Concordance calculation hangs in recent versions of plink/seq

[Former user]

Testing on recent versions of pseq (git commit ids eae9674b, bea1a74) seems to reveal a bug in the concordance calculation process. I am loading two datasets from PLINK format:

      pseq db new-project --resources resources
      pseq db load-plink --file chip.data --id ichip --check-reference
      pseq db load-plink --file seq.data --id seq --check-reference
      pseq db concordance --mask reg=chr20 > concordance

This works correctly up to about 312K of output and then hangs. The pseq process still appears to be running on around 100% CPU, but no output is produced. In Plink/Seq 1.0.8 this error does not occur and the full output it produced rapidly.

Unfortunately I can't share the data this is running on, but if you have ideas I can try to rerun tests.

[Joshua Randall]

Note that eae9674b is on my fork, though that is unrelated to this issue as it occurs also in bea1a74. Not sure why Nick from my team shows up as Anonymous, but this is an issue we have seen with the latest development versions (it is not an issue in 0.08).