I received a call from Jeff Bartlett that the 3 working HWS have all stopped (ITMX was not running due to the clipping issue in the vertex).

After getting permission from Jeff, I logged into the HWS machines and found that they all stopped at the same time. Checking the disk usage confirmed my suspicion that the data disk (/data) was indeed full. We'll have to look into whether the clean-up script is still working - it might be but it's possible that it wasn't clearing out everything and this, like Thanos, was just inevitable. Once I confirm that all data more than 48 hours old is archived at CIT, we can remove all the old data and get the HWS code running again.