Michael MacDonald

> Seems like a reasonable change, what about the output of `dmg system query`? does the "reason" field in the table still get updated when the rank terminates? requesting changes...

Is this patch necessary for 2.4? Given its size and complexity, it seems somewhat risky. If I understand correctly, the main goal is to correctly avoid failures in bdev scan...

@tanabarr: I'm really having a hard time with the global variables... I think it's asking for a lot of trouble with flaky tests, particularly down the road when other people...

@daos-stack/daos-gatekeeper: This PR is a small change to a SConscript file, and it has passed the build stages multiple times successfully. Not sure what's going on with the functional/NLT tests...

> > > If I understand the ticket and the PR changes correctly, this patch aims to reduce CI instability by reserving a hard-coded maximum amount of hugepages at startup,...

Runs 3 & 4 both hit a timeout in the offline reintegration tests, which appears to be tracked in [DAOS-15608](https://daosio.atlassian.net/browse/DAOS-15608). I don't think it's related to any changes made in...

Test failure is [DAOS-15608](https://daosio.atlassian.net/browse/DAOS-15608).

> Should the last merge commit have ran with `Features: control`? ~~Can't~~ ~~Shouldn't~~ ~~Probably~~ Hopefully(?) won't hurt... /me rolls the dice

Ping reviewers... @mchaarawi, @kjacque, @liw, @kccain Looks like metadata-owners wasn't tagged automatically.

The failure on the previous run appears to be an instance of [DAOS-15608](https://daosio.atlassian.net/browse/DAOS-15608). I restarted it in hopes of getting a clean run.