Skip to content

Cellranger arc running out of disk space #436

@nchernia

Description

@nchernia

I'm running the latest version of Cumulus (3.0.0). I was able to successfully run CellRanger arc on some of my samples, but two of them have been crashing, it appears because they do not have enough disk space.

I have rerun setting disk space and memory very high:
arc_disk_space 1250
arc_memory "400G"
atac_disk_space 1250
atac_memory "250G"
count_disk_space 1000

However they are still crashing (after running for longer) and this is getting to be very expensive.

Any advice?

Here is the end of the log file:

2025-04-04 13:25:19 [runtime] (update) ID.results.SC_ATAC_GEX_COUNTER_CS.SC_ATAC_GEX_COUNTER._ATAC_MATRIX_COMPUTER.ALIGN_ATAC_READS.fork0 chunks running (144/239 completed)
2025-04-04 13:30:33 [runtime] (update) ID.results.SC_ATAC_GEX_COUNTER_CS.SC_ATAC_GEX_COUNTER._ATAC_MATRIX_COMPUTER.ALIGN_ATAC_READS.fork0 chunks running (144/239 completed)
2025-04-04 13:35:43 [runtime] (update) ID.results.SC_ATAC_GEX_COUNTER_CS.SC_ATAC_GEX_COUNTER._ATAC_MATRIX_COMPUTER.ALIGN_ATAC_READS.fork0 chunks running (144/239 completed)
2025-04-04 13:40:45 [runtime] (update) ID.results.SC_ATAC_GEX_COUNTER_CS.SC_ATAC_GEX_COUNTER._ATAC_MATRIX_COMPUTER.ALIGN_ATAC_READS.fork0 chunks running (144/239 completed)
2025-04-04 13:45:52 [runtime] (update) ID.results.SC_ATAC_GEX_COUNTER_CS.SC_ATAC_GEX_COUNTER._ATAC_MATRIX_COMPUTER.ALIGN_ATAC_READS.fork0 chunks running (145/239 completed)
2025-04-04 13:50:56 [runtime] (update) ID.results.SC_ATAC_GEX_COUNTER_CS.SC_ATAC_GEX_COUNTER._ATAC_MATRIX_COMPUTER.ALIGN_ATAC_READS.fork0 chunks running (150/239 completed)
2025-04-04 13:56:08 [runtime] (update) ID.results.SC_ATAC_GEX_COUNTER_CS.SC_ATAC_GEX_COUNTER._ATAC_MATRIX_COMPUTER.ALIGN_ATAC_READS.fork0 chunks running (156/239 completed)
2025-04-04 13:58:20 [runtime] Pipestance directory out of disk space.
/cromwell_root/results/SC_ATAC_GEX_COUNTER_CS has only 9004kB remaining space available.
To ignore this error, set MRO_DISK_SPACE_CHECK=disable in your environment.
2025-04-04 13:58:23 [runtime] (failed) ID.results.SC_ATAC_GEX_COUNTER_CS.SC_ATAC_GEX_COUNTER._ATAC_MATRIX_COMPUTER.ALIGN_ATAC_READS

[error] Pipestance failed. Error log at:
results/SC_ATAC_GEX_COUNTER_CS/SC_ATAC_GEX_COUNTER/_ATAC_MATRIX_COMPUTER/ALIGN_ATAC_READS/fork0/chnk158-u0a3bef7605/_errors

Log message:
/cromwell_root/results/SC_ATAC_GEX_COUNTER_CS has only 9004kB remaining space available.
To ignore this error, set MRO_DISK_SPACE_CHECK=disable in your environment.

Waiting 6 seconds for UI to do final refresh.
Pipestance failed. Use --noexit option to keep UI running after failure.

2025-04-04 13:58:29 Shutting down.
cellranger-arc count --id=results --libraries=libraries.csv --reference=genome_dir --jobmode=local --localcores=64 --localmem=500
Traceback (most recent call last):
File "", line 40, in
tee: /cromwell_root/stderr: No space left on device
File "/usr/lib/python3.9/subprocess.py", line 373, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['cellranger-arc', 'count', '--id=results', '--libraries=libraries.csv', '--reference=genome_dir', '--jobmode=local', '--localcores=64', '--localmem=500']' returned non-zero exit status 1.
/cromwell_root/script: line 78: echo: write error: No space left on device
sh: write error: No space left on device

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions