isoseq cluster2 crashes on "truncated" file generated by isoseq refine

[f-ferraro]

Operating system
Which operating system and version are you using?
Red Hat Enterprise Linux 8.6 (Ootpa)

Conda environment

# Name                    Version                   Build  Channel
isoseq                    4.2.0                h9ee0642_0    bioconda
lima                      2.12.0               h9ee0642_1    bioconda
pbmm2                     1.16.0               h9ee0642_0    bioconda

Describe the bug
Hi, I am trying to process some Kinnex RNASeq data.
After isoseq refine, which ends without error, isoseq cluster2 runs for a while and then stops, I think at the sorting step, complaining for the file (generated by refine) to be corrupted. I tried running it a couple of times from scratch but with no success. Could suggest how to fix this, please? :)
Thank you in advance!

Error message and to reproduce
pacbioerror.txt

[armintoepfer]

Your input is corrupt.

[E::bgzf_read_block] Invalid BGZF header at offset 7865347630

Maybe there's an issue with you HPC? Is the data local or mounted via NFS?

[f-ferraro]

Thanks for the reply!

Yes, the data is mounted via NFS. Do you have perhaps any pointers to see if it's an issue of the HPC?

[armintoepfer]

Maybe the random access doesn't work properly. Is there any way to keep the intermediate files local to that node?

[f-ferraro]

I'll talk with our admin. Thank you for the help!