On Thu, Apr 5, 2012 at 1:17 PM, Ori Mamluk <omam...@zerto.com> wrote:
This series looks like patches on top of your last series. Please consolidate them into logical commits that build up the feature rather than a series that shows the editing steps you took when writing this code. This makes it easier to review because there will be no code churn (code that gets introduced in one patch and then deleted a few patches later). We should see the final picture as much as possible, not the intermediate steps. > Points and open issues: I think security hasn't been mentioned. At the moment there's no way to authenticate the client or server. That means a malicious client may be able to connect to the server and replicate junk data to destroy existing replicas. Also, guest I/O buffers will be transferred unencrypted over the network connection. What's the plan here? At least the protocol should include extensibility so that authentication and/or encryption can be added later. Stefan
