'std::bad_alloc' on EC2

Hello,

I find KB python super useful and efficient. Recently I have moved to work EC2 instances and to test that everything was good I decided to re-run a sample in there using KB, the same version and index. I was previously working in a dragen server and sometimes my kb jobs would fail but sometimes the same sample would finish. Because of that I was eager to see if moving to EC2 would be better, but now I cannot run kb in my EC2 instance. The installation seems to be fine.

This is my command (notice I am trying to run the workflow lamanno)

IDX=/home/ubuntu/kb_lamanno_index_CAR/index.idx 
T2G=/home/ubuntu/kb_lamanno_index_CAR/t2g.txt 
SPL=/home/ubuntu/kb_lamanno_index_CAR/spliced_t2c.txt
UNSPL=/home/ubuntu/kb_lamanno_index_CAR/unspliced_t2c.txt

kb count --loom -i $IDX -g $T2G -x 10xv2 -o $sample -c1 $SPL -c2 $UNSPL --workflow lamanno --filter bustools \
sample_01_L001_R1.fastq.gz sample_01_L001_R2.fastq.gz sample_01_L002_R1.fastq.gz sample_01_L002_R2.fastq.gz

It runs for about 5 min and then I get:

[2022-06-28 22:07:53,285]    INFO [count_lamanno] Using index /home/ubuntu/kb_lamanno_index_CAR/index.idx to generate BUS file to SM135_batch1_01_S1 from
[2022-06-28 22:07:53,285]    INFO [count_lamanno]         /home/ubuntu/BioInfo-SM-135/batch1/SM135_batch1_01_S1_L001_R1_001.fastq.gz
[2022-06-28 22:07:53,285]    INFO [count_lamanno]         /home/ubuntu/BioInfo-SM-135/batch1/SM135_batch1_01_S1_L001_R2_001.fastq.gz
[2022-06-28 22:07:53,285]    INFO [count_lamanno]         /home/ubuntu/BioInfo-SM-135/batch1/SM135_batch1_01_S1_L002_R1_001.fastq.gz
[2022-06-28 22:07:53,285]    INFO [count_lamanno]         /home/ubuntu/BioInfo-SM-135/batch1/SM135_batch1_01_S1_L002_R2_001.fastq.gz
[2022-06-28 22:10:41,559]   ERROR [count_lamanno] 
[bus] Note: Strand option was not specified; setting it to --fr-stranded for specified technology
[index] k-mer length: 31
[index] number of targets: 1,455,417
[index] number of k-mers: 1,578,640,741
[index] number of equivalence classes: 128
terminate called after throwing an instance of 'std::bad_alloc'
what():  std::bad_alloc
[2022-06-28 22:10:41,560]   ERROR [main] An exception occurred
Traceback (most recent call last):
  File "/home/ubuntu/anaconda3/lib/python3.9/site-packages/kb_python/main.py", line 1301, in main
    COMMAND_TO_FUNCTION[args.command](parser, args, temp_dir=temp_dir)
  File "/home/ubuntu/anaconda3/lib/python3.9/site-packages/kb_python/main.py", line 482, in parse_count
    count_velocity(
  File "/home/ubuntu/anaconda3/lib/python3.9/site-packages/ngs_tools/logging.py", line 62, in inner
    return func(*args, **kwargs)
  File "/home/ubuntu/anaconda3/lib/python3.9/site-packages/kb_python/count.py", line 1915, in count_velocity
    bus_result = kallisto_bus(
  File "/home/ubuntu/anaconda3/lib/python3.9/site-packages/kb_python/validate.py", line 116, in inner
    results = func(*args, **kwargs)
  File "/home/ubuntu/anaconda3/lib/python3.9/site-packages/kb_python/count.py", line 190, in kallisto_bus
    run_executable(command)
  File "/home/ubuntu/anaconda3/lib/python3.9/site-packages/kb_python/dry/__init__.py", line 25, in inner
    return func(*args, **kwargs)
  File "/home/ubuntu/anaconda3/lib/python3.9/site-packages/kb_python/utils.py", line 203, in run_executable
    raise sp.CalledProcessError(p.returncode, ' '.join(command))

I have tried to increase the memory up to 100 and my instance (r4.8xlarge) should have no issue. I tried looking into other issues like #49 but that seems to be more 'platform based'. When using cell ranger this sample detected 6798 cells after filtering.

Maybe I am missing something obvious, thank you so much for your support.

BUStools / bustools

'std::bad_alloc' on EC2 #79