reachableceo/ultix
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

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

Browse Source 
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:
mrcharles
2026-08-31 16:55:38 -05:00
parent cba2a64d67
commit acd7242604
5 changed files with 35 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
5-activate-multiqueue.sh
Executable
+10
View File
@@ -0,0 +1,10 @@
#!/usr/bin/env bash
# Activate the staged net multiqueue (queues=4/2 already in qm config) with
# ONE graceful bounce of VM 5111. Run any time from this VM; the host-side
# work is detached, so this script returns before the VM goes down.
# WARNING: every crush session on this VM dies with the bounce (~4 min).
set -euo pipefail
echo "bouncing VM 5111 (detached host-side); ~4 min downtime"
timeout 60 ssh root@pfv-tsys5.knel.net \
'setsid bash -c "echo bounce-start \$(date -Is); qm shutdown 5111 --timeout 120; qm wait 5111 --timeout 180 || true; qm start 5111; echo bounce-done \$(date -Is)" </dev/null >/var/log/ukrrs-vm5111-bounce.log 2>&1 &'
echo "dispatched. after reboot verify: ethtool -l ens18 -> Combined 4/4, ens19 -> 2/2"