A good amount of the enterprise IT infrastructure runs on VMware. Increase in digital traffic requires the business services to remain highly available to the end users. The best operating practise in such cases is to continuously validate the service reliability and optimise the recovery procedures. Faults of various types are designed and injected into VMware infrastructure and application recovery is validated. Service Level Objectives are continuously monitored by the ops teams while chaos is in play at virtual infrastructure. ChaosNative Litmus offers various chaos experiments for VMware infrastructure using which chaos workflows are designed and executed.
Continuous validation of SLOs when VMware vSphere is the underlying infrastructure
Evaluate the resilience of Kubernetes infrastructure that is built on VMware vSphere
Continuous chaos testing of virtual infrastructure in CI pipelines
Validate the functioning of monitoring and observability platforms built on virtual machine infrastructure
DR validation scenarios of business services built with virtual machine infrastructure
Chaos Experiment
Description
Tunable Parameters
Disk Loss Experiment
Description
VMWare Disk Loss experiment causes the detachment of Virtual Disks attached to a VM for a specified duration of time and later attaches them back to the VM.
Tunables
Necessary Inputs:
Disk IDs
Disk VM MOIDs
Secrets Inputs:
vCenter Server URL
vCenter Username
vCenter Password
VM PowerOff Experiment
Description
VMware VM Power-off experiment causes power-off of the VMWare VM(s) provided by the Moid(s) of the instance.
Tunables
Necessary Inputs:
VM MOIDs
Secrets Inputs:
vCenter Server URL
vCenter Username
vCenter Password
Process Kill Experiment
Description
VMWare Process Kill experiment kills target processes in a Linux-based VMware VM
Tunables
Necessary Inputs:
Process IDs
VM Name
Secrets Inputs:
vCenter Server URL
vCenter Username
vCenter Password
VM Username
VM Password
Service Kill Experiment
Description
VMware Service Kill experiment kills given services running on a Linux VM for a specified chaos interval before restarting it. It also validates the self-healing behaviour of the services, if specified.
Tunables
Necessary Inputs:
Service Names
VM Name
Tunable Inputs:
Self healing Service (enable /disable)
Secrets Inputs:
vCenter Server URL
vCenter Username
vCenter Password
VM Username
VM Password
Host Reboot Experiment
Description
VMWare Host Reboot experiment causes a target VMWare ESX Host to reboot. It can also validate the behaviour of hosts belonging to HA clusters.
Tunables
Necessary Inputs:
Host Name
Host Datacenter
Tunable Inputs:
High Availability Cluster (enable /disable)
Secrets Inputs:
vCenter Server URL
vCenter Username
vCenter Password
HTTP Chaos Experiment
Description
VMware HTTP Chaos experiment will induce HTTP chaos into the VMWare VM(s) for the specified duration. It uses toxiproxy to inject HTTP chaos.
Tunables
Necessary Inputs:
Toxicity
Latency
Request Timeout
Tunable Inputs:
Stream Port
Stream Type
Listen Port
Stream URL
Listen URL
HTTP Chaos Type
Operating System
Rate Limit
Data Limit
Secrets Inputs:
vCenter Server URL
vCenter Username
vCenter Password
VM Username
VM Password