSRAFiles" -X 100G -v SRR#

2019-09-05T19:10:10 prefetch.2.9.3: KClientHttpOpen - connected from 'IP ADDRESS' to www.ncbi.nlm.nih.gov (IP ADDRESS)
2019-09-05T19:10:11 prefetch.2.9.3: KClientHttpOpen - verifying CA cert
2019-09-05T19:10:11 prefetch.2.9.3: KClientHttpOpen - connected from 'IP ADDRESS' to sra-download.ncbi.nlm.nih.gov (IP ADDRESS)
2019-09-05T19:10:11 prefetch.2.9.3: KClientHttpOpen - verifying CA cert
2019-09-05T19:10:11 prefetch.2.9.3: 1) Downloading 'SRR#'...
2019-09-05T19:10:14 prefetch.2.9.3:  Downloading via https...
2019-09-05T19:10:14 prefetch.2.9.3: https://sra-download.ncbi.nlm.nih.gov/traces/sra61/SRR/006699/SRR# -> /f/ProjectSRAFiles/SRR#.sra.tmp.7523.tmp
2019-09-05T19:10:14 prefetch.2.9.3: KClientHttpOpen - connected from 'IP ADDRESS' to sra-download.ncbi.nlm.nih.gov (IP ADDRESS)
2019-09-05T19:10:14 prefetch.2.9.3: KClientHttpOpen - verifying CA cert
2019-09-05T19:13:16 prefetch.2.9.3 err: unknown while writing file within file system module - error writing system file - There is not enough space on the disk.(112)
2019-09-05T19:13:16 prefetch.2.9.3 int: unknown while writing file within file system module - Cannot KFileWrite: /f/ProjectSRAFiles/SRR#.sra.tmp.7523.tmp
2019-09-05T19:13:18 prefetch.2.9.3: 1) failed to download SRR#

UPDATE (09.06.19): Now with aspera fasp download, I am not even getting an error. Just a message that the download fails after downloading only ~4.5Gb of the data. I have tried a brand new hard drive, and it is the same problem so I don't think it is an issue of file size.

USER-MacBook-Pro-2:ncbi USER$ prefetch -v SRR#

2019-09-06T15:34:36 prefetch.2.9.3: Using '/Users/USER/Applications/Aspera Connect.app/Contents/Resources/ascp'
2019-09-06T15:34:36 prefetch.2.9.3: KClientHttpOpen - connected from 'IP ADDRESS' to www.ncbi.nlm.nih.gov (IP ADDRESS) 
2019-09-06T15:34:36 prefetch.2.9.3: KClientHttpOpen - verifying CA cert 
2019-09-06T15:34:36 prefetch.2.9.3: KClientHttpOpen - connected from 'IP ADDRESS' to sra-download.ncbi.nlm.nih.gov (IP ADDRESS) 
2019-09-06T15:34:36 prefetch.2.9.3: KClientHttpOpen - verifying CA cert 
2019-09-06T15:34:37 prefetch.2.9.3: KClientHttpOpen - connected from 'IP ADDRESS' to www.ncbi.nlm.nih.gov (IP ADDRESS) 
2019-09-06T15:34:37 prefetch.2.9.3: KClientHttpOpen - verifying CA cert 
2019-09-06T15:34:37 prefetch.2.9.3: 1) Downloading 'SRR#'...
2019-09-06T15:34:37 prefetch.2.9.3:  Downloading via fasp...
/Users/USER/Applications/Aspera Connect.app/Contents/Resources/ascp /Users/USER/Applications/Aspera Connect.app/Contents/Resources/ascp -i /Users/USER/Applications/Aspera Connect.app/Contents/Resources/asperaweb_id_dsa.openssh -pQTk1 -l 450m [e-mail]:data/sracloud/traces/sra61/SRR/006699/SRR# /Volumes/USER/CTAProjectSRAFiles/ncbi/public/sra/SRR#.sra.tmp.16642.tmp
SRR#                                      
2019-09-06T15:46:15 prefetch.2.9.3:  fasp download failed
2019-09-06T15:46:15 prefetch.2.9.3: 1) failed to download SRR#

[Arup Ghosh]

Open sratoolkit options vdb-config.exe -i to check if the default download path is set to the 1Tb hard drive.

[JJDollar]

It is set to the correct hard drive

[ATpoint]

Do yourself a favor and get a small Linux VM or partition and then simply download directly as fastq: Fast download of FASTQ files from the European Nucleotide Archive (ENA)

[JJDollar]

The sequences I am trying to download are not on ENA (yet?)

[ATpoint]

That is unfortunate. Still, the error message is clear, your disk is full. In the linked tutorial there is also a suggestion on how to efficiently use prefetch/parallel-fastq-dump, the paragraph starts with Edit 28.2.19: For matters of completeness. I suggest you do the download chunkwise, so get a sra file, convert to fastq, remove sra, maybe save fastq somewhere else to free disk space. As said, the tutorial is for Unix so you would probably need a VM.

[Mensur Dlakic]

It is not only a matter of setting a correct download directory. A temporary directory also must have enough disk space. See this thread for more information.

It says clearly in the message above that /f/ProjectSRAFiles/SRR#.sra.tmp.7523.tmp can't be written, meaning that /f/ProjectSRAFiles/ likely does not have enough space. I would make sure that old or incomplete downloads are purged from that directory as well.

[JJDollar]

The hard drive reports having plenty of space, and all incomplete files have been removed (I checked for hidden files as well).

[ATpoint]

Be also sure to deactivate the caching option in vdb-config. I cannot even tell you what exactly these cache files are that prefetch saves on disk, but they are large and take up space while not necessary for proper download.

[ATpoint]

Can you share the accession number? Alternatively, contact the NCBI support.

[GenoMax]

Please add:

I was trying to save to hard drives with Fat32 file system, which has wont save files >4Gb. Reformatted the hard drive, and the files downloaded.

as an answer so closure can be provided to this thread.

Answer 1

[JJDollar]

I was trying to save to hard drives with Fat32 file system, which has wont save files >4Gb. Reformatted the hard drive, and the files downloaded. Thank you for everyone' help!