Compare commits
6 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 156ef97667 | |||
| 2f74266bd9 | |||
| 222bdb08ab | |||
| 8dd744b039 | |||
| 9e2be150b5 | |||
| ed5ba5c59e |
@@ -174,17 +174,26 @@ _PAGE_LIMIT = 200 # max events returned per request
|
|||||||
|
|
||||||
|
|
||||||
def _annotate_suppressions(events: list, suppressions: list) -> None:
|
def _annotate_suppressions(events: list, suppressions: list) -> None:
|
||||||
"""Annotate each event dict in-place with an is_suppressed bool."""
|
"""Annotate each event dict in-place with an is_suppressed bool.
|
||||||
|
|
||||||
|
Mirrors the suppression check order in monitor.py exactly:
|
||||||
|
interface_down → interface OR host
|
||||||
|
unifi_device_* → unifi_device
|
||||||
|
everything else → host
|
||||||
|
"""
|
||||||
for ev in events:
|
for ev in events:
|
||||||
sup_type = (
|
etype = ev.get('event_type', '')
|
||||||
'unifi_device' if ev.get('event_type') == 'unifi_device_offline'
|
name = ev.get('target_name', '')
|
||||||
else 'interface' if ev.get('event_type') == 'interface_down'
|
detail = ev.get('target_detail', '') or ''
|
||||||
else 'host'
|
if etype == 'interface_down':
|
||||||
)
|
ev['is_suppressed'] = (
|
||||||
ev['is_suppressed'] = db.check_suppressed(
|
db.check_suppressed(suppressions, 'interface', name, detail) or
|
||||||
suppressions, sup_type,
|
db.check_suppressed(suppressions, 'host', name)
|
||||||
ev.get('target_name', ''), ev.get('target_detail', '') or '',
|
)
|
||||||
)
|
elif etype == 'unifi_device_offline':
|
||||||
|
ev['is_suppressed'] = db.check_suppressed(suppressions, 'unifi_device', name, detail)
|
||||||
|
else:
|
||||||
|
ev['is_suppressed'] = db.check_suppressed(suppressions, 'host', name, detail)
|
||||||
|
|
||||||
|
|
||||||
# ---------------------------------------------------------------------------
|
# ---------------------------------------------------------------------------
|
||||||
|
|||||||
+1
-1
@@ -68,7 +68,7 @@ class DiagnosticsRunner:
|
|||||||
f' echo "=== ip_route ===";'
|
f' echo "=== ip_route ===";'
|
||||||
f' ip route show dev {q} 2>/dev/null;'
|
f' ip route show dev {q} 2>/dev/null;'
|
||||||
f' echo "=== dmesg ===";'
|
f' echo "=== dmesg ===";'
|
||||||
f' dmesg 2>/dev/null | grep {q} | tail -50;'
|
f' dmesg 2>/dev/null | grep -F -- {q} | tail -50;'
|
||||||
f' echo "=== lldpctl ===";'
|
f' echo "=== lldpctl ===";'
|
||||||
f' lldpctl 2>/dev/null || echo "lldpd not running";'
|
f' lldpctl 2>/dev/null || echo "lldpd not running";'
|
||||||
f' echo "=== end ==="'
|
f' echo "=== end ==="'
|
||||||
|
|||||||
+7
-6
@@ -734,7 +734,7 @@ class NetworkMonitor:
|
|||||||
f'Interface {iface} on {host} went link-down ({_now_utc()})',
|
f'Interface {iface} on {host} went link-down ({_now_utc()})',
|
||||||
)
|
)
|
||||||
if not sup and consec >= self.fail_thresh:
|
if not sup and consec >= self.fail_thresh:
|
||||||
self._ticket_interface(event_id, is_new, host, iface, consec)
|
self._ticket_interface(event_id, host, iface, consec)
|
||||||
|
|
||||||
if host_has_regression:
|
if host_has_regression:
|
||||||
hosts_with_regression.append(host)
|
hosts_with_regression.append(host)
|
||||||
@@ -771,7 +771,7 @@ class NetworkMonitor:
|
|||||||
db.resolve_event('cluster_network_issue', self.cluster_name, '')
|
db.resolve_event('cluster_network_issue', self.cluster_name, '')
|
||||||
|
|
||||||
def _ticket_interface(
|
def _ticket_interface(
|
||||||
self, event_id: int, is_new: bool, host: str, iface: str, consec: int
|
self, event_id: int, host: str, iface: str, consec: int
|
||||||
) -> None:
|
) -> None:
|
||||||
title = (
|
title = (
|
||||||
f'[{host}][auto][production][issue][network][single-node] '
|
f'[{host}][auto][production][issue][network][single-node] '
|
||||||
@@ -810,11 +810,11 @@ class NetworkMonitor:
|
|||||||
f'UniFi {name} ({d.get("ip","")}) offline ({_now_utc()})',
|
f'UniFi {name} ({d.get("ip","")}) offline ({_now_utc()})',
|
||||||
)
|
)
|
||||||
if not sup and consec >= self.fail_thresh:
|
if not sup and consec >= self.fail_thresh:
|
||||||
self._ticket_unifi(event_id, is_new, d)
|
self._ticket_unifi(event_id, d)
|
||||||
else:
|
else:
|
||||||
db.resolve_event('unifi_device_offline', name, d.get('type', ''))
|
db.resolve_event('unifi_device_offline', name, d.get('type', ''))
|
||||||
|
|
||||||
def _ticket_unifi(self, event_id: int, is_new: bool, device: dict) -> None:
|
def _ticket_unifi(self, event_id: int, device: dict) -> None:
|
||||||
name = device['name']
|
name = device['name']
|
||||||
title = (
|
title = (
|
||||||
f'[{name}][auto][production][issue][network][single-node] '
|
f'[{name}][auto][production][issue][network][single-node] '
|
||||||
@@ -850,12 +850,12 @@ class NetworkMonitor:
|
|||||||
f'Host {name} ({ip}) unreachable via ping ({_now_utc()})',
|
f'Host {name} ({ip}) unreachable via ping ({_now_utc()})',
|
||||||
)
|
)
|
||||||
if not sup and consec >= self.fail_thresh:
|
if not sup and consec >= self.fail_thresh:
|
||||||
self._ticket_unreachable(event_id, is_new, name, ip, consec)
|
self._ticket_unreachable(event_id, name, ip, consec)
|
||||||
else:
|
else:
|
||||||
db.resolve_event('host_unreachable', name, ip)
|
db.resolve_event('host_unreachable', name, ip)
|
||||||
|
|
||||||
def _ticket_unreachable(
|
def _ticket_unreachable(
|
||||||
self, event_id: int, is_new: bool, name: str, ip: str, consec: int
|
self, event_id: int, name: str, ip: str, consec: int
|
||||||
) -> None:
|
) -> None:
|
||||||
title = (
|
title = (
|
||||||
f'[{name}][auto][production][issue][network][single-node] '
|
f'[{name}][auto][production][issue][network][single-node] '
|
||||||
@@ -970,6 +970,7 @@ class NetworkMonitor:
|
|||||||
except Exception as e:
|
except Exception as e:
|
||||||
logger.error(f'Monitor loop error: {e}', exc_info=True)
|
logger.error(f'Monitor loop error: {e}', exc_info=True)
|
||||||
time.sleep(30)
|
time.sleep(30)
|
||||||
|
continue
|
||||||
|
|
||||||
time.sleep(self.poll_interval)
|
time.sleep(self.poll_interval)
|
||||||
|
|
||||||
|
|||||||
@@ -324,6 +324,7 @@
|
|||||||
</div>
|
</div>
|
||||||
</div>
|
</div>
|
||||||
<div class="host-grid" id="host-grid">
|
<div class="host-grid" id="host-grid">
|
||||||
|
{%- set has_global_sup = suppressions | selectattr('target_type', 'equalto', 'all') | list | length > 0 -%}
|
||||||
{% for name, host in snapshot.hosts.items() %}
|
{% for name, host in snapshot.hosts.items() %}
|
||||||
{% set suppressed = suppressions | selectattr('target_name', 'equalto', name) | list %}
|
{% set suppressed = suppressions | selectattr('target_name', 'equalto', name) | list %}
|
||||||
<div class="host-card host-card-{{ host.status }}" data-host="{{ name }}">
|
<div class="host-card host-card-{{ host.status }}" data-host="{{ name }}">
|
||||||
@@ -331,7 +332,7 @@
|
|||||||
<div class="host-name-row">
|
<div class="host-name-row">
|
||||||
<span class="host-status-dot dot-{{ host.status }}"></span>
|
<span class="host-status-dot dot-{{ host.status }}"></span>
|
||||||
<span class="host-name">{{ name }}</span>
|
<span class="host-name">{{ name }}</span>
|
||||||
{% if suppressed %}
|
{% if suppressed or has_global_sup %}
|
||||||
<span class="badge-suppressed" title="Suppressed">🔕</span>
|
<span class="badge-suppressed" title="Suppressed">🔕</span>
|
||||||
{% endif %}
|
{% endif %}
|
||||||
</div>
|
</div>
|
||||||
@@ -468,7 +469,7 @@
|
|||||||
{% block scripts %}
|
{% block scripts %}
|
||||||
<script>
|
<script>
|
||||||
// Start auto-refresh using saved settings interval (default 30 s)
|
// Start auto-refresh using saved settings interval (default 30 s)
|
||||||
const _savedInterval = (window.gandalfSettings && window.gandalfSettings.refreshInterval) || 30;
|
const _savedInterval = window.gandalfSettings?.refreshInterval ?? 30;
|
||||||
if (_savedInterval > 0) lt.autoRefresh.start(refreshAll, _savedInterval * 1000);
|
if (_savedInterval > 0) lt.autoRefresh.start(refreshAll, _savedInterval * 1000);
|
||||||
|
|
||||||
// When settings change, restart auto-refresh with new interval
|
// When settings change, restart auto-refresh with new interval
|
||||||
|
|||||||
@@ -571,7 +571,7 @@ async function loadLinks() {
|
|||||||
}
|
}
|
||||||
|
|
||||||
loadLinks();
|
loadLinks();
|
||||||
const _linksInterval = (window.gandalfSettings && window.gandalfSettings.refreshInterval) || 60;
|
const _linksInterval = window.gandalfSettings?.refreshInterval ?? 60;
|
||||||
if (_linksInterval > 0) lt.autoRefresh.start(loadLinks, Math.max(_linksInterval, 15) * 1000);
|
if (_linksInterval > 0) lt.autoRefresh.start(loadLinks, Math.max(_linksInterval, 15) * 1000);
|
||||||
|
|
||||||
window.onGandalfSettingsChanged = function(s) {
|
window.onGandalfSettingsChanged = function(s) {
|
||||||
|
|||||||
@@ -36,6 +36,12 @@ class TestBuildSshCommand:
|
|||||||
cmd = DiagnosticsRunner.build_ssh_command('10.0.0.1', 'eth0')
|
cmd = DiagnosticsRunner.build_ssh_command('10.0.0.1', 'eth0')
|
||||||
assert 'ethtool' in cmd
|
assert 'ethtool' in cmd
|
||||||
|
|
||||||
|
def test_dmesg_uses_fixed_string_grep(self):
|
||||||
|
# grep -F prevents iface names with dots (e.g. eth0.1) being treated as
|
||||||
|
# regex wildcards; -- prevents leading - from being parsed as a flag
|
||||||
|
cmd = DiagnosticsRunner.build_ssh_command('10.0.0.1', 'eth0')
|
||||||
|
assert 'grep -F --' in cmd
|
||||||
|
|
||||||
|
|
||||||
# ── parse_output ─────────────────────────────────────────────────────────────
|
# ── parse_output ─────────────────────────────────────────────────────────────
|
||||||
|
|
||||||
|
|||||||
Reference in New Issue
Block a user