Dataplane reconcile fixes; LB counters cleanup; SPA scope cookie

Checker / reload:
- Reload's update-in-place branch now mirrors b.Address onto the
  runtime health.Backend. Without this, GetBackend kept returning
  the pre-reload address indefinitely after a config edit that
  touched addresses but not healthcheck settings — the VPP sync
  path reads cfg.Backends directly so the dataplane moved on
  while the gRPC and SPA view stayed wedged on the old IPv4/IPv6.

Sync (internal/vpp/lbsync.go):
- reconcileVIP now detects encap mismatch in addition to
  src-ip-sticky mismatch and takes the full tear-down / re-add
  path via a new shared recreateVIP helper. Triggered when every
  backend flips address family (gre4 <-> gre6) and the existing
  VIP can no longer accept new ASes — previously the sync wedged
  with 'Invalid address family' until a full maglevd restart.
- setASWeight is issued whenever the state machine requests
  flush (a.Flush=true), not only on the weight-value transition
  edge. Fixes the case where a backend reached StateDisabled
  after its effective weight had already been drained to 0 by
  pool failover — the sticky-cache entries pointing at it were
  previously never cleared.

maglev-frontend:
- signal.Ignore(SIGHUP) so a controlling-terminal disconnect
  doesn't kill the daemon.
- debian/vpp-maglev.service grants CAP_SYS_ADMIN in addition to
  CAP_NET_RAW so setns(CLONE_NEWNET) can join the healthcheck
  netns. Comment documents the 'operation not permitted' symptom
  and notes the knob can be dropped if the deployment doesn't use
  the 'netns:' healthcheck option.

LB plugin counters (internal/vpp/lbstats.go + friends):
- Fix the VIP counter regex: the LB plugin registers
  vlib_simple_counter_main_t names without a leading '/'
  (vlib_validate_simple_counter in counter.c:50 uses cm->name
  verbatim; only entries that set cm->stat_segment_name get a
  slash). first/next/untracked/no-server now read through as
  live values instead of zero.
- Drop the per-backend FIB counter block end-to-end (proto,
  grpcapi, metrics, vpp.Client, lbstats, maglevc). Traced from
  lb/node.c:558 into ip{4,6}_forward.h:141 — the LB plugin
  forwards by writing adj_index[VLIB_TX] directly and bypassing
  ip{4,6}_lookup_inline, which is the only path that increments
  lbm_to_counters. The backend's FIB load_balance stats_index
  literally never ticks for LB-forwarded traffic, so the column
  was always zero and misleading. docs/implementation/TODO
  records the full investigation and the recommended upstream
  path (new lb_as_stats_dump API message) for when we're ready
  to carry that VPP patch.
- maglevc show vpp lb counters: plain-text tabular headers.
  label() wraps strings in ANSI escapes (~11 bytes of overhead),
  but tabwriter counts bytes, not rendered width — so a header
  row with label()'d cells and data rows with plain cells drifts
  column alignment on every row. color.go comment now spells
  out the constraint: label() only works when column N is
  wrapped identically in every row (key-value layouts are fine,
  multi-column tables with header-only labelling are not).

SPA:
- stores/scope.ts is cookie-backed (maglev_scope, 1 year,
  SameSite=Lax). App.tsx hydrates from the cookie then validates
  against the fetched snapshots: a cookie referencing a maglevd
  that no longer exists falls through to snaps[0] instead of
  leaving the user on a ghost selection.
- components/Flash.tsx wraps props.value in createMemo. Solid's
  on() fires its callback on every dep notification, not on
  value change — source is right in solid-js/dist/solid.js:460,
  no equality check. Without the memo, flipping scope between
  two 'connected' maglevds (or any other cross-store reactive
  re-eval that doesn't actually change the concrete string)
  replays the animation every time. createMemo's default ===
  dedupe fixes it in one place for every Flash consumer,
  superseding the local createMemo workaround we'd added in
  BackendRow earlier.
This commit is contained in:
2026-04-14 14:39:52 +02:00
parent 4288e22b71
commit 224167ce39
20 changed files with 435 additions and 471 deletions

View File

@@ -94,6 +94,12 @@ func run() error {
ReadHeaderTimeout: 10 * time.Second,
}
// Ignore SIGHUP so a controlling-terminal disconnect (or any
// stray process-group SIGHUP) doesn't kill the daemon — the
// default Go handler terminates the process with "Hangup",
// which is the wrong behaviour for a long-running network
// service. SIGTERM / SIGINT remain the clean-shutdown signals.
signal.Ignore(syscall.SIGHUP)
sigCh := make(chan os.Signal, 1)
signal.Notify(sigCh, syscall.SIGTERM, syscall.SIGINT)

File diff suppressed because one or more lines are too long

File diff suppressed because one or more lines are too long

View File

@@ -5,7 +5,7 @@
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
<link rel="icon" type="image/x-icon" href="/view/favicon.ico" />
<title>maglev</title>
<script type="module" crossorigin src="/view/assets/index-DCJJqBMY.js"></script>
<script type="module" crossorigin src="/view/assets/index-3m4Pjc8_.js"></script>
<link rel="stylesheet" crossorigin href="/view/assets/index-3BvNJ7QB.css">
</head>
<body>

View File

@@ -19,7 +19,14 @@ const App: Component = () => {
const [snaps, ver] = await Promise.all([fetchAllState(), fetchVersion()]);
replaceAll(snaps);
setVersion(ver);
if (!scope() && snaps.length > 0) {
// Hydrate the scope: prefer a cookie-loaded value, but only if it
// still matches a maglevd we actually got back in the snapshot.
// Otherwise fall back to the first server in the list so the user
// never sees a "ghost selection" pointing at a maglevd that was
// removed or renamed since their last visit.
const current = scope();
const valid = current && snaps.some((s) => s.maglevd.name === current);
if (!valid && snaps.length > 0) {
setScope(snaps[0].maglevd.name);
}
openEventStream();

View File

@@ -1,4 +1,4 @@
import { createEffect, on, type Component, type JSX } from "solid-js";
import { createEffect, createMemo, on, type Component, type JSX } from "solid-js";
type Props = {
// value is used solely for change detection. When it changes the
@@ -26,9 +26,26 @@ type Props = {
const Flash: Component<Props> = (props) => {
let el: HTMLSpanElement | undefined;
// Solid's on() fires its callback whenever a tracked dep *notifies*,
// not whenever the tracked value actually changes — the source is
// right there in dist/solid.js:460, there's no equality check, just
// `fn(input, prevInput, prevValue)` on every notification. That's a
// problem for Flash because props.value is typically a reactive
// expression chained through the store (e.g. a per-maglevd
// vpp_state read), and changing an upstream signal like scope()
// causes every downstream reactive to re-evaluate even when the
// concrete value it produces is identical. Without this memo, the
// badge flashes every time you flip from one maglevd to another
// where both are 'connected'.
//
// createMemo gives us the equality dedupe we need: its default
// equality is ===, so an upstream notification that doesn't change
// the memoized output never propagates to the effect below.
const value = createMemo(() => props.value);
createEffect(
on(
() => props.value,
value,
() => {
el?.animate(
[

View File

@@ -1,6 +1,48 @@
import { createSignal } from "solid-js";
// The currently selected maglevd name, or undefined before first fetch.
const [scope, setScope] = createSignal<string | undefined>(undefined);
// Persisted selection of which maglevd the SPA is currently scoped to.
// The cookie is a best-effort hint: if it's missing, corrupt, or names a
// maglevd that no longer exists, we fall back to whatever App.tsx's
// hydration path picks (typically the first server in byName order).
// SameSite=Lax keeps it from leaking to third-party iframes; Max-Age is
// a year so the selection survives browser restarts.
const COOKIE_NAME = "maglev_scope";
const COOKIE_MAX_AGE = 60 * 60 * 24 * 365;
function readCookie(): string | undefined {
try {
const raw = document.cookie.split("; ").find((c) => c.startsWith(COOKIE_NAME + "="));
if (!raw) return undefined;
const value = decodeURIComponent(raw.slice(COOKIE_NAME.length + 1));
return value || undefined;
} catch {
return undefined;
}
}
function writeCookie(name: string | undefined) {
try {
if (!name) {
// Clear by setting Max-Age=0. Works across every mainstream browser.
document.cookie = `${COOKIE_NAME}=; Path=/; Max-Age=0; SameSite=Lax`;
return;
}
const value = encodeURIComponent(name);
document.cookie = `${COOKIE_NAME}=${value}; Path=/; Max-Age=${COOKIE_MAX_AGE}; SameSite=Lax`;
} catch {
// quota / third-party-cookie blocks / private window — best effort
}
}
const [scope, setScopeRaw] = createSignal<string | undefined>(readCookie());
// setScope wraps the raw signal setter so every selection change writes
// back to the cookie. Callers use this exactly like the old setScope —
// the cookie plumbing is invisible at the call site.
function setScope(name: string | undefined) {
setScopeRaw(name);
writeCookie(name);
}
export { scope, setScope };

View File

@@ -377,12 +377,7 @@ export function applyConfiguredWeight(
// worse — typical maintenance / outage state
// "unknown" → fallthrough; should be unreachable, kept as
// a safety net for logic bugs in this function
export type FrontendHealth =
| "ok"
| "bug-buckets"
| "primary-drained"
| "degraded"
| "unknown";
export type FrontendHealth = "ok" | "bug-buckets" | "primary-drained" | "degraded" | "unknown";
export function frontendHealth(snap: StateSnapshot, fe: FrontendSnapshot): FrontendHealth {
const stateOf: Record<string, string> = {};

View File

@@ -17,9 +17,7 @@ const COOKIE_MAX_AGE = 60 * 60 * 24 * 365; // 1 year
function readCookie(): Set<string> {
try {
const raw = document.cookie
.split("; ")
.find((c) => c.startsWith(COOKIE_NAME + "="));
const raw = document.cookie.split("; ").find((c) => c.startsWith(COOKIE_NAME + "="));
if (!raw) return new Set();
const value = decodeURIComponent(raw.slice(COOKIE_NAME.length + 1));
if (!value) return new Set();