A node being lent says when a replica with nowhere to go is stopped
CI / test (push) Successful in 9m37s

Tried on the PoC: lending cuda01, the one GPU node, its model had nowhere
else to go and waited out the drain's five-minute grace, while ctl slurm
status said only "1 replica(s) still to move off". It now says the time
that replica is stopped at. Also recorded: Slurm on a GPU node, tried on
cuda01 (gres, ConstrainDevices, accounting of gres/gpu).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Ilker ManapandClaude Opus 5.5 committed 2026-11-14 20:45:57 +01:00
1 parent 4b4306bb6a
commit c3a75dcf3f
4 files changed
+56 -5

No files matched your search

+5
View File
@@ -12336,6 +12336,11 @@
"replicas": {
"type": "integer"
},
"stops_by": {
"type": "string",
"format": "date-time",
"description": "StopsBy is, while it is being lent, when a replica still on it is stopped even with nowhere else to go: the drain's grace after the lend began. Zero when not known (a leader that took over meanwhile)."
},
"slurm": {
"$ref": "#/components/schemas/slurmsvc.Reported"
},