fix(ceph): align the spice mclock model with the live cluster (#411)

This commit is contained in:
Andy Molenda
2026-08-03 15:13:09 +00:00
committed by GitHub
parent 5d60aae2fd
commit caf120d8e5
@@ -51,22 +51,22 @@ clusters = {
# upgrade or the deadlock returns silently.
# https://www.mail-archive.com/ceph-users@ceph.io/msg28007.html
#
# osd_scrub_cost, osd_deep_scrub_stride and the custom mclock profile are
# provisional: measured gains were inside noise, under conditions that no
# longer apply. Drop them once the cluster is quiet enough to A/B properly,
# the profile first (it makes us owner of the client/recovery reservations).
# osd_scrub_cost and osd_deep_scrub_stride are provisional: measured gains
# were inside noise, under conditions that no longer apply. Drop them once
# the cluster is quiet enough to A/B properly. The custom mclock profile
# from the same incident is already gone: the cluster runs `balanced` (set
# live; the old model still said custom and would have re-imposed it on
# converge). balanced owns the scheduler reservations, so no _res overrides.
ceph_config = {
global = {
osd_deep_scrub_interval = "2419200" # 28 days
}
osd = {
osd_scrub_disable_reservation_queuing = "true"
osd_max_scrubs = "6"
osd_scrub_cost = "50" # provisional
osd_deep_scrub_stride = "2097152" # provisional, 2 MiB
osd_mclock_profile = "custom" # provisional
osd_mclock_scheduler_background_recovery_res = "0.2"
osd_mclock_scheduler_background_best_effort_res = "0.3"
osd_scrub_disable_reservation_queuing = "true"
osd_max_scrubs = "6"
osd_scrub_cost = "50" # provisional
osd_deep_scrub_stride = "2097152" # provisional, 2 MiB
osd_mclock_profile = "balanced"
}
}
hosts = [