Water Resource Group

This document will list tips for WRgroup

Computers

  • Linux Server, 128.118.212.69
  • Small Linux, 130.203.207.218
  • Windows Work Station, 130.203.207.223
  • My Computer, 130.203.208.181

tight VNC

tightVNC tutorial to setup

Start local VNC:

# small Linux
ssh -L 5901:127.0.0.1:5901 -N -f -l kxf227 130.203.207.218
# Linux Server
ssh -L 5902:127.0.0.1:5902 -N -f -l kxf227 128.118.212.69

Restart local VNC:

netstat -ntlp
kill -9 PID

Restart server VNC:

vncserver -kill :1
vncserver :1
sudo systemctl daemon-reload
sudo systemctl enable vncserver@1.service
sudo systemctl start vncserver@1

hack to fix VScode / Atom display:

mkdir ~/lib
cp /usr/lib/x86_64-linux-gnu/libxcb.so.1 ~/lib
cp /usr/lib/x86_64-linux-gnu/libxcb.so.1.1.0 ~/lib
sed -i 's/BIG-REQUESTS/_IG-REQUESTS/' /usr/lib/x86_64-linux-gnu/libxcb.so.1.1.0

Memory Manage

  1. cgroup 可以限制用户的内存. Steps:
  • 更改:

    /etc/cgconfig.conf  #内存大小
    /etc/cgrules.conf  #限制用户
    
  • 执行下面的命令:

    cgconfigparser -l /etc/cgconfig.conf
    cgrulesengd
    
  • cgroup tutorial

2. 更改 /etc/security/limits.conf 然后重启 这个我没太搞明白。但应该是限制单个process的。

Mount Drives

  • find uuid:

    sudo blkid
    
  • modify /etc/fstab

    just follow this example:

    UUID=b4aefb54-e1c7-4378-9380-13eea35ad943   /mnt/sdc/   ext4    defaults    2   2
    
  • restart or mount -a

CUDA

Add path:

export PATH=/usr/local/cuda-8.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64\ $LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

Jupyter Notebook

Remote notebook:

  • terminal 1:

    ssh kxf227@130.203.207.218
    jupyter notebook --no-browser --port=8889
    
  • terminal 2:

    ssh -nNT -L localhost:8888:localhost:8889 kxf227@130.203.207.218
    
  • browser:

    localhost:8888