KRAKEN2_BUILD fails on AWS megatest with 134 but insane amount of memory requested

[jfy133]

Description of the bug

https://cloud.seqera.io/orgs/nf-core/workspaces/AWSmegatests/watch/45RClgbxCJEiN6/v2/tasks

Command used and terminal output

Relevant files

No response

System information

No response

[jfy133]

Work around for now that should be implemented: should skip kraken2/brakcken in AWS megatest

[jfy133]

Error executing process > 'NFCORE_CREATETAXDB:CREATETAXDB:FASTA_BUILD_ADD_KRAKEN2_BRACKEN:KRAKEN2_BUILD (test_full)'

Caused by:
  Essential container in task exited


Command executed:

  kraken2-build \
      --build \
       \
      --threads 18 \
      --db kraken2-database/
  
  
  
  cat <<-END_VERSIONS > versions.yml
  "NFCORE_CREATETAXDB:CREATETAXDB:FASTA_BUILD_ADD_KRAKEN2_BRACKEN:KRAKEN2_BUILD":
      kraken2: $(echo $(kraken2 --version 2>&1) | sed 's/^.*Kraken version //; s/ .*$//')
  END_VERSIONS

Command exit status:
  134

Command output:
  Creating sequence ID to taxonomy ID map (step 1)...
  Sequence ID to taxonomy ID map already present, skipping map creation.
  Estimating required capacity (step 2)...
  Estimated hash table requirement: 514931564 bytes
  Capacity estimation complete. [1.252s]
  Building database files (step 3)...

Command error:
  10:39AM INF [seqera-fusion] setting up AWS Batch spot instance monitoring provider=aws-batch
  10:39AM INF [seqera-fusion] instance metadata ami-id=ami-068145e3a19edb6fd instance-id=i-01c7b4b79eb1b8307 instance-type=m6id.32xlarge
  10:39AM INF [seqera-fusion] starting incremental snapshot timer interval=300000
  10:44AM INF [seqera-fusion] sending incremental snapshot signal (SIGUSR1)
  grep: write error: Broken pipe
  terminate called after throwing an instance of 'std::invalid_argument'
    what():  stoul
  cat: write error: Broken pipe
  /opt/conda/share/kraken2-2.1.6/libexec/build_kraken2_db.sh: line 143:   293 Done                    list_sequence_files
         294 Exit 123                | xargs -0 cat
         295 Aborted                 (core dumped) | build_db -k $KRAKEN2_KMER_LEN -l $KRAKEN2_MINIMIZER_LEN -S $KRAKEN2_SEED_TEMPLATE $KRAKEN2XFLAG -H hash.k2d.tmp -t taxo.k2d.tmp -o opts.k2d.tmp -n taxonomy/ -m $seqid2taxid_map_file -c $required_capacity -p $KRAKEN2_THREAD_CT $max_db_flag -B $KRAKEN2_BLOCK_SIZE -b $KRAKEN2_SUBBLOCK_SIZE -r $KRAKEN2_MIN_TAXID_BITS $fast_build_flag
  FUSION_GPU_USED=false
  Creating sequence ID to taxonomy ID map (step 1)...
  Sequence ID to taxonomy ID map already present, skipping map creation.
  Estimating required capacity (step 2)...
  Estimated hash table requirement: 514931564 bytes
  Capacity estimation complete. [1.252s]
  Building database files (step 3)...
  Fusion Info:
      kernel_version: 6.1
      disk_cache_size: 7076Gb
      max_open_files: 1048576
      fusion_version: 2.4.20-ce52500
      clone_namespace: false

Work dir:
  s3://nf-core-awsmegatests/work/createtaxdb/work-55ace05e4276e0c5f29895499cd843c615c05f3e/6a/da508111ada00ab8bc005795c69ff2

But no furhter info