I am getting a no space left on device error even though i am removing the files in the tmp directory. Can you please look into this
I am unable to take a look right now.
Could you please give me a Run ID?
RUN ID #31266 - This is the runid
The error is very weird:
filename = ‘/tmp/patches/xxx’, file descriptor = 17, errno = 28, error message = ‘No space left on device’, buf = 0x1220dcb8, total write size = 96, bytes this sub-write = 96, bytes actually written = 18446744073709551615, offset = 0)
Especially: bytes actually written = 18446744073709551615
Which is equivalent to 17179869184 Gb…
GPU Runs already have access to 80Gb.
If you really need more, I can increase, but the bug is likely due to:
- your code
- or library version mismatch
I am removing the temporary files that i am creating can you confirm that the files are indeed removed because that is the issue also i ran this code on my local and i dont write such huge files i will look at my code again but if you can confirm that i am removig my tmp files that would be great
I cannot really confirm that for you, but what you can do is this:
import os
os.system("ls /tmp")
It will list the content in the console so that you can visually confirm that the directory is empty.
is it possible to double the space on the device to 160 GB
I made a mistake and misunderstood the big number, it is actually an unsigned long max value. So likely a -1 if signed. Meaning nothing has even been wrote yet.
I increased the size to 300 gb for GPU Runs only.