There is very little info level logs during replication startup sequence and reconfigs, which makes it difficult to triage customer issues. There should be some info logs before some major steps in the startup sequence for example when it starts to do recoverFromOplog, or change some import states.