[FEATURE] High availability #2587
Description
Problem to solve
Currently operating one sequencer is performant, but when you go to upgrade the node and/or it goes down the sequencer then it will cause downtime. This makes have a high availability system a requirement for mainnets.
Proposed solution or feature
The OP stack has a piece of software, conductor. This approach seems sufficient and compatible with our needs. We should implement our own solution for ev-node.
Alternatives considered
No response
Additional context
When using a system like this in the simplest case there is only a small increase in latency, but in the worst case, a node goes offline then there will be a latency increase for a short period of time