Skip to content

Crash after multiple black hole merger in Eagle 50Mpc box

I've had a 50Mpc Eagle run crash with the following on stdout:

  410482   7.766382e-03    0.6041746    0.6551507   2.364302e-08   35   35           61           61            0            0               186.073      0
[0013] [01411.2] runner_iact_nonsym_bh_bh_swallow: BH 6275087515187 wants to swallow BH particle 5821773462317
[0013] [01411.2] runner_iact_nonsym_bh_bh_swallow: BH 6275041571923 wants to swallow BH particle 6275087515187
[0013] [01411.2] runner_do_bh_swallow: BH 6275041571923 swallowing BH particle 6275087515187
[0013] [01411.2] runner_do_bh_swallow: BH 6275041571923 removing BH particle 6275087515187

and a segfault message on stderr:

srun: error: irene2195: task 13: Segmentation fault

This simulation was started with a version of swift from a week or so ago - it's commit 77dc3b54 with the fix for !1034 (merged) merged in.

In the commit log I see that 39740c15 fixes a problem with multiple black hole mergers, which I think this is. Does anyone know if this is likely to be the same problem and whether I can cherry pick the change and continue from the same restart file?

To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information