Skip to content

fix: bound emergency cleanup task results #981

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
chubes4 merged 1 commit into from
Jul 29, 2026
Merged

fix: bound emergency cleanup task results #981

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
89 changes: 85 additions & 4 deletions inc/Tasks/WorkspaceDiskEmergencyCleanupTask.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,8 +16,9 @@

class WorkspaceDiskEmergencyCleanupTask extends SystemTask {



private const MAX_DURABLE_REASONS = 10;
private const MAX_DURABLE_REASON_BYTES = 120;
private const MAX_DURABLE_JOB_IDS = 25;
/**
* PluginSettings key that gates threshold-triggered emergency cleanup.
*/
Expand Down Expand Up @@ -125,6 +126,7 @@ public function executeTask( int $jobId, array $params ): void {
! empty($result['action_required']) ? 'yes' : 'no'
)
: 'Workspace disk emergency cleanup skipped: disk thresholds not crossed.';
$durable_report = $this->compact_durable_report($result);

do_action(
'datamachine_log',
Expand All @@ -133,11 +135,90 @@ public function executeTask( int $jobId, array $params ): void {
array(
'task' => $this->getTaskType(),
'jobId' => $jobId,
'report' => $result,
'report' => $durable_report,
)
);

$this->completeJob($jobId, $result);
$this->completeJob($jobId, $durable_report);
}

/**
* Project a recurring task result into bounded durable evidence.
*
* Workspace abilities and CLI commands intentionally retain the complete
* plan for explicit operator inspection. Recurring jobs retain only the
* operational facts needed to diagnose disk pressure and scheduled work.
*
* @param array<string,mixed> $result Full emergency cleanup result.
* @return array<string,mixed>
*/
private function compact_durable_report( array $result ): array {
$budget = (array) ( $result['disk_budget'] ?? array() );
$summary = (array) ( $result['scheduled_summary'] ?? array() );
$selection = (array) ( $result['inode_recovery_plan'] ?? array() );
$capacity = (array) ( $result['capacity_evidence'] ?? array() );

return array(
'success' => ! empty($result['success']),
'triggered' => ! empty($result['triggered']),
'skipped' => ! empty($result['skipped']),
'dry_run' => ! empty($result['dry_run']),
'generated_at' => (string) ( $result['generated_at'] ?? '' ),
'trigger_reasons' => $this->compact_reasons($budget['trigger_reasons'] ?? array()),
'selected_artifact_count' => (int) ( $result['selected_artifact_count'] ?? 0 ),
'selected_worktree_count' => (int) ( $result['selected_worktree_count'] ?? 0 ),
'scheduled' => array(
'chunks' => (int) ( $summary['scheduled_chunks'] ?? 0 ),
'artifact_rows' => (int) ( $summary['scheduled_artifact_rows'] ?? 0 ),
'batch_job_id' => (int) ( $summary['batch_job_id'] ?? 0 ),
'direct_job_ids' => array_slice(array_map('intval', (array) ( $summary['direct_job_ids'] ?? array() )), 0, self::MAX_DURABLE_JOB_IDS),
),
'measured_recovery' => array(
'target_bytes' => (int) ( $budget['target_recovery_bytes'] ?? 0 ),
'target_inodes' => (int) ( $budget['target_recovery_inodes'] ?? 0 ),
'planned_bytes' => (int) ( $selection['planned_measured_recovery_bytes'] ?? 0 ),
'planned_inodes' => (int) ( $selection['planned_measured_recovery_inodes'] ?? 0 ),
'target_met' => ! empty($selection['target_met']),
'reclaimed_bytes' => is_numeric($capacity['reclaimed_bytes'] ?? null) ? (int) $capacity['reclaimed_bytes'] : null,
'reclaimed_inodes' => is_numeric($capacity['reclaimed_inodes'] ?? null) ? (int) $capacity['reclaimed_inodes'] : null,
),
'action_required' => ! empty($result['action_required']),
'action_required_reasons' => $this->compact_reasons($result['action_required_reasons'] ?? array()),
'capacity_evidence' => $this->compact_capacity_evidence($capacity),
);
}

/**
* @param mixed $reasons Candidate reason values.
* @return array<int,string>
*/
private function compact_reasons( mixed $reasons ): array {
$compact = array();
foreach ( array_slice((array) $reasons, 0, self::MAX_DURABLE_REASONS) as $reason ) {
$compact[] = substr((string) $reason, 0, self::MAX_DURABLE_REASON_BYTES);
}
return $compact;
}

/**
* @param array<string,mixed> $capacity Full capacity snapshots.
* @return array<string,mixed>
*/
private function compact_capacity_evidence( array $capacity ): array {
$fields = array( 'filesystem_total_bytes', 'filesystem_free_bytes', 'filesystem_free_inodes', 'workspace_allocated_bytes' );
$compact = array(
'reclaimed_bytes' => is_numeric($capacity['reclaimed_bytes'] ?? null) ? (int) $capacity['reclaimed_bytes'] : null,
'reclaimed_inodes' => is_numeric($capacity['reclaimed_inodes'] ?? null) ? (int) $capacity['reclaimed_inodes'] : null,
'before' => array(),
'after' => array(),
);

foreach ( $fields as $field ) {
$compact['before'][ $field ] = is_numeric($capacity['before'][ $field ] ?? null) ? (int) $capacity['before'][ $field ] : null;
$compact['after'][ $field ] = is_numeric($capacity['after'][ $field ] ?? null) ? (int) $capacity['after'][ $field ] : null;
}

return $compact;
}

/**
Expand Down
109 changes: 109 additions & 0 deletions tests/workspace-disk-emergency-cleanup-durable-result.php
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,109 @@
<?php

declare(strict_types=1);

namespace DataMachine\Core {
class PluginSettings {
public static function get( string $key, mixed $default ): mixed {
return $default;
}
}
}

namespace DataMachine\Engine\AI\System\Tasks {
class SystemTask {
public array $completed = array();

public function completeJob( int $job_id, array $result ): void {
$this->completed[ $job_id ] = $result;
}

public function failJob( int $job_id, string $message ): void {}
}
}

namespace DataMachineCode\Workspace {
class Workspace {
public static array $result;

public function workspace_disk_emergency_cleanup( array $opts ): array {
return self::$result;
}
}
}

namespace {
if ( ! defined('ABSPATH') ) {
define('ABSPATH', __DIR__ . '/fixtures/');
}

$GLOBALS['emergency_cleanup_logs'] = array();
function do_action( string $hook, mixed ...$args ): void {
if ( 'datamachine_log' === $hook ) {
$GLOBALS['emergency_cleanup_logs'][] = $args;
}
}

require_once dirname(__DIR__) . '/inc/Tasks/WorkspaceDiskEmergencyCleanupTask.php';

function durable_result_assert( bool $condition, string $message ): void {
if ( ! $condition ) {
throw new RuntimeException($message);
}
}

$candidates = array();
for ( $index = 0; $index < 128; ++$index ) {
$candidates[] = array(
'handle' => 'repo@candidate-' . $index,
'evidence' => str_repeat('candidate-evidence-', 1024),
);
}

DataMachineCode\Workspace\Workspace::$result = array(
'success' => true,
'triggered' => true,
'dry_run' => true,
'generated_at' => '2026-07-28T00:00:00+00:00',
'disk_budget' => array(
'trigger_reasons' => array( 'filesystem_free_bytes_below_threshold' ),
'target_recovery_bytes' => 123456,
'target_recovery_inodes' => 789,
),
'emergency_plan' => array( 'artifact_candidates' => $candidates, 'worktree_candidates' => $candidates ),
'apply_plan' => array( 'artifact_candidates' => $candidates, 'worktree_candidates' => $candidates ),
'selected_artifact_count' => 128,
'selected_worktree_count' => 0,
'inode_recovery_plan' => array( 'planned_measured_recovery_bytes' => 123456, 'planned_measured_recovery_inodes' => 789, 'target_met' => true ),
'action_required' => true,
'action_required_reasons' => array( 'worktree_deletion_requires_human_approval' ),
'scheduled_summary' => array( 'scheduled_chunks' => 1, 'scheduled_artifact_rows' => 128, 'batch_job_id' => 44, 'direct_job_ids' => array( 45, 46 ) ),
'capacity_evidence' => array(
'reclaimed_bytes' => 900,
'reclaimed_inodes' => 12,
'before' => array( 'filesystem_total_bytes' => 10000, 'filesystem_free_bytes' => 100, 'filesystem_free_inodes' => 10, 'workspace_allocated_bytes' => 9000, 'unbounded' => str_repeat('x', 1024 * 1024) ),
'after' => array( 'filesystem_total_bytes' => 10000, 'filesystem_free_bytes' => 1000, 'filesystem_free_inodes' => 22, 'workspace_allocated_bytes' => 8000, 'unbounded' => str_repeat('x', 1024 * 1024) ),
),
);

$full_size = strlen((string) json_encode(DataMachineCode\Workspace\Workspace::$result, JSON_THROW_ON_ERROR));
durable_result_assert($full_size > 4 * 1024 * 1024, 'Fixture must contain a multi-megabyte explicit workspace report.');

$task = new DataMachineCode\Tasks\WorkspaceDiskEmergencyCleanupTask();
$task->executeTask(42, array( 'dry_run' => true ));
$log_context = $GLOBALS['emergency_cleanup_logs'][0][2];
$completed = $task->completed[42];
$log_size = strlen((string) json_encode($log_context, JSON_THROW_ON_ERROR));
$completed_size = strlen((string) json_encode($completed, JSON_THROW_ON_ERROR));

durable_result_assert($log_size <= 8192, 'Durable log context must remain at or below 8192 serialized bytes.');
durable_result_assert($completed_size <= 8192, 'Completed job result must remain at or below 8192 serialized bytes.');
durable_result_assert(! isset($log_context['report']['emergency_plan']) && ! isset($completed['apply_plan']), 'Durable task records must not persist candidate plans.');
durable_result_assert(128 === $completed['selected_artifact_count'] && array( 45, 46 ) === $completed['scheduled']['direct_job_ids'], 'Durable result must retain selected counts and scheduled job IDs.');
durable_result_assert(array( 'filesystem_free_bytes_below_threshold' ) === $log_context['report']['trigger_reasons'], 'Durable log context must retain trigger reasons.');
durable_result_assert(900 === $completed['measured_recovery']['reclaimed_bytes'] && 12 === $completed['measured_recovery']['reclaimed_inodes'], 'Durable result must retain measured recovery totals.');
durable_result_assert(true === $completed['action_required'] && array( 'worktree_deletion_requires_human_approval' ) === $completed['action_required_reasons'], 'Durable result must retain action-required evidence.');
durable_result_assert(100 === $completed['capacity_evidence']['before']['filesystem_free_bytes'] && 1000 === $completed['capacity_evidence']['after']['filesystem_free_bytes'], 'Durable result must retain compact capacity evidence.');

echo "workspace-disk-emergency-cleanup-durable-result: log={$log_size} completed={$completed_size} bytes\n";
}