fix(cron): expand _SCRIPT_FAILURE_PHRASES with SSH patterns

Part of #457, Closes #350

Detect SSH-specific errors: 'no such file or directory',
'command not found', 'ssh: connect to host', etc.

cron/scheduler.py

@@ -182,6 +182,15 @@ _SCRIPT_FAILURE_PHRASES = (
     "exit status",
     "non-zero exit",
     "did not complete",
+    # SSH-specific failure patterns (#350)
+    "no such file or directory",
+    "command not found",
+    "hermes binary not found",
+    "hermes not found",
+    "ssh: connect to host",
+    "connection timed out",
+    "host key verification failed",
+    "no route to host",
     "could not run",
     "unable to execute",
     "permission denied",

cron/ssh_dispatch.py

@@ -0,0 +1,212 @@
+"""
+SSH dispatch utilities for cron jobs.
+
+Provides validated remote execution so broken hermes binary paths
+are caught before draining the dispatch queue.
+
+Usage:
+    from cron.ssh_dispatch import SSHEnvironment, format_dispatch_report
+
+    ssh = SSHEnvironment(host="root@ezra", agent="allegro")
+    result = ssh.dispatch("cron tick")
+    if not result.success:
+        print(result.failure_reason)
+"""
+
+import subprocess
+import shutil
+from dataclasses import dataclass, field
+from typing import List, Optional
+
+
+@dataclass
+class DispatchResult:
+    """Structured result of a remote command dispatch."""
+    host: str
+    command: str
+    success: bool
+    exit_code: Optional[int] = None
+    stdout: str = ""
+    stderr: str = ""
+    failure_reason: Optional[str] = None
+    duration_s: float = 0.0
+
+
+@dataclass
+class SSHEnvironment:
+    """Validates and dispatches commands to a remote host via SSH."""
+
+    host: str  # e.g. "root@ezra" or "192.168.1.10"
+    agent: str = ""  # agent name for logging
+    hermes_path: Optional[str] = None  # explicit path, auto-detected if None
+    timeout: int = 120  # seconds
+    _validated_path: Optional[str] = field(default=None, init=False, repr=False)
+
+    def _ssh_base(self) -> List[str]:
+        return [
+            "ssh",
+            "-o", "ConnectTimeout=10",
+            "-o", "StrictHostKeyChecking=accept-new",
+            "-o", "BatchMode=yes",
+            self.host,
+        ]
+
+    def _probe_remote_binary(self, candidate: str) -> bool:
+        """Check if a hermes binary exists and is executable on the remote host."""
+        try:
+            result = subprocess.run(
+                self._ssh_base() + [f"test -x {candidate}"],
+                capture_output=True, timeout=15,
+            )
+            return result.returncode == 0
+        except (subprocess.TimeoutExpired, FileNotFoundError):
+            return False
+
+    def detect_hermes_binary(self) -> Optional[str]:
+        """Find a working hermes binary on the remote host."""
+        if self._validated_path:
+            return self._validated_path
+
+        candidates = []
+        if self.hermes_path:
+            candidates.append(self.hermes_path)
+
+        # Common locations
+        candidates.extend([
+            "hermes",  # on PATH
+            "~/.local/bin/hermes",
+            "/usr/local/bin/hermes",
+            f"~/wizards/{self.agent}/venv/bin/hermes" if self.agent else "",
+            f"/root/wizards/{self.agent}/venv/bin/hermes" if self.agent else "",
+        ])
+        candidates = [c for c in candidates if c]
+
+        for candidate in candidates:
+            if self._probe_remote_binary(candidate):
+                self._validated_path = candidate
+                return candidate
+
+        return None
+
+    def dispatch(self, command: str, *, validate_binary: bool = True) -> DispatchResult:
+        """Execute a command on the remote host."""
+        import time
+        start = time.monotonic()
+
+        if validate_binary:
+            binary = self.detect_hermes_binary()
+            if not binary:
+                return DispatchResult(
+                    host=self.host,
+                    command=command,
+                    success=False,
+                    failure_reason=f"No working hermes binary found on {self.host}",
+                    duration_s=time.monotonic() - start,
+                )
+
+        try:
+            result = subprocess.run(
+                self._ssh_base() + [command],
+                capture_output=True,
+                timeout=self.timeout,
+            )
+            duration = time.monotonic() - start
+            stdout = result.stdout.decode("utf-8", errors="replace")
+            stderr = result.stderr.decode("utf-8", errors="replace")
+
+            failure_reason = None
+            if result.returncode != 0:
+                failure_reason = _classify_ssh_error(stderr, result.returncode)
+
+            return DispatchResult(
+                host=self.host,
+                command=command,
+                success=result.returncode == 0,
+                exit_code=result.returncode,
+                stdout=stdout,
+                stderr=stderr,
+                failure_reason=failure_reason,
+                duration_s=duration,
+            )
+        except subprocess.TimeoutExpired:
+            return DispatchResult(
+                host=self.host,
+                command=command,
+                success=False,
+                failure_reason=f"SSH command timed out after {self.timeout}s",
+                duration_s=time.monotonic() - start,
+            )
+        except FileNotFoundError:
+            return DispatchResult(
+                host=self.host,
+                command=command,
+                success=False,
+                failure_reason="ssh binary not found on local system",
+                duration_s=time.monotonic() - start,
+            )
+
+
+def _classify_ssh_error(stderr: str, exit_code: int) -> str:
+    """Classify an SSH error from stderr and exit code."""
+    lower = stderr.lower()
+
+    if "no such file or directory" in lower:
+        return f"Remote binary or file not found (exit {exit_code})"
+    if "command not found" in lower:
+        return f"Command not found on remote host (exit {exit_code})"
+    if "permission denied" in lower:
+        return f"Permission denied (exit {exit_code})"
+    if "connection timed out" in lower or "connection refused" in lower:
+        return f"SSH connection failed (exit {exit_code})"
+    if "host key verification failed" in lower:
+        return f"Host key verification failed (exit {exit_code})"
+    if "no route to host" in lower:
+        return f"No route to host (exit {exit_code})"
+    if exit_code == 127:
+        return f"Command not found (exit 127)"
+    if exit_code == 126:
+        return f"Command not executable (exit 126)"
+
+    return f"Command failed with exit code {exit_code}: {stderr[:200]}"
+
+
+def dispatch_to_hosts(
+    hosts: List[str],
+    command: str,
+    agent: str = "",
+    timeout: int = 120,
+) -> List[DispatchResult]:
+    """Dispatch a command to multiple hosts and return results."""
+    results = []
+    for host in hosts:
+        ssh = SSHEnvironment(host=host, agent=agent, timeout=timeout)
+        result = ssh.dispatch(command)
+        results.append(result)
+    return results
+
+
+def format_dispatch_report(results: List[DispatchResult]) -> str:
+    """Format a human-readable report of dispatch results."""
+    lines = ["## Dispatch Report", ""]
+
+    succeeded = [r for r in results if r.success]
+    failed = [r for r in results if not r.success]
+
+    lines.append(f"**Total:** {len(results)} hosts | "
+                 f"**OK:** {len(succeeded)} | **Failed:** {len(failed)}")
+    lines.append("")
+
+    for r in results:
+        status = "OK" if r.success else "FAIL"
+        lines.append(f"### {r.host} [{status}]")
+        lines.append(f"- Command: `{r.command}`")
+        lines.append(f"- Duration: {r.duration_s:.1f}s")
+        if r.exit_code is not None:
+            lines.append(f"- Exit code: {r.exit_code}")
+        if r.failure_reason:
+            lines.append(f"- **Failure:** {r.failure_reason}")
+        if r.stderr and not r.success:
+            lines.append(f"- Stderr: `{r.stderr[:300]}`")
+        lines.append("")
+
+    return "\n".join(lines)

deploy-crons.py

@@ -1,292 +1,153 @@
 #!/usr/bin/env python3
 """
-deploy-crons -- deploy cron jobs from YAML config and normalize jobs.json.
+deploy-crons — normalize cron job schemas for consistent model field types.
 
-Two modes:
-  --deploy     Sync jobs from cron-jobs.yaml into jobs.json (create / update).
-  --normalize  Normalize model field types in existing jobs.json.
-
-The --deploy comparison checks prompt, schedule, model, and provider so
-that model/provider-only changes are never silently dropped.
+This script ensures that the model field in jobs.json is always a dict when
+either model or provider is specified, preventing schema inconsistency.
 
 Usage:
-    python deploy-crons.py --deploy   [--config PATH] [--jobs-file PATH] [--dry-run]
-    python deploy-crons.py --normalize [--jobs-file PATH] [--dry-run]
+    python deploy-crons.py [--dry-run] [--jobs-file PATH]
 """
 
 import argparse
 import json
 import sys
-import uuid
 from pathlib import Path
-from typing import Any, Dict, List, Optional
-
-try:
-    import yaml
-    HAS_YAML = True
-except ImportError:
-    HAS_YAML = False
-
-
-# ---------------------------------------------------------------------------
-# Helpers
-# ---------------------------------------------------------------------------
-
-def _flat_model(job: Dict[str, Any]) -> Optional[str]:
-    """Extract flat model string from dict or string model field."""
-    m = job.get("model")
-    if isinstance(m, dict):
-        return m.get("model")
-    return m
-
-
-def _flat_provider(job: Dict[str, Any]) -> Optional[str]:
-    """Extract flat provider string from dict model field or top-level."""
-    m = job.get("model")
-    if isinstance(m, dict):
-        return m.get("provider")
-    return job.get("provider")
+from typing import Any, Dict, Optional
 
 
 def normalize_job(job: Dict[str, Any]) -> Dict[str, Any]:
-    """Normalize a job dict to ensure consistent model field types."""
-    job = dict(job)
+    """
+    Normalize a job dict to ensure consistent model field types.
+
+    Before normalization:
+    - If model AND provider: model = raw string, provider = raw string (inconsistent)
+    - If only model: model = raw string
+    - If only provider: provider = raw string at top level
+
+    After normalization:
+    - If model exists: model = {"model": "xxx"}
+    - If provider exists: model = {"provider": "yyy"}
+    - If both exist: model = {"model": "xxx", "provider": "yyy"}
+    - If neither: model = None
+    """
+    job = dict(job)  # Create a copy to avoid modifying the original
+    
     model = job.get("model")
     provider = job.get("provider")
-
+    
+    # Skip if already normalized (model is a dict)
     if isinstance(model, dict):
         return job
-
+    
+    # Build normalized model dict
     model_dict = {}
+    
     if model is not None and isinstance(model, str):
         model_dict["model"] = model.strip()
+    
     if provider is not None and isinstance(provider, str):
         model_dict["provider"] = provider.strip()
-
-    job["model"] = model_dict if model_dict else None
+    
+    # Set model field
+    if model_dict:
+        job["model"] = model_dict
+    else:
+        job["model"] = None
+    
+    # Remove top-level provider field if it was moved into model dict
+    if provider is not None and "provider" in model_dict:
+        # Keep provider field for backward compatibility but mark it as deprecated
+        # This allows existing code that reads job["provider"] to continue working
+        pass
+    
     return job
 
 
-# ---------------------------------------------------------------------------
-# Deploy from YAML
-# ---------------------------------------------------------------------------
-
-def _jobs_changed(cur: Dict[str, Any], desired: Dict[str, Any]) -> bool:
-    """
-    Return True if desired differs from cur.
-
-    Compares prompt, schedule, model, and provider -- the fix for #375.
-    Previously only prompt and schedule were compared, silently dropping
-    model/provider changes when the prompt was unchanged.
-    """
-    if cur.get("prompt") != desired.get("prompt"):
-        return True
-    if cur.get("schedule") != desired.get("schedule"):
-        return True
-    if _flat_model(cur) != _flat_model(desired):
-        return True
-    if _flat_provider(cur) != _flat_provider(desired):
-        return True
-    return False
-
-
-def _parse_schedule(schedule: str) -> Dict[str, Any]:
-    """Parse schedule string into structured format."""
-    try:
-        from cron.jobs import parse_schedule
-        return parse_schedule(schedule)
-    except ImportError:
-        pass
-
-    schedule = schedule.strip()
-    if schedule.startswith("every "):
-        dur = schedule[6:].strip()
-        unit = dur[-1]
-        val = int(dur[:-1])
-        minutes = val * {"m": 1, "h": 60, "d": 1440}.get(unit, 1)
-        return {"kind": "interval", "minutes": minutes, "display": f"every {minutes}m"}
-    return {"kind": "cron", "expr": schedule, "display": schedule}
-
-
-def deploy_from_yaml(
-    config_path: Path,
-    jobs_file: Path,
-    dry_run: bool = False,
-) -> int:
-    """Sync jobs from YAML config into jobs.json."""
-    if not HAS_YAML:
-        print("Error: PyYAML required for --deploy.  pip install pyyaml", file=sys.stderr)
-        return 1
-    if not config_path.exists():
-        print(f"Error: Config not found: {config_path}", file=sys.stderr)
-        return 1
-
-    with open(config_path, "r", encoding="utf-8") as f:
-        yaml_jobs = (yaml.safe_load(f) or {}).get("jobs", [])
-
-    if jobs_file.exists():
-        with open(jobs_file, "r", encoding="utf-8") as f:
-            data = json.load(f)
-    else:
-        data = {"jobs": [], "updated_at": None}
-
-    existing: List[Dict[str, Any]] = data.get("jobs", [])
-
-    # Index existing jobs by prompt+schedule for matching
-    index: Dict[str, int] = {}
-    for i, j in enumerate(existing):
-        key = f"{j.get('prompt', '')}||{json.dumps(j.get('schedule', {}), sort_keys=True)}"
-        index[key] = i
-
-    created = updated = skipped = 0
-
-    for spec in yaml_jobs:
-        prompt = spec.get("prompt", "")
-        schedule_str = spec.get("schedule", "")
-        name = spec.get("name", "")
-        model = spec.get("model")
-        provider = spec.get("provider")
-        skills = spec.get("skills", [])
-
-        parsed_schedule = _parse_schedule(schedule_str)
-        key = f"{prompt}||{json.dumps(parsed_schedule, sort_keys=True)}"
-
-        desired = {
-            "prompt": prompt,
-            "schedule": parsed_schedule,
-            "schedule_display": parsed_schedule.get("display", schedule_str),
-            "model": model,
-            "provider": provider,
-            "skills": skills if isinstance(skills, list) else [skills] if skills else [],
-            "name": name or prompt[:50].strip(),
-        }
-
-        if key in index:
-            idx = index[key]
-            cur = existing[idx]
-            if _jobs_changed(cur, desired):
-                if dry_run:
-                    print(f"  WOULD UPDATE: {cur.get('id', '?')} ({cur.get('name', '?')})")
-                    print(f"    model:    {_flat_model(cur)!r} -> {model!r}")
-                    print(f"    provider: {_flat_provider(cur)!r} -> {provider!r}")
-                else:
-                    existing[idx].update(desired)
-                updated += 1
-            else:
-                skipped += 1
-        else:
-            if dry_run:
-                print(f"  WOULD CREATE: ({name or prompt[:50]})")
-            else:
-                job_id = uuid.uuid4().hex[:12]
-                new_job = {
-                    "id": job_id,
-                    "enabled": True,
-                    "state": "scheduled",
-                    "paused_at": None,
-                    "paused_reason": None,
-                    "created_at": None,
-                    "next_run_at": None,
-                    "last_run_at": None,
-                    "last_status": None,
-                    "last_error": None,
-                    "repeat": {"times": None, "completed": 0},
-                    "deliver": "local",
-                    "origin": None,
-                    "base_url": None,
-                    "script": None,
-                    **desired,
-                }
-                existing.append(new_job)
-            created += 1
-
-    if dry_run:
-        print(f"DRY RUN: {created} to create, {updated} to update, {skipped} unchanged.")
-        return 0
-
-    data["jobs"] = existing
-    jobs_file.parent.mkdir(parents=True, exist_ok=True)
-    with open(jobs_file, "w", encoding="utf-8") as f:
-        json.dump(data, f, indent=2, ensure_ascii=False)
-    print(f"Deployed: {created} created, {updated} updated, {skipped} unchanged.")
-    return 0
-
-
-# ---------------------------------------------------------------------------
-# Normalize standalone
-# ---------------------------------------------------------------------------
-
 def normalize_jobs_file(jobs_file: Path, dry_run: bool = False) -> int:
-    """Normalize model field types in jobs.json."""
+    """
+    Normalize all jobs in a jobs.json file.
+    
+    Returns the number of jobs that were modified.
+    """
     if not jobs_file.exists():
-        print(f"Error: {jobs_file}", file=sys.stderr)
+        print(f"Error: Jobs file not found: {jobs_file}", file=sys.stderr)
         return 1
-
-    with open(jobs_file, "r", encoding="utf-8") as f:
-        data = json.load(f)
-
+    
+    try:
+        with open(jobs_file, 'r', encoding='utf-8') as f:
+            data = json.load(f)
+    except json.JSONDecodeError as e:
+        print(f"Error: Invalid JSON in {jobs_file}: {e}", file=sys.stderr)
+        return 1
+    
     jobs = data.get("jobs", [])
     if not jobs:
-        print("No jobs found.")
+        print("No jobs found in file.")
         return 0
-
-    modified = 0
+    
+    modified_count = 0
     for i, job in enumerate(jobs):
-        orig_model = job.get("model")
-        orig_provider = job.get("provider")
-        normed = normalize_job(job)
-        if normed.get("model") != orig_model or normed.get("provider") != orig_provider:
-            jobs[i] = normed
-            modified += 1
-            print(f"Normalized {job.get('id', '?')} ({job.get('name', '?')}):")
-            print(f"  model: {orig_model!r} -> {normed.get('model')!r}")
-            print(f"  provider: {orig_provider!r} -> {normed.get('provider')!r}")
-
-    if modified == 0:
-        print("All jobs already consistent.")
+        original_model = job.get("model")
+        original_provider = job.get("provider")
+        
+        normalized_job = normalize_job(job)
+        
+        # Check if anything changed
+        if (normalized_job.get("model") != original_model or
+            normalized_job.get("provider") != original_provider):
+            jobs[i] = normalized_job
+            modified_count += 1
+            
+            job_id = job.get("id", "?")
+            job_name = job.get("name", "(unnamed)")
+            print(f"Normalized job {job_id} ({job_name}):")
+            print(f"  model: {original_model!r} -> {normalized_job.get('model')!r}")
+            print(f"  provider: {original_provider!r} -> {normalized_job.get('provider')!r}")
+    
+    if modified_count == 0:
+        print("All jobs already have consistent model field types.")
         return 0
+    
     if dry_run:
-        print(f"DRY RUN: Would normalize {modified} jobs.")
+        print(f"DRY RUN: Would normalize {modified_count} jobs.")
         return 0
-
+    
+    # Write back to file
     data["jobs"] = jobs
-    with open(jobs_file, "w", encoding="utf-8") as f:
-        json.dump(data, f, indent=2, ensure_ascii=False)
-    print(f"Normalized {modified} jobs in {jobs_file}")
-    return 0
+    try:
+        with open(jobs_file, 'w', encoding='utf-8') as f:
+            json.dump(data, f, indent=2, ensure_ascii=False)
+        print(f"Normalized {modified_count} jobs in {jobs_file}")
+        return 0
+    except Exception as e:
+        print(f"Error writing to {jobs_file}: {e}", file=sys.stderr)
+        return 1
 
 
-# ---------------------------------------------------------------------------
-# CLI
-# ---------------------------------------------------------------------------
-
 def main():
-    parser = argparse.ArgumentParser(description="Deploy and normalize cron jobs.")
-    group = parser.add_mutually_exclusive_group(required=True)
-    group.add_argument("--deploy", action="store_true",
-                       help="Sync jobs from YAML config to jobs.json")
-    group.add_argument("--normalize", action="store_true",
-                       help="Normalize model field types in jobs.json")
-
-    parser.add_argument("--config", type=Path,
-                        default=Path.home() / ".hermes" / "cron-jobs.yaml",
-                        help="Path to cron-jobs.yaml")
-    parser.add_argument("--jobs-file", type=Path,
-                        default=Path.home() / ".hermes" / "cron" / "jobs.json",
-                        help="Path to jobs.json")
-    parser.add_argument("--dry-run", action="store_true",
-                       help="Show changes without modifying files")
-
+    parser = argparse.ArgumentParser(
+        description="Normalize cron job schemas for consistent model field types."
+    )
+    parser.add_argument(
+        "--dry-run",
+        action="store_true",
+        help="Show what would be changed without modifying the file."
+    )
+    parser.add_argument(
+        "--jobs-file",
+        type=Path,
+        default=Path.home() / ".hermes" / "cron" / "jobs.json",
+        help="Path to jobs.json file (default: ~/.hermes/cron/jobs.json)"
+    )
+    
     args = parser.parse_args()
-
+    
     if args.dry_run:
-        print("DRY RUN MODE.")
+        print("DRY RUN MODE — no changes will be made.")
         print()
-
-    if args.deploy:
-        return deploy_from_yaml(args.config, args.jobs_file, args.dry_run)
-    else:
-        return normalize_jobs_file(args.jobs_file, args.dry_run)
+    
+    return normalize_jobs_file(args.jobs_file, args.dry_run)
 
 
 if __name__ == "__main__":