Host disk bars were showing virtual filesystems like tmpfs, /dev, /run,
/sys, and Docker overlay mounts. These clutter the UI and don't represent
meaningful disk usage.
Changed from `shouldIgnoreReadOnlyFilesystem` (read-only only) to the
full `fsfilters.ShouldSkipFilesystem` which also excludes:
- Virtual FS types: tmpfs, devtmpfs, sysfs, proc, cgroup, etc.
- Special mountpoints: /dev, /proc, /sys, /run, /var/lib/docker, /snap
- Network filesystems: fuse, nfs, cifs, etc.
Related to #790
Add 13 test cases for convertPBSSnapshots covering:
- Empty/nil input handling
- Basic snapshot field mapping
- Files as string array
- Files as object array with filename field
- Invalid file objects ignored gracefully
- Verification status as string ("ok" vs other)
- Verification status as object with state field
- Verification object without state field
- Multiple snapshots with unique IDs
- Empty namespace handling in ID generation
- Container (ct) backup type
Coverage: convertPBSSnapshots 0% -> 100%
- Test nil input returns nil
- Test empty slice returns nil
- Test single task with all fields populated
- Test task with error state (failed tasks)
- Test multiple tasks conversion
- Test nil time pointers preserved as nil
- Test zero time values converted to nil (defensive handling)
- Test all optional time fields (UpdatedAt, StartedAt, CompletedAt)
Improves monitoring package coverage for Docker Swarm task handling.
- sanitizeGuestAddressStrings: IPv6 loopback with CIDR (::1/128) edge case
(96.2% -> 100%)
- parseContainerConfigNetworks: parts without equals, mac/ips/ip6addr/ip6prefix
key variants, whitespace-only values, all-empty results (97.2% -> 100%)
- mergeNVMeTempsIntoDisks: NVMe disk with no legacy temps (continue branch),
more disks than temps (break branch) (97% -> 100%)
Go map iteration order is non-deterministic, causing flaky test failures.
This fix ensures "Composite" sensor is always preferred over "Sensor 1"
by checking them in separate loops rather than relying on iteration order.
- recordGuestSnapshot: nil handling, field setting, map initialization
- GetDiagnosticSnapshots: nil/empty handling, sorting verification
Both functions now at 100% coverage.
When tasks have identical NextRun and Priority, the Less function
falls back to comparing InstanceName alphabetically. Add test to
cover this edge case branch, improving Less coverage to 100%.
Tests for LXC container config parsing functions:
- sanitizeRootFSDevice: rootfs device extraction
- collectIPsFromInterface: recursive IP extraction from various types
- parseContainerRawIPs: JSON IP parsing
- parseContainerConfigNetworks: network interface parsing with MAC
normalization, CIDR stripping, sorting
- parseContainerMountMetadata: mount point metadata parsing
- extractContainerRootDeviceFromConfig: root device extraction
- mergeContainerNetworkInterface: interface merging by name/MAC
- convertContainerDiskInfo: Proxmox disk info conversion
- ensureContainerRootDiskEntry: root disk entry creation
91+ test cases covering edge cases, invalid data, and complex scenarios.
Coverage improved from 40.6% to 42.2%.
Tests for all 9 functions handling Docker host identification:
- uniqueNonEmptyStrings: dedupe with order preservation
- sanitizeDockerHostSuffix: string cleaning for IDs
- tokenHintFromRecord: redacted token hints
- dockerHostIDExists: ID existence check
- dockerHostSuffixCandidates: candidate suffix generation
- fallbackDockerHostID: hash-based ID generation
- findMatchingDockerHost: host matching by various criteria
- generateDockerHostIdentifier: unique ID generation with suffixes
- resolveDockerHostIdentifier: main resolver orchestration
86 test cases covering edge cases (nil, empty, whitespace), Unicode
handling, and complex matching scenarios.
Tests cover nil/empty inputs, basic field copying, time fields
(CreatedAt/UpdatedAt with nil and zero value handling), update status
conversion, endpoint ports, and labels cloning.
- Use public API AcknowledgeDockerHostCommand instead of private function
- Use strings.Contains instead of custom helper functions
- Add test case for empty hostID (skips validation, succeeds)
The error message referenced "Settings -> Docker -> Removed hosts" but
that UI path no longer exists. The correct path is now
"Settings -> Agents -> Removed Docker Hosts".
Related to #778
Add 56 test cases covering:
- isTransientError: 13 cases for context errors, retryable MonitorErrors,
and standard error types (nil, Canceled, DeadlineExceeded, Timeout, etc.)
- shouldTryPortlessFallback: 14 cases for connection error patterns that
trigger port fallback (refused, reset, no such host, timeout variants)
- shouldAttemptFallback: 13 cases for timeout/deadline/canceled patterns
used in storage polling fallback
- classifyDLQReason: 8 cases for dead letter queue reason classification
(max_retry_attempts vs permanent_failure)
- Edge cases: 3 cases for overlapping error patterns and partial matches
First test file for these error classification utilities used in retry
and fallback decision logic.
Add 97 test cases for four utility functions in monitor.go:
- stringValue (30 cases): Type-to-string conversion for various Go types
including string, int types, float types, json.Number, fmt.Stringer
- anyToInt64 (34 cases): Interface-to-int64 conversion with overflow
handling, string parsing, and json.Number support
- parseInterfaceStat (15 cases): Interface stat map parsing for network
interface statistics
- extractGuestOSInfo (18 cases): QEMU guest agent OS info extraction
with fallback logic for various Linux distros, Windows, FreeBSD
Coverage for these functions: 0% → 100%
Package coverage: 37.5% → 38.5%
Added comprehensive tests for 6 previously untested pure utility functions:
- sortContent: sorts comma-separated storage content values
- formatSeconds: converts total seconds to HH:MM:SS format
- dedupeStringsPreserveOrder: deduplicates strings while preserving order
- sanitizeGuestAddressStrings: validates and filters IP addresses
- copyFloatPointer: creates independent copy of float64 pointer
- clampInterval: clamps time.Duration to specified bounds
Tests cover edge cases including empty inputs, boundary conditions,
special values (DHCP, loopback addresses), and various separators.
Coverage increased from 36.8% to 37.5%.
32 test cases covering makeNodeSnapshotKey and makeGuestSnapshotKey
utility functions used for diagnostic memory snapshot storage. Tests
cover simple inputs, edge cases (empty strings, negative VMIDs),
special characters, and key uniqueness verification. Also includes
struct field tests for NodeMemoryRaw and VMMemoryRaw.
First test file for diagnostic_snapshots.go. Coverage 36.7% → 36.8%.
The backup status indicator feature was incomplete - it added the UI
component but never populated VM/Container LastBackup from actual
backup data. Now SyncGuestBackupTimes() is called after storage
backups and PBS backups are polled, matching each guest's VMID to
its most recent backup timestamp.
Fixes#786
46 test cases covering:
- clampFloat helper (9 cases: range checks, boundary conditions)
- Staleness score impact (5 cases: 0/0.25/0.5/0.75/1 scores)
- Error penalty (5 cases: 0 to 10 errors, min clamping)
- Queue depth stretching (5 cases: 1 to 50 depth, max clamping)
- EMA smoothing (convergence behavior with alpha=0.6)
- Jitter (bounds checking, deterministic tests with seeded RNG)
- Boundary conditions (6 cases: zero/negative intervals, max<min)
- Combined factors (3 cases: all factors interacting)
- State persistence (per-instance EMA state isolation)
First unit test file for scheduler.go (previously only had integration tests).
Add focused unit tests for four utility functions in temperature.go:
- extractTempInput: 16 test cases for sensor value extraction
- extractCoreNumber: 18 test cases for core number parsing
- extractHostname: 21 test cases for URL hostname extraction
- normalizeSMARTEntries: 15 test cases for SMART data normalization
70 test cases total covering type conversions, edge cases,
boundary conditions, and error handling paths.
Test coverage for isCephStorageType, countServiceDaemons,
extractCephCheckSummary, and summarizeCephHealth functions.
These parse Ceph storage types and health status JSON.
Test normalizeLabel, normalizeNodeLabel, splitInstanceKey,
breakerStateToValue, sanitizeInstanceLabels, makeMetricKey,
and makeNodeMetricKey with 36 test cases covering edge cases
like empty strings, whitespace, and invalid inputs.
Move the inline filesystem skip logic from pollVMsAndContainersEfficient
into a reusable ShouldSkipFilesystem function. This consolidates filtering
for virtual filesystems (tmpfs, cgroup, etc.), network mounts (nfs, cifs,
fuse), and special mountpoints (/dev, /proc, /snap, etc.) into one tested
location.
Reduces cyclomatic complexity of pollVMsAndContainersEfficient and adds
28 test cases covering virtual fs types, network mounts, special mounts,
Windows paths, and edge cases.
When api_tokens.json is modified on disk, the ConfigWatcher reloads
the tokens into memory. However, the Monitor's dockerTokenBindings and
hostTokenBindings maps were not synchronized with the new token set,
causing orphaned bindings when agents reconnect after reinstall.
Add SetAPITokenReloadCallback to ConfigWatcher that triggers Monitor's
new RebuildTokenBindings method after token reload. This method
reconstructs the binding maps from current Docker host and host agent
state, keeping only bindings for tokens that still exist in config.
Related to #773
Mark intentionally unused parameters with underscore to:
- Silence unparam warnings for legitimate unused parameters
- Keep function signatures intact for API compatibility
- Remove unused req from serveChecksum helper
- Replace custom maxInt64 helper with Go 1.21+ builtin max()
- Mark unused cfg parameter in newAdaptiveIntervalSelector
- Remove test for deleted helper function
- Merge variable declaration with assignment (S1021)
- Use unconditional strings.TrimPrefix (S1017)
- Remove unnecessary nil checks around range (S1031)
- Remove unnecessary fmt.Sprintf (S1039)
- Use copy() instead of manual loop (S1001)
- Use time.Until instead of t.Sub(time.Now()) (S1024)
- Use buf.String() instead of string(buf.Bytes()) (S1030)
- Fix SA4006 unused value issues in ssh.go, validation.go, generator.go
- Replace deprecated ioutil with io/os in config.go
- Replace deprecated tar.TypeRegA with tar.TypeReg
- Remove deprecated rand.Seed calls (auto-seeded in Go 1.20+)
- Fix always-true nil check in main.go
- Fix impossible nil comparison in tempproxy/client.go
- Add nil check for config in monitor.New()
- Use defer for tempCancel() to ensure context is always cancelled
- Remove redundant shouldCollect variable that was always true
- Fix indentation after removing the unnecessary conditional block
Add seamless migration path from legacy agents to unified agent:
- Add AgentType field to report payloads (unified vs legacy detection)
- Update server to detect legacy agents by type instead of version
- Add UI banner showing upgrade command when legacy agents are detected
- Add deprecation notice to install-host-agent.ps1
- Create install-docker-agent.sh stub that redirects to unified installer
Legacy agents (pulse-host-agent, pulse-docker-agent) now show a "Legacy"
badge in the UI with a one-click copy command to upgrade to the unified
agent.
- Implemented adaptive layout for NodeSummaryTable with responsive columns and sticky name column.
- Fixed GuestRow background display issues.
- Added IsLegacy field to Host and DockerHost models to flag legacy agents (version < 1.0.0).
- Updated monitor to populate IsLegacy based on agent version.
- Removed global legacySSHDisabled flag that was triggered by any single node auth failure
- Changed disableLegacySSHOnAuthFailure to only log warnings
- Fixed potential context leak in monitor.go
- Updated tests to reflect removal of global disable logic
Fixes#727. Previously, if temperature monitoring was enabled and a node wasn't found in ClusterEndpoints, the entire node processing was skipped. This change ensures we only skip temperature collection.
During cluster startup, nodes were temporarily using the primary cluster
endpoint for temperature collection before cluster metadata validation
completed. This caused all nodes to show the same (incorrect) temperature
values for ~4 minutes until validation finished and per-node endpoints
were established.
Example: minipc would show delly's temperature (90°C) instead of its own
(50°C) from startup until cluster validation completed.
Root cause:
- Temperature collection started immediately at startup
- Cluster endpoint validation happened asynchronously
- Code fell back to primary endpoint when ClusterEndpoints was empty
- All nodes used same endpoint, got same temperature data
Fix: Skip temperature collection for cluster nodes until:
1. ClusterEndpoints array is populated (validation complete)
2. Node's specific endpoint is found in the cluster metadata
This ensures correct temperature data from the very first collection,
maintaining data integrity during startup. When persisted config exists,
endpoints are available immediately so no delay occurs. For new clusters,
temperature collection begins once validation completes (~30s).
Preserves Pulse's correctness guarantee: users can trust metrics
immediately after restart without waiting for "warm-up" period.
