You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, the only way to update which ENIs are monitored in a particular user-VPC is invoke remove-vpc & add-vpc to tear down capture and stand it back up, a process with currently takes ~10 minutes. This is obviously not the desired experience.
This task is to investigate options for more nimble, ideally automatic, creation and destruction of per-ENI mirroring configuration and propose a initial solution to the problem. The proposal should ideally address the following topics:
Automated creation/destruction of per-ENI configuration in each user-VPC
A way for users to initiate this process themselves if they wish
How the system will behave for longer-lived (EC2 instances) and shorter lived (Lambda containers) resources
How the system will handle multiple, concurrent and/or conflicting instructions
This task will NOT address changes in the network topography of the user's VPC (e.g. adding/removing subnets); that will be addressed with the refresh-cluster command (see: #32)
Acceptance Criteria
Proposal design, written up in GitHub
The text was updated successfully, but these errors were encountered:
Taking a look at things, AWS Event Bridge [1] seems like a good option to explore further. I'm thinking we'll have one or more Lambda functions that are capable of setting up/tearing down the per-ENI resources. We'll probably schedule them to fire on a regular cadence as a backstop, and then use Event Bridge to catch things like EC2 ASG and ECS scale up/down events and react sooner than backstop would fire. LOTS of AWS Services fire these events [2].
Description
Currently, the only way to update which ENIs are monitored in a particular user-VPC is invoke
remove-vpc
&add-vpc
to tear down capture and stand it back up, a process with currently takes ~10 minutes. This is obviously not the desired experience.This task is to investigate options for more nimble, ideally automatic, creation and destruction of per-ENI mirroring configuration and propose a initial solution to the problem. The proposal should ideally address the following topics:
This task will NOT address changes in the network topography of the user's VPC (e.g. adding/removing subnets); that will be addressed with the
refresh-cluster
command (see: #32)Acceptance Criteria
The text was updated successfully, but these errors were encountered: