[16:11:17] 10netops, 10Operations, 10ops-eqdfw: eqdfw: Patch GTT cross-connect - https://phabricator.wikimedia.org/T194515#4200613 (10ayounsi) p:05Triage>03Normal [16:56:08] bblack, ema, about T187962, there are 12 cp servers that will need to move racks, is that something we can do ahead of time? And would they benefit from beeing distributed (instead of all in one rack like currently)? [16:56:09] T187962: Rack/cable/configure asw2-c-eqiad switch stack - https://phabricator.wikimedia.org/T187962 [16:56:28] list is https://www.irccloud.com/pastebin/6D8L1XyV/ [16:56:38] also is there any that is due to be decomissioned? [17:11:11] 10netops, 10Operations, 10ops-eqiad, 10Patch-For-Review: Rack/cable/configure asw2-c-eqiad switch stack - https://phabricator.wikimedia.org/T187962#4200809 (10ayounsi) [17:15:53] 10netops, 10Operations, 10ops-eqdfw: eqdfw: Patch GTT cross-connect - https://phabricator.wikimedia.org/T194515#4200820 (10ayounsi) [17:34:26] XioNoX: those are all due for replacement "sometimes this Q", probably fairly late [17:35:01] do they really need to move racks, or is this just to minimize temporary downtime? [17:35:22] (because we can depool all the edge nodes at the site for a downtime) [17:35:53] (although, eqiad is a bit trickier, but I guess we can do it like a edge-only core dc failover and use codfw to reach applayer) [17:36:01] they do need to move, rack C8 is currently the only 10g rack, but new layout has C2/4/7 as 10g [17:39:46] bblack: Actually, keeping them on the old switch stack (and rack) might be the best option if they get decommed before end of June [17:40:25] I'd avoid leaving too many servers behind, but those seem to be good candidates [18:27:55] 10netops, 10Operations, 10ops-eqiad, 10Patch-For-Review: Rack/cable/configure asw2-c-eqiad switch stack - https://phabricator.wikimedia.org/T187962#4200965 (10chasemp) >>! In T187962#4198886, @Andrew wrote: > I'm flying on the 29th. If Chase wants to manage these things without me that's fine with me thou...