plan for October hardware slip: presume 8 vCPU/48G, multiqueue bounce script

Hardware window #601 slipped to ~Oct 2026: interim fences and daynight
values become the operating config, cpuset reservation defers with it, and
the compose cgroup_parent wiring becomes the gateway starvation fix. Adds
5-activate-multiqueue.sh — the one remaining VM-5111 reboot (human-timed)
to bring the staged net queues live. Tracking/README/crush prefs updated.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2
[#602]
This commit is contained in:
2026-08-31 16:55:38 -05:00
parent cba2a64d67
commit acd7242604
5 changed files with 35 additions and 8 deletions
+13 -4
View File
@@ -5,8 +5,10 @@
- Host flags live: ssd=1/discard (ROTA=0), iothread, onboot=1, startup
order=10, cpuunits 9000/50, ide2 gone.
- Net multiqueue: queues=4/2 now IN live qm config (staged 16:40 by crush
via ssh; activates at next VM start = Friday). Guest oneshot unit
ukrrs-net-multiqueue auto-runs ethtool -L on that boot.
via ssh); activates at the next VM start — ONE reboot still owed, human-
timed: run ~/optimize/5-activate-multiqueue.sh whenever convenient
(kills crush sessions, ~4 min). Guest oneshot ukrrs-net-multiqueue
auto-runs ethtool -L on that boot.
- Gateway boot-race + live-restore endpoint loss: permanently fixed by
ukrrs-gateway-ensure unit (enabled, both lanes).
- Root cause of the 15:59 no-op: the script copy executed then predates the
@@ -34,5 +36,12 @@
- mopac-demo/mcli fake containers: restart=no, exited at the first bounce,
left down (test stubs; your call).
- Prometheus/harness daemons not deployed here yet (that is OPT-9's wiring).
- Friday unchanged: CPU/RAM swap (plan NFS outage for -02), 2nd USB3 card,
q35 + 20 vCPU/128G reshape, GPUs. Enable the ensure unit BEFORE Friday.
## HARDWARE SLIPPED TO ~OCT 2026 (ruling 2026-08-31)
- PRESUME 8 vCPU / 48G on this VM until the #601 window: interim fences
(accounts 3G/4.5G) and the daynight interim values ARE the operating
config; the commented post-upgrade block stays dormant.
- Gateway cpuset hard reservation (16-17) waits for the window → the
compose cgroup_parent wiring is THE starvation fix meanwhile (ticketed).
- q35 / 20 vCPU / 128G / GPUs / USB3 reseat / NFS-outage-for--02 all move
to the same October window (#601, #606).