Compare commits

...
2 Commits
Author SHA1 Message Date
andreas a42d783c0c version 5.4.2
Release / release (push) Successful in 1m10s
2026-08-17 09:24:23 -04:00
andreasandClaude Sonnet 5 18e33656c0 fix: preserve chart history across reconnects, show gaps for missing data
Only wipe real (non-RTT) plugin data on an actual client reboot (boot
flag), not on every ordinary OVERDUE/DOWN -> UP recovery. A transient
network blip no longer erases CPU/memory/etc. history.

Also split the shared time-series chart into separate line/area segments
wherever the gap between samples is much larger than the typical spacing,
so missing data (host overdue, or history that simply hasn't accumulated
across a drop) renders as a visual gap instead of an interpolated line.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DEimzMv4Q5EjFg3hoiZ69T
2026-08-17 09:13:49 -04:00
8 changed files with 88 additions and 25 deletions
+7
View File
@@ -2,6 +2,13 @@
All notable changes to this project are documented here, organized by release. All notable changes to this project are documented here, organized by release.
## [5.4.2]
### Fixed
- preserve chart history across reconnects, show gaps for missing data
---
## [5.4.1] ## [5.4.1]
### Added ### Added
+1 -1
View File
@@ -20,7 +20,7 @@ A lightweight UDP-based host monitoring system. Monitored hosts run a client (`h
└────────────────────┘ └────────────────────────────┘ └────────────────────┘ └────────────────────────────┘
``` ```
**Package:** `hbd` v5.4.1 **Package:** `hbd` v5.4.2
**Python:** 3.11+ **Python:** 3.11+
### Subpackages ### Subpackages
+1 -1
View File
@@ -14,4 +14,4 @@ Install options:
""" """
__all__ = ["__version__"] __all__ = ["__version__"]
__version__ = "5.4.1" __version__ = "5.4.2"
+30 -8
View File
@@ -847,16 +847,38 @@
const yRange = yHigh - yLow || 1; const yRange = yHigh - yLow || 1;
const y = v => PAD.top + cH - ((v - yLow) / yRange) * cH; const y = v => PAD.top + cH - ((v - yLow) / yRange) * cH;
// Build polyline points and filled area path
const linePoints = pts.map(p => `${x(p.t).toFixed(1)},${y(p.v).toFixed(1)}`).join(' ');
const areaPath = `M${x(pts[0].t).toFixed(1)},${(PAD.top + cH).toFixed(1)} ` +
pts.map(p => `L${x(p.t).toFixed(1)},${y(p.v).toFixed(1)}`).join(' ') +
` L${x(pts[pts.length-1].t).toFixed(1)},${(PAD.top + cH).toFixed(1)} Z`;
// Color based on latest value // Color based on latest value
const latest = pts[pts.length - 1].v; const latest = pts[pts.length - 1].v;
const { stroke: strokeColor, fill: fillColor } = opts.colorFor(latest); const { stroke: strokeColor, fill: fillColor } = opts.colorFor(latest);
// Split into segments wherever the gap between consecutive samples is
// much larger than the typical spacing (e.g. the host was overdue/down
// for a while) — draw each segment separately so missing data reads as
// a visual gap instead of an interpolated line across dead time.
const deltas = [];
for (let i = 1; i < pts.length; i++) deltas.push(pts[i].t - pts[i - 1].t);
deltas.sort((a, b) => a - b);
const medianDelta = deltas[Math.floor(deltas.length / 2)];
const gapThreshold = medianDelta * 2.5;
const segments = [[pts[0]]];
for (let i = 1; i < pts.length; i++) {
if (pts[i].t - pts[i - 1].t > gapThreshold) segments.push([]);
segments[segments.length - 1].push(pts[i]);
}
let linePolylines = '';
let areaPaths = '';
for (const seg of segments) {
if (seg.length < 2) continue;
const segPoints = seg.map(p => `${x(p.t).toFixed(1)},${y(p.v).toFixed(1)}`).join(' ');
linePolylines += `<polyline points="${segPoints}" fill="none" stroke="${strokeColor}" stroke-width="1.5" stroke-linejoin="round"/>`;
const segArea = `M${x(seg[0].t).toFixed(1)},${(PAD.top + cH).toFixed(1)} ` +
seg.map(p => `L${x(p.t).toFixed(1)},${y(p.v).toFixed(1)}`).join(' ') +
` L${x(seg[seg.length-1].t).toFixed(1)},${(PAD.top + cH).toFixed(1)} Z`;
areaPaths += `<path d="${segArea}" fill="${fillColor}" opacity="0.6"/>`;
}
// Compute nice tick step for ~3-5 grid lines // Compute nice tick step for ~3-5 grid lines
const rawStep = yRange / 4; const rawStep = yRange / 4;
const mag = Math.pow(10, Math.floor(Math.log10(rawStep || 1))); const mag = Math.pow(10, Math.floor(Math.log10(rawStep || 1)));
@@ -890,8 +912,8 @@
<line x1="${PAD.left}" y1="${PAD.top}" x2="${PAD.left}" y2="${PAD.top + cH}" stroke="#ccc" stroke-width="1"/> <line x1="${PAD.left}" y1="${PAD.top}" x2="${PAD.left}" y2="${PAD.top + cH}" stroke="#ccc" stroke-width="1"/>
<line x1="${PAD.left}" y1="${PAD.top + cH}" x2="${PAD.left + cW}" y2="${PAD.top + cH}" stroke="#ccc" stroke-width="1"/> <line x1="${PAD.left}" y1="${PAD.top + cH}" x2="${PAD.left + cW}" y2="${PAD.top + cH}" stroke="#ccc" stroke-width="1"/>
<g clip-path="url(#${opts.clipId})"> <g clip-path="url(#${opts.clipId})">
<path d="${areaPath}" fill="${fillColor}" opacity="0.6"/> ${areaPaths}
<polyline points="${linePoints}" fill="none" stroke="${strokeColor}" stroke-width="1.5" stroke-linejoin="round"/> ${linePolylines}
</g> </g>
${xLabels} ${xLabels}
</svg>`; </svg>`;
+5 -1
View File
@@ -521,9 +521,13 @@ def handle_datagram(msg: dict, addr, transport, ctx: dict):
# Transition to UP and log/notify if appropriate # Transition to UP and log/notify if appropriate
lasts = conn.state lasts = conn.state
d = conn.newstate(hbdcls.Connection.UP, now) d = conn.newstate(hbdcls.Connection.UP, now)
if boot:
# On reboot, pre-boot plugin data and derived alerts are stale. # On reboot, pre-boot plugin data and derived alerts are stale.
# Cancel all plugin timers and wipe plugin state so timers restart # Cancel all plugin timers and wipe plugin state so timers restart
# cleanly from the first two post-boot samples. # cleanly from the first two post-boot samples. An ordinary
# reconnect (no boot flag) doesn't invalidate the client's
# already-collected data, so it's left alone — this keeps chart
# history intact across a transient network blip.
for pname in list(host.plugin_timers): for pname in list(host.plugin_timers):
host.cancel_plugin_timer(pname) host.cancel_plugin_timer(pname)
for pname in [k for k in host.plugin_data if not _is_rtt_key(k)]: for pname in [k for k in host.plugin_data if not _is_rtt_key(k)]:
+1 -1
View File
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
[project] [project]
name = "hbd" name = "hbd"
version = "5.4.1" version = "5.4.2"
description = "Heartbeat monitoring system — client (hbc) and server (hbd)" description = "Heartbeat monitoring system — client (hbc) and server (hbd)"
readme = "README.md" readme = "README.md"
requires-python = ">=3.11" requires-python = ">=3.11"
+1 -1
View File
@@ -41,7 +41,7 @@ from pathlib import Path
from typing import Any, Dict, List, Optional, Tuple from typing import Any, Dict, List, Optional, Tuple
# updated by scripts/bumpminor.sh # updated by scripts/bumpminor.sh
__version__ = "5.4.1" __version__ = "5.4.2"
# --------------------------------------------------------------------------- # ---------------------------------------------------------------------------
# Protocol (mirrors hbd/common/proto.py) # Protocol (mirrors hbd/common/proto.py)
+36 -6
View File
@@ -14,10 +14,12 @@ class _FakeTransport:
self.sent.append((data, addr)) self.sent.append((data, addr))
def _htb(name, rtt=None, interval=0): def _htb(name, rtt=None, interval=0, boot=0):
d = {"name": name, "interval": interval, "id": 0} d = {"name": name, "interval": interval, "id": 0}
if rtt is not None: if rtt is not None:
d["rtt"] = rtt d["rtt"] = rtt
if boot:
d["boot"] = boot
return parse_message(dicttos("HTB", d)) return parse_message(dicttos("HTB", d))
@@ -94,10 +96,11 @@ def test_request_update_fires_on_recovery_even_with_rtt_history():
assert ack.get("request_update") assert ack.get("request_update")
def test_recovery_clears_real_plugin_data_but_preserves_rtt_history(): def test_ordinary_recovery_preserves_real_plugin_data_and_rtt_history():
"""Regression for Finding 2: host.plugin_data.clear() on recovery must """An ordinary reconnect (no boot flag) — e.g. OVERDUE->UP after a
wipe real client-collected plugin data while leaving rtt_* history transient network blip — must NOT wipe already-collected real plugin
samples intact. data (e.g. cpu_monitor, os_info) or rtt_* history. Only an actual
client reboot invalidates that data (see the boot-flag test below).
""" """
hbdclass.Host.hosts.pop("rtt-hist-host5", None) hbdclass.Host.hosts.pop("rtt-hist-host5", None)
transport = _FakeTransport() transport = _FakeTransport()
@@ -116,8 +119,35 @@ def test_recovery_clears_real_plugin_data_but_preserves_rtt_history():
host.add_plugin_data("os_info", {"os": "linux"}, timestamp=time.time()) host.add_plugin_data("os_info", {"os": "linux"}, timestamp=time.time())
assert "os_info" in host.plugin_data assert "os_info" in host.plugin_data
# Recovery heartbeat. # Ordinary recovery heartbeat — no boot flag.
handle_datagram(_htb("rtt-hist-host5", rtt=13.0), ("127.0.0.1", 50000), transport, ctx) handle_datagram(_htb("rtt-hist-host5", rtt=13.0), ("127.0.0.1", 50000), transport, ctx)
assert "os_info" in host.plugin_data
assert len(host.plugin_data["rtt_ipv4"]) == 4
def test_boot_recovery_clears_real_plugin_data_but_preserves_rtt_history():
"""A recovery heartbeat carrying the boot flag (client process actually
restarted) must still wipe stale real plugin data, while rtt_* history
(still a valid measurement, unaffected by a client reboot) survives.
"""
hbdclass.Host.hosts.pop("rtt-hist-host6", None)
transport = _FakeTransport()
ctx = _base_ctx()
for rtt in (10.0, 11.0, 12.0):
handle_datagram(_htb("rtt-hist-host6", rtt=rtt), ("127.0.0.1", 50000), transport, ctx)
host = hbdclass.Host.hosts["rtt-hist-host6"]
assert len(host.plugin_data["rtt_ipv4"]) == 3
conn = host.connections["IPv4"]
conn.state = hbdclass.Connection.DOWN
host.add_plugin_data("os_info", {"os": "linux"}, timestamp=time.time())
assert "os_info" in host.plugin_data
# Recovery heartbeat with boot=1 — client process actually restarted.
handle_datagram(_htb("rtt-hist-host6", rtt=13.0, boot=1), ("127.0.0.1", 50000), transport, ctx)
assert "os_info" not in host.plugin_data assert "os_info" not in host.plugin_data
assert len(host.plugin_data["rtt_ipv4"]) == 4 assert len(host.plugin_data["rtt_ipv4"]) == 4