How to wake up fastq-dump --split 3 process?

Question

[Cy]

Hi everyone,

I run the fastq-dump --split 3 to split the sra.file but it took a very long time. When I check the process viewer through htop, I found that the fastq-dump --split 3 command is in s (interruptible sleep) status. May I know how to solve this problem?

[GenoMax]

If you need data for a few accession numbers then just searching EBI-ENA directly and getting the fastq files is the most efficient way.

[Cy]

Yes, I found it on EBI-ENA. Thank you. ^^ But I want to learn how to solve this problem also. I waited the process to run and the system come out with this result:

2018-07-10T17:40:50 fastq-dump.2.9.1 sys: timeout exhausted while reading file within network system module - mbedtls_ssl_read returned -76 ( NET - Reading information from the socket failed )

2018-07-10T20:05:43 fastq-dump.2.9.1 sys: timeout exhausted while reading file within network system module - mbedtls_ssl_read returned -76 ( NET - Reading information from the socket failed )

2018-07-10T21:10:01 fastq-dump.2.9.1 sys: timeout exhausted while reading file within network system module - mbedtls_ssl_read returned -76 ( NET - Reading information from the socket failed )

Read 64821551 spots for SRR5257023

Written 64821551 spots for SRR5257023

So, may I consider the process was finished?

[GenoMax]

Did you get any errors downloading from ENA? That appears to be some kind of network error. Are you using a network proxy (have a firewall between you and NCBI) and have you properly configured SRAtoolkit to use the proxy? There was a past similar issue reported for fastq-dump that appears to be still not closed (may not mean much).

[Cy]

Thank you for your help. I had solved the problem.

[ATpoint]

How? Maybe you can give some input for others.

Answer 1

[ATpoint]

Fastq-dump is a terribly annoying piece of software. I had issues myself many times and cannot really provide a solution. Check by ls -lh if the file is progressingly getting larger. if so, just wait. S means that the tool is waiting for I/O. I would check if the file is backed up at the European nucleotide Archive (ENA). If so, you can get the fastq file there directly. Have a look at my recent tutorial on how to retrieve data from the ENA.

[Cy]

Hi, thank you for your suggestion. I found the file at the ENA. But I want to solve this problem also. I waited the process to run and the system come out with this result:

2018-07-10T17:40:50 fastq-dump.2.9.1 sys: timeout exhausted while reading file within network system module - mbedtls_ssl_read returned -76 ( NET - Reading information from the socket failed )

2018-07-10T20:05:43 fastq-dump.2.9.1 sys: timeout exhausted while reading file within network system module - mbedtls_ssl_read returned -76 ( NET - Reading information from the socket failed )

2018-07-10T21:10:01 fastq-dump.2.9.1 sys: timeout exhausted while reading file within network system module - mbedtls_ssl_read returned -76 ( NET - Reading information from the socket failed )

Read 64821551 spots for SRR5257023

Written 64821551 spots for SRR5257023

So, may I consider the process was finished?

[ATpoint]

Well, the number of spots matches the description of the file, so yes it probably finished. That is exactly the point with this terrible software. It sometimes causes strange errors and leaves you with riddles. To be safe, run fastQvalidator on it, or simply align it and see if the aligner complains about anything. THere is also a new tool, fasterq-dump in the latest release of the sratoolkit, which is faster but (and at this point a hugh WTF to NCBI), does no longer has a --gzip option. If you have the disk space, you might use it in the future, but better avoid SRA files whenever possible.

[Cy]

Thank you for your help. I had solved the problem.