start_qrl core dumps

[pibara]

Describe the bug When running start_qrl with the option --qrldir, the command eventually core dumps. When restarted is core dumps promptly.

Expected behavior Something else than a coredump

Screenshots Added a bit of code to start_qrl to make it use faulthandler

#!/usr/bin/python3
# -*- coding: utf-8 -*-
import re
import sys
import faulthandler
from qrl.main import main
if __name__ == '__main__':
    faulthandler.enable()
    sys.argv[0] = re.sub(r'(-script\.pyw|\.exe)?$', '', sys.argv[0])
    sys.exit(main())

The result:

When running with logging set to DEBUG, the last logged info is:

Desktop:

• OS: Ubuntu 18.04

[cyyber]

Please check your available disk space for /export/home/rob/qrl

[pibara]

Please check your available disk space for /export/home/rob/qrl

[cyyber]

Does your CPU support aes-ni ? Check via following command if it gives you any output cat /proc/cpuinfo | grep aes

[pibara]

[cyyber]

Please provide output of following commands

lsb_release -a

free

cmake --version

gcc --version

g++ --version

python3 --version

[pibara]

[cyyber]

We tried to setup the node with same environment as mentioned above (with 2 GB RAM) and we are not able to repeat the above issue. I think this is something more related to your environment or hardware. This issue may also happen if for some reason the filesystem is broken. I will suggest you to setup a fresh node.