Notification channels are now managed through a proper web form instead
of a raw YAML textarea. Any authenticated user can create channels; private
channels (owner-scoped) are hidden from other users. The user profile
channel selector becomes a tag/chip picker with a "My Channels" CRUD section.
- settings.py: add CHANNEL_TYPE_SCHEMAS for all 6 notifier types; channel
section switches to section_mode="channels"; cards include owner/private/min_level
- configio.py: add apply_channel() and delete_channel() for per-entry CRUD
- notify.py: strip owner/private metadata before dispatching to drivers
- http.py: add GET/POST /api/0/notification_channels, PUT/DELETE /{name},
GET /api/0/notification_channel_types; visibility helper filters private
channels per user; PUT /api/0/users/me validates against visible channels
- settings.html: card grid with edit/delete per channel; add/edit modal
with type dropdown and dynamically rendered type-specific fields
- profile.html: chip picker replaces checkbox list; My Channels section
for creating/editing/deleting user-owned channels
- tests: update test_settings_sections, test_http_users_me; add
test_notification_channels_api (16 new tests, 46 total passing)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Setting enabled: false at the plugin level (e.g. memory_monitor: {enabled: false})
was silently ignored because the non-dict value was skipped by the metric parser,
leaving THRESHOLD_DEFAULTS entries active.
- _parse_plugin_thresholds: detect plugin-level enabled/enable flag and delete
all matching entries from target_dict (covers legacy and default config paths)
- _parse_multi_config named configs: inject disabled stubs from effective_defaults
into raw_overrides so the merge step overwrites inherited defaults
- Accept 'enable' as a tolerated alias for 'enabled' in both code paths
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
fetchHostGlance was only called for the initially expanded host, leaving
all other hosts showing "—" until manually expanded. Now fetches glance
for every host-card on DOMContentLoaded and refreshes all (not just
expanded) on the 30s auto-refresh interval.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
DOMContentLoaded was calling fetchHostGlance but not fetchHostInfo,
leaving the info-meta section stuck on "Loading…". Both the URL-hash
and default first-host paths now call fetchHostInfo and populate
infoCache on load.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Rename memory_monitor threshold key from 'percent' to 'memory_percent'
so it matches exactly rather than relying on suffix stripping, which was
causing swap_percent to be evaluated against the memory threshold
- Add swap_percent default thresholds (warning: 40%, critical: 75%)
- Add zfs_monitor pool capacity default thresholds (warning: 80%, critical: 90%)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Extracts host info assembly (owner, managers, hbc version/type,
last packet timestamp, threshold configs) into a testable module-level
helper, with 10 covering tests.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Allows any authenticated user to update their own full_name, avatar,
notification_channels, and password via the config YAML write path.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Move field-extraction inside the try/except in fetch_user so non-dict
responses from providers with empty profile_data_path (Gitea, GitHub)
raise OAuthError instead of an uncaught AttributeError. Apply
html.escape() to provider name, label, and logo URL in the login page.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replace hardcoded Gitea OAuth handlers with generic {name}-parameterized
routes and update the login page to render a button for each configured
provider via oauth_mod.get_providers().
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
purge_stale_alerts used _find_threshold to validate alert state keys,
but _find_threshold has no wildcard matching. A threshold configured as
"zfs_monitor.*.status" never matched the concrete alert state key
"zfs_monitor.tank.status", so every restart silently purged active ZFS
pool alert states and reset the grace period from scratch.
Also fix _check_pending_or_renotify to set last_notification after the
grace-period notification fires, so the re-notification interval is
anchored to when the alert was actually sent rather than the next PLG cycle.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add regex filter input to the Alerts dashboard that filters displayed
hosts on every keystroke; invalid regex turns the border red
- Initialise the filter from ?filter= in the URL query string
- Change _build_url() to produce /alerts?filter=<hostname> so
notification links (Pushover, email, Matrix, etc.) land on the
alerts page pre-filtered to the alerting host
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Reads oauth.gitea.logo from config and, when set, renders an <img>
inside the button with flex alignment.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Older hbd clients send zfs_monitor data with a `health` string but no
`health_ok` numeric field (added in a recent plugin update). Without
health_ok in the data, the wildcard threshold check found nothing and
no CRITICAL alert was raised for DEGRADED/SUSPENDED pools.
Synthesize health_ok from the health string in the server's nested-
metric loop so alerts fire regardless of client version.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Remove unused `_gitea_cfg_url` module-level helper from http.py
- Add logger.warning on invalid/expired state in oauth_gitea_callback
- Add test_callback_invalid_state_rejects and test_full_oauth_flow_chain to tests/test_oauth.py (21 tests total, all passing)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Creates or updates a user from an OAuth2 provider: new users are
inserted with an empty password_hash (OAuth-only login); existing users
have their display name and avatar refreshed while all other attributes
(admin flag, password_hash, notification_channels) are preserved.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add hbd/server/oauth.py with OAuthError, _gitea_cfg(), and is_enabled()
to detect when all three required Gitea OAuth2 config keys are present.
Add "oauth": {} default to SERVER_DEFAULTS in config.py.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Store owner in data-owner attribute; updateHostHeader always prepends it
so it survives innerHTML replacement. Render it immediately on page load
before JS fetches plugin data.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- When os_info arrives with no owner field, apply default_owner from server config
- Stop applying default_owner unconditionally in get_host_access (now deferred to os_info handling)
- os_info plugin logs debug message when injecting owner from client config
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Server sets request_update=1 in ACK when host.plugin_data is empty
- hbc: AsyncConnection.request_info_event; handle_ack sets it on request_update
- hbc: _info_plugin_refresh_loop clears InfoPlugin caches and resends on demand
- hbc_mini: same via _request_info event and _info_refresh_loop
- docs/USERS.md: document client-declared owner config key
- docs/PLUGIN_DEVELOPMENT.md: document server-initiated InfoPlugin refresh
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- owner: optional top-level config key in ~/.hbc.yaml / ~/.hbc.json
- Propagated into plugin configs at load time so os_info can include it
- os_info PLG data carries owner field when set
- udp: sets host.owner from os_info if not already configured server-side
- live.html: format event log timestamps as YYYY-MM-DD HH:MM:SS (24-hour)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- eventlog() now stores {ts, host, level, service, message} dicts instead of strings
- WebSocket sends/broadcasts filter event log messages by the user's managed hosts
- live.html renders structured log entries with level-coloured spans
- Swiss railway clock minute hand now holds until second hand reaches 12, then steps
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Alerts page: move metric name into the header row alongside hostname.
Notifications: include metric name in title (hostname metric) and
strip the metric prefix from the body so it contains only value/detail.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
These fields were never read by the plugin; thresholds are configured
server-side. Also document the -b flag in README.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
error_received() no longer sets _dead=True; it just closes the transport
so the existing retry loop in heartbeat_sender (hbc) and sendto (hbc_mini)
reopens the connection on the next interval. This allows hbc to recover
when it starts before network connectivity is established.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- threshold: fix crash when display is None (_format_display now falls
back to default format string instead of calling None.format())
- threshold: shorten notification messages by stripping plugin-name prefix
from metric_path (cpu_percent instead of cpu_monitor.cpu_percent)
- main: demote aiohttp.access log records from INFO to DEBUG
- udp: replace debug print with proper logger.info for new host sign-on
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Seed threshold_configs["default"] from THRESHOLD_DEFAULTS at the start
of _parse_config() so the Settings page displays built-in defaults
regardless of whether the server config uses the multi-config format,
the legacy thresholds: format, or has no threshold config at all.
_parse_multi_config() overwrites the seed with the fully-merged
effective defaults when a threshold_configs section is present.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add ComparisonOperator.NAGIOS ("nagios") that maps Nagios exit codes
directly to alert levels (0=OK 1=WARNING 2=CRITICAL 3=UNKNOWN) without
requiring numeric warning/critical thresholds. Hysteresis is bypassed for
discrete codes. Display template defaults to "{check_name}: {output}".
_format_display() handles None threshold_value gracefully.
Add nagios_runner.status_code as a built-in default threshold config so
nagios checks alert out of the box.
Also: fix alerts.html scrolling (override html,body), make hostname a link
to /plugins#<hostname>, remove overall_status/overall_status_code/plugin_count
from nagios_runner and hbc_mini, replace with computed worst-status in
plugins.html via nagiosWorstStatus() helper.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- nagios_runner: remove overall_status/overall_status_code/plugin_count fields;
each command still reports its own <name>_status and <name>_status_code
- threshold: expose {output} and {status} aliases in display templates for
nagios_runner generic matches (mapped from <check_name>_output/status)
- alerts.html: fix scrolling by overriding html,body height/overflow (style.css
sets both); make hostname a link to /plugins/<hostname>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Andreas WredeandClaude Sonnet 4.6
andreasandClaude Sonnet 4.6