Today I made a couple of attempts at Colab. None of the four instances ran for more than 20 minutes. This was with one notebook per account. One account is still paid, though now cancelled. Colab has been my main processing contribution since I got hooked into it. I have retired my one functional GPU. I have been running P-1 and DCs on two CPUs, but I think I'll let it all rest for a bit. I'll even be able to turn them off once in a while! :smile:

[QUOTE=Aramis Wyler;564679]I noticed today that after being especially persistant about re-starting the sessions when they die, it started giving me a different error - that there simply were no gpus available. Not that I'd hit my quots, or couldn't connect to a gpu, but that there were none.[/QUOTE]

I quite consistently get this message on my 22nd attempt from an individual session.

Newe colab error message today
 

[CODE]/bin/bash: LD_LIBRARY_PATH]=lib:/usr/lib64-nvidia: No such file or directory[/CODE]

This is my process

[CODE]import subprocess
import os
import os.path
from google.colab import drive
import sys
if not os.path.exists('/content/drive/My Drive'):
drive.mount('/content/drive')

%cd '/content/drive/My Drive/HOOTER/'
!cp libstdc* /usr/lib/x86_64-linux-gnu/

!LD_LIBRARY_PATH]="lib:${LD_LIBRARY_PATH}"

!./gpuowl.exe >>gpuowllog.txt 2>&1 &
print('gpuowl launched in background')[/CODE]

Oh and for the last couple days, instead of 8 hours per session per day I am getting 4 or 5.

I haven't even opened up Colab for well over a week. Didn't seem worth the effort.

Today I tried for kicks and got a T4 for an hour, enough to complete and submit four 73-74 TFs. I had to complete the 5th partially completed one manually using mfaktc.

Still not worth it for me.

[QUOTE=petrw1;565988][CODE]/bin/bash: LD_LIBRARY_PATH]=lib:/usr/lib64-nvidia: No such file or directory[/CODE]
[/QUOTE]This notebook continues to work with P100, K-80 without any problems with gpuOwl 6.11.380 at PRP/LL. For T4, P4 you have to leave out the gpuOwl options in the last line so that you get the "optimal" performance for this cards.
[CODE]
import os.path
from google.colab import drive
if not os.path.exists('/content/drive/My Drive'):
drive.mount('/content/drive')

#!unzip -o -d '/' '/content/drive/My Drive/Directory.zip'
#!zip -r '/content/drive/My Drive/Directory.zip' '/content/drive/My Drive/gpuowl-master'

[COLOR="Red"]os.chdir('/content/drive/My Drive/')
[/COLOR]
!cp libstdc* /usr/lib/x86_64-linux-gnu/

!nvidia-smi -L
[COLOR="Red"]os.chdir ('/content/drive/My Drive/gpuowl-master/')[/COLOR]
!chmod 755 '/content/drive/My Drive/gpuowl-master/gpuowl.exe'
!ls -la
!cd '.' && /content/drive/My\ Drive/gpuowl-master/gpuowl.exe -carry short -use CARRY32,ORIG_SLOWTRIG,IN_WG=128,IN_SIZEX=16,IN_SPACING=4,OUT_WG=128,OUT_SIZEX=16,OUT_SPACING=4 -nospin -block 100 -maxAlloc 10000 -B1 750000 -rB2 20
[/CODE]