Methods for managing self-healing cloud infrastructure and applications

A method to manage cloud applications, the method comprising:

• determining current states of virtual machines;
• determining if a virtual machine has been in an operational state of “init” or “ready” state for longer than a threshold amount of time;
• in response to determining that the virtual machine has been in the “init” or “ready” state for longer than the threshold amount of time, setting the virtual machine to a “fail” state;
• comparing the current states of the virtual machines to target states to determine a difference between the number of operational virtual machines and the number of target virtual machines requested in the target states
• in response to determining that the difference indicates that there are not enough healthy virtual machines satisfying a health check to meet the target states, adding virtual machines based on the difference;
• and
• in response to determining that the difference indicates that more healthy virtual machines are active than the number of target virtual machines requested in the target states, instruct the cloud manager to remove an additional virtual machine based on the difference and a health-based ranking of operational virtual machines.

A non-transitory computer readable storage medium comprising instructions that, when executed, cause a machine to at least:

• determine current states of virtual machines;
• communicate with a hybrid cloud infrastructure;
• determine if a virtual machine has been in an operational state of “init” or “ready” state for longer than a threshold amount of time;
• in response to determining that the virtual machine has been in the “init” or “ready” state for longer than the threshold amount of time, set the virtual machine to a “fail” state;
• compare the current states of the virtual machines to target states to determine a difference between the number of operational virtual machines and the number of target virtual machines requested in the target states;
• and
• in response to determining that the difference indicates that there are not enough healthy virtual machines to meet the target states, add virtual machines based on the difference;
• and
• in response to determining that the difference indicates that there are more healthy virtual machines than the number of target virtual machines requested in the target states, instruct the cloud manager to remove an additional virtual machine based on the difference and a health-based ranking of operational virtual machines.