jakman85
a819c50ffe
roles/etcd_io_priority + playbooks/etcd-io-priority.yml, enabled for [k3s_prod]
only. Mitigation for the 2026-09-07 k3s-prod-1 outage: etcd's WAL fdatasync took
3.4-6.7 s (expects 1 s), the node went NotReady while the host stayed up, and
/proc/pressure/io full avg300 read 11.36 there against 0.75 (node 2, NVMe) and
1.29 (node 3).
WHAT IT DOES, on the SATA nodes only:
1. k3s-etcd-io-priority.service (oneshot, Before=k3s.service) switches the
shared device to BFQ and drops nr_requests 64 -> 32.
2. IOWeight=10000 drop-ins on system.slice AND k3s.service, which BFQ turns
into io.bfq.weight=1000 — a 10:1 share against the default-weighted
kubepods.slice where Longhorn's instance-managers live.
3. Reads all of it back out of the kernel and ASSERTS. If the scheduler did
not switch or the weight did not land, the play fails.
THE THREE OBVIOUS ANSWERS ARE INERT ON THESE NODES, ALL CHECKED LIVE, ALL
DELIBERATELY ABSENT:
* IOSchedulingClass=realtime on k3s.service. mq-deadline DOES honour the
priority class, so this looks right — but ioprio is inherited across fork
and kept across exec (block/blk-ioc.c __copy_io copies ioc->ioprio) and
CONTAINERD IS A CHILD OF THE K3S SERVER PROCESS. Verified on k3s-prod-1:
containerd pid 84553 has ppid 84333 = k3s-server, and
ionice -c2 -n0 -- sh -c 'ionice -p $$; sh -c "ionice -p \$\$"'
prints best-effort prio 0 at both levels. Realtime for k3s is realtime for
every container on the node, Longhorn included. No relative gain.
* IOSchedulingPriority (the level inside a class): mq-deadline buckets by
CLASS only, so BE-0 vs BE-4 is a no-op.
* io.latency, which would be the ideal scheduler-independent protection:
CONFIG_BLK_CGROUP_IOLATENCY is not built in Debian trixie's 7.1.8 kernel
(only IOCOST is) and no io.latency file exists under /sys/fs/cgroup.
AND ONE NON-OBVIOUS TRAP: bare IOWeight= is DECORATION here. systemd writes it
to io.weight (blk-iocost — compiled in but disabled, io.cost.qos is empty) and
io.bfq.weight (BFQ — not loaded). Neither reader exists on the stock nodes, so
the scheduler change and the weights have to be one role with one verification.
The value is 10000 rather than a "reasonable" 1000 for the same reason: systemd
rescales 1..10000 onto io.bfq.weight's 1..1000, so 1000 maps to ~101 against
BFQ's unweighted default of 100. The role asserts the result is >= 500.
k3s-prod-2 IS SKIPPED ON PURPOSE. The role discovers the device under
k3s_data_dir per node (verified: prod-1/3 -> /dev/sda5 -> sda, prod-2 ->
/dev/nvme0n1p5 -> nvme0n1) and writes NOTHING on NVMe rather than leaving inert
files behind — BFQ is a poor fit for NVMe and node 2 measured 0.75 IO PSI.
DEV IS UNTOUCHED. Master switch defaults false and is set true only in
group_vars/k3s_prod; dev has a dedicated etcd device (docs/plans/ssd-etcd-
isolation.md) and nothing to arbitrate. Verified with --list-hosts: no extra
vars -> the three dev hosts (where the role is a no-op); -e k3s_hosts=k3s_prod
-> the three prod hosts; --limit k3s-prod-3 -> one.
NOTHING IS RESTARTED. The oneshot writes two sysfs files; the weights go in
through daemon-reload. If daemon-reload turns out not to re-realize a running
k3s.service's cgroup, the verification FAILS and names the rolling restart as
the fix rather than quietly leaving the weight unapplied.
WHAT IT CANNOT DO, stated in the role README and worth repeating: it does not
separate the two tenants. Both still write one ext4 filesystem, and no cgroup
control arbitrates the shared jbd2 journal that etcd's fdatasync waits on. It is
also UNPROVEN on this hardware — it fails loudly if the settings do not land, it
cannot tell you they helped. Apply to k3s-prod-3 first and watch
NodeIOPressureHigh. Replacing the C400 / giving etcd its own device remains the
real fix and the operator's (PROD-PROVISIONING.md §2 exit condition).
Syntax-checked, and the fact-derivation and assert expressions were exercised
offline against all three node shapes plus negative controls (a node still on
mq-deadline fails the gate; IOWeight=1000 -> 101 fails the floor).
Co-Authored-By: Claude Opus 5 <[email protected]>
(cherry picked from commit de8c1f832724c011b03466842e18296de07a6776)
infra/ — node provisioning and the toolchain
OpenTofu (Proxmox VE) + cloud-init + Ansible bring up the 3-node k3s/Cilium dev cluster.
Spec: ../home-network-k3s/docs/02-environments-and-provisioning.md.
| Path | What |
|---|---|
tofu/ |
The "Create" stage: 3 cloud-init'd VMs on the Proxmox box (bpg/proxmox provider) |
cloud-init/ |
First-boot baseline shipped as a Proxmox user-data snippet |
ansible/ |
The "Baseline"/"Configure" stages: k3s HA, Cilium, kube-vip, the VPN egress gateway |
devcontainer/ |
The pinned toolchain (tofu, ansible, sops, age, flux, kubectl, helm and the whole CI gate toolchain) |
cluster.sh |
RETIRED — drove the old libvirt VMs; does not work against Proxmox |
PHASE1-STATUS.md |
Current state of the cluster and how to reconnect to it |
PHASE1-SCOPE.md |
Historical Phase-1 scoping doc (libvirt era — superseded) |
Where the cluster actually lives
The dev cluster is not on this workstation. It runs on a Proxmox VE host; the nodes sit on the Proxmox LAN and are reached across a pfSense NAT:
| Nodes | k3s-dev-1/2/3 at 192.168.1.50/.51/.52 — 6 vCPU, 24 GiB, Debian 13, k3s v1.36.4 |
| kube-API | kube-vip VIP 192.168.1.49:6443, reached from the workstation via NAT 192.168.2.50:6443 |
| Node SSH | pfSense NAT 192.168.2.102:2250/2251/2252 → node :22 (Ansible only) |
| Ingress VIP | 192.168.1.53 (Cilium L2, shared Gateway); .53–.60 reserved for LB VIPs |
The earlier libvirt/KVM cluster on 10.73.0.0/24 is retired. Any doc or command referencing
virsh, qemu:///system, or 10.73.0.x for the cluster is historical.