NVTOP

To install nvtop, download Makefile.nvtop, and issue the following commands

make -f Makefile.nvtop
sudo make -f Makefile.nvtop install

Authentication

auth3.py is a python3 avatar of our auth.py functionality. And an example of beautiful coding by @R. Harish - Thanks.
This was posted about 4 months ago, but I was a little loose... 😜

08:04 bvr@bvr ~> python -m auth3 -h
usage: auth3.py [-h] [-u USERNAME] [-p PASSWORD] [-q]
Authentication script for FortiNet
optional arguments:
  -h, --help            show this help message and exit
  -u USERNAME, --username USERNAME
                        FortiNet username
  -p PASSWORD, --password PASSWORD
                        FortiNet password
  -q, --quiet           disable verbose output

This can also be seamlessly configured to be used as a command, say authenticate, as follows:

1. cp auth3.py ~/.local/bin/authenticate
2. chmod +x ~/.local/bin/authenticate: Yes, this has been written to be a script; with a shebang header
3. Add export PATH=$PATH:~/.local/bin to ~/.profile, and source ~/.profile
4. authenticate --- and it works...

GPU Monitor

This is a GPU server monitor script, to extract values from nvidia-smi and stack into a space delimited list.

usage: gpu_monitor.py [-h] [-v] [-i PATH] [-o PATH]

Parse `nvidia-smi' and write a line of log for each gpu.

optional arguments:
  -h, --help            show this help message and exit
  -v, --verbose         Verbose logging. (default: False)
  -i PATH, --input PATH
                        Input Filename (default: -)
  -o PATH, --output PATH
                        Output Filename (default: -)

This has been used to set up a gpu monitoring cron job, with a frequency of 5 mins, through gpu_monitor.sh. The output is a space delimited list as follows:

1576212658.9938178 0 30.0 52.0 46.0 0.0 0.0
1576212658.993862 1 30.0 60.0 47.0 0.0 0.0
1576212658.9938943 2 28.0 46.0 39.0 0.0 0.0
1576212658.9939241 3 29.0 38.0 38.0 0.0 0.0

The entries in each column respectively represent,

1. Time stamp as no. of secs since epoch
2. GPU id
3. Fan Speed (in %)
4. GPU Temperature (in deg C)
5. Power consumption (in W)
6. Memory Occupancy (in MiB)
7. GPU utility (in %)