[08:42:28] <jayme>	 heads up: I've applied a change that will trigger image garbage collection on worker nodes earlier then usual (at 80% imagefs disk usage rather than 85%). See https://phabricator.wikimedia.org/T437539 for context.
[10:25:40] <brouberol>	 nice! Is that a fleet-wide change (as in across clusters)?
[10:48:28] <btullis>	 Hello. I would be interested in your collective take on https://phabricator.wikimedia.org/T430658#12337206 if anyone has a chance to have a look. This is about the required re-numbering of dse-k8s-eqiad. Thanks.
[11:01:36] <jayme>	 brouberol: it's fleet-wide, yes
[11:01:54] <brouberol>	 ack thanks
[11:02:09] <jayme>	 I've also opened https://phabricator.wikimedia.org/T438465 for the related issue regarding ephemeral storage (calculation and evictions)
[11:31:24] <jayme>	 btullis: nice writeup, thanks! I wonder: Did you find anything that would prevent us to change the detectLocal mode fleet wide? I'd love to keep the basic cluster config somewhat identical across the fleet
[11:31:40] <jayme>	 and it sounds as if switching to InterfaceNamePrefix only has benefits
[12:05:25] <btullis>	 jayme: No, I haven't found anything that would prevent us from doing that fleet-wide, and I agree that it would be good to migrate them all, if it works as expected.
[12:05:54] <jayme>	 ack
[12:07:26] <jayme>	 nice that the possible options for localmode are not even documented 🤦
[12:11:04] <jayme>	 oh, you even wrote about the other clusters in your last sentence...sorry
[13:00:03] <cdanis>	 btullis: nice writeup :)
[13:04:32] <btullis>	 Thank you :-)