spades-hammer finishing with abnormally, OS return value: -7 but with an (possibly) error message saying 'None' after 'writing down entries'

[jfy133]

Description of bug

For quite a while now I've been receiving errors of the following when running metaSPAdes.

== Error ==  system call for: "['/usr/local/bin/spades-hammer', '/fusion/s3/nf-core-awsmegatests/work/mag/work-409d36e985e8762373533852bcec10a78434ba8e/5d/9167bdfabd754e97516dcbf2eb48e5/corrected/configs/config.info']" finished abnormally, OS return value: -7
None

Based on other issues this is normally associated with memory issues, but I'm now running wth ~1TB of RAM and it's still happening at the same point, I think possibly related to this:

  0:35:06.762  6933M / 8740M INFO    General                 (concurrent_dsu.cpp        :  64)   Writing down entries

However there is no further error (as seen above, it just says None).

Can you provide any suggestions how to further debug this?

(Note there are upstream malloc warnings, but in other issues I've seen the advice is to ignore them)

spades.log

spades.log

params.txt

params.txt

SPAdes version

v4.0.0

Operating System

Debian GNU/Linux 12 (bookworm)

Python Version

3.12.5

Method of SPAdes installation

docker container (biocontainers)

No errors reported in spades.log

• Yes

[handesome]

same error，did you solved it yet？

[jfy133]

No, had it for a while :(

[TrinyEG]

Any solution to this error?

[jfy133]

Nope :(

[asl]

Nope :(

Your filesystem likely does not support memory mapped files properly. So, you'd need to move scratch directory to something more reliably

[jfy133]

Nope :(

Your filesystem likely does not support memory mapped files properly. So, you'd need to move scratch directory to something more reliably

That's helpful @asl !

In my case this happens on AWS but with a 'special' filesystem (fusion by seqera), so I will follow up with them about this and see if that cou6be the case :)