Cuju: An Open Source Project for Virtualization-Based Fault Tolerance

================================================================

Summary

Virtualization technology could provide a unique benefit to protect any legacy application systems from hardware failures. The reliability of virtual machines running on virtualized servers is not only threatened by hardware failures beneath the whole virtual infrastructure, but also nosy hypervisors that essentially support virtual machines cannot be trusted.

We develop the opensource tool, Cuju, which is a virtualization based fault tolerance technique with epoch-based synchronization. There are several performance optimization technologies are applied by Cuju, including a non-stop/pipelined, continuously migration, dirty tracking for guest virtual memory/virtual device status, and eliminate data transfer between QEMU and KVM.

Cuju shows that these optimizations have greatly saved the processor usage, synchronization bandwidth and have significantly improved VM network throughput and latency at the same time.

For more information see: https://cuju-ft.github.io/cuju-web/home.html

Cuju Install Guide

The environment prepare

All Node Install

Assume you have already builded Primary, Backup and NFS node.

A recommended topology below:

If you only have one or two machine, you can reference this setting.

Another recommended topology below:

Open the Intel virtualization support (VT-x) in your bios.
Install OS in all nodes: (pick 1 of 2)
- Ubuntu-16.04-desktop-amd64.iso (Ubuntu 16.04.0)
- ubuntu-18.04.1-live-server-amd64.iso (Ubuntu 18.04.1)
Install related packages in all nodes

 $ sudo apt-get update

ubuntu-16:
 $ sudo apt-get install ssh vim gcc make gdb fakeroot build-essential \
kernel-package libncurses5 libncurses5-dev zlib1g-dev \
libglib2.0-dev qemu xorg bridge-utils openvpn vncviewer \
libssl-dev libpixman-1-dev nfs-common git

ubuntu-18:
 $ sudo apt-get install ssh vim gcc make gdb fakeroot build-essential \
kernel-package libncurses5 libncurses5-dev zlib1g-dev \
libglib2.0-dev qemu xorg bridge-utils openvpn libelf-dev \
libssl-dev libpixman-1-dev nfs-common git tigervnc-viewer

Set up the bridge and network environment
- You can follow our recommended topology to set up the network environment
- The example of network interfaces set up below (edit your /etc/network/interfaces):
NFS node

auto lo
iface lo inet loopback

auto eth0
iface eth0 inet static
address 192.168.11.1
netmask 255.255.255.0
gateway 192.168.11.254
dns-nameservers 8.8.8.8

eth0 is your physical NIC name, please modify it according to your actual NIC name

Primary node

auto lo
iface lo inet loopback

auto br0
iface br0 inet static
bridge_ports eth0
bridge_maxwait 0
address 192.168.11.2
netmask 255.255.255.0
gateway 192.168.11.254
dns-nameservers 8.8.8.8

auto eth0
iface eth0 inet static
address 0.0.0.0

auto eth1
iface eth1 inet static
address 192.168.111.1
netmask 255.255.255.0

Backup node

auto lo
iface lo inet loopback

auto br0
iface br0 inet static
bridge_ports eth0
bridge_maxwait 0
address 192.168.11.3
netmask 255.255.255.0
gateway 192.168.11.254
dns-nameservers 8.8.8.8

auto eth0
iface eth0 inet static
address 0.0.0.0

auto eth1
iface eth1 inet static
address 192.168.111.2
netmask 255.255.255.0

Build the high-speed connections (ex. 10G NIC) with Primary and Backup nodes by the eth1
After editing these network interfaces, type /etc/init.d/networking restart or reboot

NFS Node Setup

Install the NFS service (Network FileSystem) in NFS node; then create a NFS folder placing the VM image

 $ sudo apt-get install nfs-kernel-server

Insert this line in /etc/exports to add your NFS folder:

 /home/[your username]/nfsfolder *(rw,no_root_squash,no_subtree_check)

After editing /etc/exports, type /etc/init.d/nfs-kernel-server restart or reboot
Go to your nfs folder, then download Cuju and build a VM image file (or download our Ubuntu-16.04 VM image file, the account/password is root/root), they will be synced with Primary and Backup node.

Primary and Backup Node Setup

Mount the NFS folder

$ sudo mkdir /mnt/nfs
$ sudo mount -t nfs 192.168.11.1:/home/[your username]/nfsfolder /mnt/nfs

Build Cuju

Install the appropriate version of the kernel for Cuju (only on ubuntu-18)

$ sudo apt-get install linux-image-4.15.0-153-generic
$ sudo apt-get install linux-headers-4.15.0-153-generic

Clone Cuju on your NFS folder from Github

$ cd /mnt/nfs
$ git clone https://github.com/Cuju-ft/Cuju.git

Configure & Compile Cuju-ft

$ cd Cuju
$ ./configure --enable-cuju --enable-kvm --disable-pie --disable-xen --target-list=x86_64-softmmu
$ make clean
$ make -j8

Configure, Compile & insmod Cuju-kvm module *1 *2

$ cd Cuju/kvm
$ ./configure
$ make clean
$ make -j8
$ ./reinsmodkvm.sh

P.S.

*1 If you meet error: incompatible type for argument 5 of '__get_user_pages_unlocked', you can use this patch:
$ cd Cuju
$ patch -p1 < ./patch/__get_user_pages_unlocked.patch
*2 If you meet error: implicit declaration of function 'use_eager_fpu' [-werror=implicit-function-declaration], you can use this patch:
$ cd Cuju
$ patch -p1 < ./patch/use_eager_fpu.patch

Execute Cuju

Before launching your VM, you should update kvm module in Primary and Backup nodes:

$ cd /mnt/nfs/Cuju/kvm
$ ./reinsmodkvm.sh

Boot VM (on Primary Host, /mnt/nfs/Cuju)
runvm.sh

sudo ./x86_64-softmmu/qemu-system-x86_64 \
-drive if=none,id=drive0,cache=none,format=raw,file=/mnt/nfs/Ubuntu20G-1604.img \
-device virtio-blk,drive=drive0 \
-m 1G -enable-kvm \
-net tap,ifname=tap0 -net nic,model=virtio,vlan=0,macaddr=ae:ae:00:00:00:25 \
-cpu host \
-vga std -chardev socket,id=mon,path=/home/[your username]/vm1.monitor,server,nowait -mon chardev=mon,id=monitor,mode=readline

You need to change the guest image path (file=/mnt/nfs/Ubuntu20G-1604.img) and monitor path (path=/home/[your username]/vm1.monitor) for your environment

Use VNC to see the console

$ vncviewer :5900 &

The default account/password is root/root if you use we provide guest image

Start Receiver (on Backup Host, /mnt/nfs/Cuju)
recv.sh

sudo x86_64-softmmu/qemu-system-x86_64 \
-drive if=none,id=drive0,cache=none,format=raw,file=/mnt/nfs/Ubuntu20G-1604.img \
-device virtio-blk,drive=drive0 \
-m 1G -enable-kvm \
-net tap,ifname=tap1 -net nic,model=virtio,vlan=0,macaddr=ae:ae:00:00:00:25 \
-vga std -chardev socket,id=mon,path=/home/[your username]/vm1r.monitor,server,nowait -mon chardev=mon,id=monitor,mode=readline \
-cpu host \
-incoming tcp:0:4441,ft_mode

You need to follow Boot VM script to change the related parameter or you can use following script to replace Receiver start script (if your VM start script is runvm.sh)
recv.sh

sed -e 's/mode=readline/mode=readline -incoming tcp\:0\:4441,ft_mode/g' -e 's/vm1.monitor/vm1r.monitor/g' -e 's/tap0/tap1/g' ./runvm.sh > tmp.sh
chmod +x ./tmp.sh
./tmp.sh

After VM boot and Receiver ready, you can execute following script to enter FT mode
ftmode.sh

ubuntu-16:
sudo echo "migrate -c tcp:192.168.111.2:4441" | sudo nc -U /home/[your username]/vm1.monitor

ubuntu-18:
sudo echo "migrate -c tcp:192.168.111.2:4441" | sudo nc -w 1 -U /home/[your username]/vm1.monitor

You need to change the ip address and port (tcp:192.168.111.2:4441) for your environment, this is Backup Host's IP And change the monitor path (/home/[your username]/vm1.monitor) for your environment

If you successfully start Cuju, you will see the following message show on Primary side:
If you want to test failover You can kill or ctrl-c VM on the Primary Host
You will need new session with vncviewer:
- If you have Primary Host and Backup Host, execute on Backup Host:
  $ vncviewer :5900 &
- If you only have Primary Host with two VM:
  $ vncviewer :5901 &

EndeavorYen/Cuju