Closed apredeus closed 10 months ago
This has been resolved; I missed the warning that, when installed with conda, parallel-fastq-dump
tends to install a really old version of sra-tools
as a dependency. Forcing version 3.0.8 has solved the issue.
Dear SRA team,
I'm trying to use
parallel-fastq-dump
, which I'm sure as you know is a thin wrapper overfastq-dump
. I keep getting error code 3 with at least one of thefastq-dump
sub-processes, while I have plenty of lustre space (over 100 Tb), and has configured the tools to not use cache (and I indeed see that it's not used). The SRA files I'm processing are pre-downloaded, and they are quite large (20 Gb or more). Are there any parameters I can tune to avoid this error?Thank you in advance!