Mothioncorrection jobs fail with strange behavior

Hi, I have quite an extensive dataset with 22000 micrographs. When I try to run the motion correction on the micrographs it tends to fail after a while. I got a couple of times that it said failed but still run quite a while without an issue because it didnt receive an heartbeat but actually did and now after restarting I got an error that it is out of space.
========= sending heartbeat
Traceback (most recent call last):
File “”, line 1, in
File “cryosparc2_worker/cryosparc2_compute/”, line 162, in
IOError: [Errno 28] No space left on device

Bt also this is not true there is more than enough space left and when I copy the same job right after the error it can go on writing. what could be the issue. Is it writing somewhere else except the project folder?

Hi @david.haselbach,

Line 162 is sys.stdout.flush()

I’m not entirely sure why you’re getting this error. Can you try running df -i? Maybe you have too many small directories?