Add Job Monitoring card to admin dashboard with:
- Hero icon for visual identification
- Brief description of monitoring capabilities
- Link to /admin/monitoring dashboard
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Add LiveView for real-time job monitoring dashboard:
- Subscribe to job:lifecycle PubSub topic for live updates
- Display health metrics (completed, failed, avg duration, active jobs)
- Show active operations with real-time updates
- Display problems section (stuck/failed job counts)
- Add router entry under /admin/monitoring
Tests verify PubSub subscription and basic rendering.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Add Events.broadcast_job_event calls to track job lifecycle:
- :started when job begins executing
- :completed when job finishes (including :discard status)
- :failed when job crashes with error
- Track duration in seconds for all job executions
Wrap main logic in try/rescue to ensure events broadcast even on crash.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Add Events.broadcast_job_event calls to track job lifecycle:
- :started when job begins executing
- :completed/:failed when job finishes
- Track duration in seconds for all job executions
Broadcast events enable real-time monitoring of polling operations.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Create JobMonitoring.Metrics module with calculate_all/0 to provide
comprehensive metrics including execution counts, success rates,
queue depths, and average execution times.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Add list_failed_jobs/0 to return retryable and cancelled jobs
and list_recent_completions/1 to return completed, cancelled,
or discarded jobs with configurable limit.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Changes:
- Fix traffic graph to handle nil interface octets gracefully
- Add comprehensive tests for nil octets scenarios
- Fix agent polling query to support devices assigned directly to org (left join on sites)
- Fix test SNMP connection to inherit credentials from site/org
- Refactor agent_channel process_polling_result to reduce nesting (Credo)
- Add alias for JobCleanupTask in application.ex (Credo)
This fixes crashes when interface stats have nil octets and ensures
devices without a site assignment can be polled by agents.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Added Logger.require for debug logging
- Refactored fetch_latest_agent_data to reduce nesting depth
- Added debug logging to trace agent assignment detection
- Check for nil/stale sensor readings and log appropriately
- Use actual reading timestamp instead of current time for data points
This will help diagnose why live polling shows 0 values and why
Phoenix SNMP disabled messages appear for agent-assigned devices.
Modified GraphLive live polling to check for agent assignments and use
agent-collected sensor data from the database instead of doing direct
SNMP polling. This respects the agent assignment and prevents Phoenix
from bypassing the agent for live graph updates.
Also refactored handle_params to reduce nesting depth.
Devices can belong directly to organizations without sites. Fixed
perform_live_poll to check device.organization_id instead of
device.site.organization_id which caused BadMapError when site is nil.
Also removed unused Repo alias.
Discovery is now working successfully with SNMPv3 after increasing
agent timeout to 60s. Removing the temporary debug logging that was
added to verify credentials were being properly decrypted and sent
to the agent.
Added logging to show what SNMPv3 parameters are being sent from Phoenix
to the agent, including password lengths to verify they're decrypted
properly.
Added JobCleanupTask that runs on production startup to cancel all
existing polling jobs and reschedule them with the latest worker code.
This ensures:
- No stale jobs run with old code after deployment (e.g., missing SNMPv3 credentials)
- All devices get fresh polling jobs with current logic
- Production deployments are seamless without manual intervention
The task:
- Only runs in production environment (skipped in dev/test)
- Runs asynchronously after supervisor initialization
- Cancels all DevicePollerWorker and DeviceMonitorWorker jobs
- Reschedules polling for all SNMP-enabled devices
- Is idempotent and safe to run multiple times
Removed debug logging from Client and DevicePollerWorker as SNMPv3
implementation is now verified and working correctly.
Added logging to show what options are being built in DevicePollerWorker,
including whether security_name (v3) or community (v2c) credentials are
being included.
The DevicePollerWorker was only passing SNMPv2c credentials (community
string) to the SNMP client, causing KeyError when polling SNMPv3 devices.
Changes:
- Updated build_client_opts in device_poller_worker.ex to detect version "3"
- Added same SNMPv3 credential handling as discovery.ex
- Includes: security_name, security_level, auth_protocol, auth_password,
priv_protocol, priv_password
This completes the SNMPv3 implementation - both discovery and polling now
support v3 authentication alongside the existing v2c support.
The discovery process was only passing SNMPv2c credentials (community
string) to the SNMP client, causing SNMPv3 devices to fail discovery
even when v3 credentials were configured.
Changes:
- Updated build_client_opts in discovery.ex to include SNMPv3 credentials
(security_name, security_level, auth_protocol, auth_password,
priv_protocol, priv_password) when version is "3"
- Updated build_snmp_opts in client.ex to pass v3 credentials to SnmpKit
- Added parse functions for security_level, auth_protocol, and priv_protocol
- Updated connection_opts typespec to include v3 parameters
- Fixed parameter names to match SnmpKit expectations (security_name vs sec_name)
This enables full SNMPv3 discovery on devices like MikroTik routers that
use v3 authentication, allowing them to discover the same sensors and
data that SNMPv2c would show.
Fixes BadMapError when loading device show page for devices without
an assigned site. Changed to use device.organization_id directly
instead of device.site.organization_id since site can be nil but
organization is always present.
Extract common pattern from perform_site_reorder/4 and perform_device_reorder/4
into a single perform_reorder/4 helper function.
Both reorder handlers now delegate to the shared helper, reducing duplication
and making the reload/regroup/flash pattern consistent.
- Reduces code from 40 lines to 28 lines
- Maintains identical behavior (all tests pass)
- Uses static gettext strings to satisfy compile-time requirements
Remove duplicate device counting logic by consolidating three identical
implementations into a single reusable function.
**Changes:**
- Refactor agent_live/index.ex mount/3 to use calculate_device_counts/1
- Delete duplicate calculate_cloud_poller_counts/1 function
- Replace all usages with calculate_device_counts/1
**Before:**
- Lines 31-47: Inline counting for agent_tokens and cloud_pollers
- Lines 312-318: calculate_device_counts/1
- Lines 320-326: calculate_cloud_poller_counts/1 (identical logic)
**After:**
- Single calculate_device_counts/1 function used everywhere
- ~20 lines removed
**Benefits:**
- Eliminates triple duplication of counting logic
- Single source of truth for device counting
- Easier to maintain and test
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Extract duplicate access control logic from multiple LiveView files into
a reusable AccessControl helper module.
**Changes:**
- NEW: lib/towerops_web/live/helpers/access_control.ex
- NEW: test/towerops_web/live/helpers/access_control_test.exs
- Refactor device_live/index.ex to use AccessControl.verify_site_access/2
and verify_device_access/2
- Refactor device_live/form.ex to use AccessControl.verify_device_access/2
- Refactor alert_live/index.ex to use AccessControl.verify_alert_access/2
- Refactor device_live/show.ex to use AccessControl.verify_device_access/2
- Remove unused Repo aliases from refactored files
**Benefits:**
- Reduces code duplication across 7+ locations
- Centralizes security-critical access checks
- Improves testability (helper module has >90% coverage)
- Consistent error handling across all LiveViews
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Features:
- Users can set a default organization in org settings
- First organization created is automatically set as default
- Default org selector only shown if user has 2+ organizations
- Migration auto-sets first org as default for existing users
Database changes:
- Added is_default boolean field to organization_memberships
- Added partial index for efficient default org lookups
- Migration sets oldest org as default for each existing user
UI changes:
- Organization settings page shows default org toggle
- Green badge displayed when org is already default
- Button to set org as default if it's not current default
Backend:
- Organizations.set_default_organization/2 manages defaults
- Organizations.get_default_membership/1 retrieves default
- Auto-set first org as default in create_organization/3
When creating devices via API, organization_id is now properly defaulted
to the authenticated organization's ID even when an empty string is sent.
This fixes Terraform provider compatibility where organization_id may be
sent as "" instead of omitted entirely.
Before: Only checked if key was missing (Map.has_key?)
After: Also handles nil and empty string values
