As some of you will have seen this morning on Twitter, the issue preventing a March 15th launch was tracked down overnight and root cause identified. This is clearly great news and the problem is not related to code or bugs.
- A network config issue occurred with deployment of the pre-mainnet, it caused a fork on startup and unexpected node behaviour
- This has been identified and fixed and the network auto recovered
- We now reset to the beginning of the sequence in the original post on this thread and there are a small number of steps from here to launch, but they take several hours to work through slowly and methodically
- If we have smooth process from here on, we expect very good news later today and will confirm as soon as we have the information concretely known
I have caught up with some of the team now and the summary version is:
- We had to add some additional security configuration to avoid account holders connecting to something that looked like Mainnet but wasn’t Mainnet, to avoid confusion and potential phishing activities and well, basically because it wasn’t yet launched or ready
- A problem arose from that config which meant the network had a stuttering start and some nodes forked early which we didn’t expect to had so understand why to be confident with the launch sequence for Mainnet and that we didn’t have an issue. We now are.
- The fantastic news from a stability perspective is that when the config was rectified, the chain came back together on its own as designed, which means it auto recovered from the fork on its own
In terms of next steps, this happened on the first point of my original post, the sequence from here is therefore exactly the same, but is starting earlier in the day (UTC).
We are in the process of resetting the pre-mainnet (fake mainnet) so the final automation run can happen.
If the remaining steps run smoothly, we anticipate very good news later today, but there are still several hours work ahead of the team befor that can be confirmed