John Spray

I think bootstrap should just refuse to run if the file is already populated -- if the user gets here they're probably confused (they already have a config file, why...

@andijcr can you merge https://github.com/redpanda-data/redpanda/pull/5315 into this branch as a starting point for tests? I just rebased that PR but I expect it to fail without the changes in this...

(didn't review the code but I notice the commit messages aren't in the usual format: see CONTRIBUTING.md in the source and/or take a peek at the commit history)

Updated title+description because while this was originally noticed on 22.1.x, it is actually present on dev as well (seen on 2022-10-6)

LGTM with the exception of the comment in partition_manifest::difference (https://github.com/redpanda-data/redpanda/pull/6387#discussion_r992318028) Needs a rebase for the overlapping changes to partition_manifest_test.cc from my timequery PR.

This fails to build

Force push: this couldn't merge because it had a merge conflict with `dev`, so while rebasing I took the opportunity to: - pull in the commit that fixes read replicas...

Push: add a commit to fix unit tests that were broken by addition of insync_offset to manifest.

> Any feedback on writing a useful non-flaky test would be appreciated Same question has come up recently for other things: `manual_clock` is probably the best way to write a...

> I've added a new unit test, however it would be nice to have a ducktape test, although i'm not sure if we could concretely control and calculate expected delay...