fix(cron): SSH dispatch validation, failure detection, broken import (#350, #541)

Three fixes:

1. New cron/ssh_dispatch.py — SSH dispatch utilities
   - SSHEnvironment: validates remote hermes binary via test -x probe
     before dispatch; returns DispatchResult with success=False on
     broken paths instead of silently reporting OK
   - DispatchResult: structured success/failure with failure_reason
   - dispatch_to_hosts / format_dispatch_report for multi-host ops

2. cron/scheduler.py — expanded _SCRIPT_FAILURE_PHRASES
   Added 7 SSH-specific patterns:
   - 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

3. cron/__init__.py — fixed broken import (#541)
   Removed stale ModelContextError and CRON_MIN_CONTEXT_TOKENS imports
   that don't exist in scheduler.py, causing ImportError on any
   'from cron import ...' statement.

Closes #350, Closes #541

cron/__init__.py

@@ -26,7 +26,7 @@ from cron.jobs import (
     trigger_job,
     JOBS_FILE,
 )
-from cron.scheduler import tick, ModelContextError, CRON_MIN_CONTEXT_TOKENS
+from cron.scheduler import tick
 
 __all__ = [
     "create_job",
@@ -39,6 +39,4 @@ __all__ = [
     "trigger_job",
     "tick",
     "JOBS_FILE",
-    "ModelContextError",
-    "CRON_MIN_CONTEXT_TOKENS",
 ]

cron/scheduler.py

@@ -186,7 +186,14 @@ _SCRIPT_FAILURE_PHRASES = (
     "unable to execute",
     "permission denied",
     "no such file",
+    "no such file or directory",
+    "command not found",
+    "hermes binary not found",
+    "hermes not found",
     "traceback",
+    "ssh: connect to host",
+    "connection timed out",
+    "host key verification failed",
 )
 
 

cron/ssh_dispatch.py

@@ -0,0 +1,192 @@
+"""SSH dispatch utilities for VPS agent operations.
+
+Provides validated SSH execution with proper failure detection.
+Used by cron jobs that dispatch work to remote VPS agents.
+"""
+
+from __future__ import annotations
+
+import logging
+import os
+import subprocess
+import time
+from typing import Optional
+
+logger = logging.getLogger(__name__)
+
+_SSH_TIMEOUT = int(os.getenv("HERMES_SSH_TIMEOUT", "30"))
+
+_DEFAULT_HERMES_PATHS = [
+    "/root/wizards/{agent}/venv/bin/hermes",
+    "/root/.local/bin/hermes",
+    "/usr/local/bin/hermes",
+    "~/.local/bin/hermes",
+    "hermes",
+]
+
+
+class DispatchResult:
+    """Structured result of a dispatch operation."""
+
+    __slots__ = (
+        "success", "host", "command", "exit_code",
+        "stdout", "stderr", "error", "duration_ms", "hermes_path",
+    )
+
+    def __init__(
+        self, success: bool, host: str, command: str,
+        exit_code: int = -1, stdout: str = "", stderr: str = "",
+        error: str = "", duration_ms: int = 0, hermes_path: str = "",
+    ):
+        self.success = success
+        self.host = host
+        self.command = command
+        self.exit_code = exit_code
+        self.stdout = stdout
+        self.stderr = stderr
+        self.error = error
+        self.duration_ms = duration_ms
+        self.hermes_path = hermes_path
+
+    def to_dict(self) -> dict:
+        return {
+            "success": self.success, "host": self.host,
+            "exit_code": self.exit_code, "error": self.error,
+            "duration_ms": self.duration_ms, "hermes_path": self.hermes_path,
+            "stderr_tail": self.stderr[-200:] if self.stderr else "",
+        }
+
+    @property
+    def failure_reason(self) -> str:
+        if self.success:
+            return ""
+        if self.error:
+            return self.error
+        if "No such file" in self.stderr or "command not found" in self.stderr:
+            return f"Hermes binary not found on {self.host}"
+        if self.exit_code != 0:
+            return f"Remote command exited {self.exit_code}"
+        return "Dispatch failed (unknown reason)"
+
+
+class SSHEnvironment:
+    """Validated SSH execution environment for VPS agent dispatch."""
+
+    def __init__(
+        self, host: str, agent: str = "", ssh_key: str = "",
+        ssh_port: int = 22, timeout: int = _SSH_TIMEOUT,
+        hermes_path: str = "",
+    ):
+        self.host = host
+        self.agent = agent
+        self.ssh_key = ssh_key
+        self.ssh_port = ssh_port
+        self.timeout = timeout
+        self.hermes_path = hermes_path
+        self._validated_path: str = ""
+
+    def _ssh_base_cmd(self) -> list[str]:
+        cmd = ["ssh", "-o", "StrictHostKeyChecking=accept-new"]
+        cmd.extend(["-o", "ConnectTimeout=10", "-o", "BatchMode=yes"])
+        if self.ssh_key:
+            cmd.extend(["-i", self.ssh_key])
+        if self.ssh_port != 22:
+            cmd.extend(["-p", str(self.ssh_port)])
+        cmd.append(self.host)
+        return cmd
+
+    def _resolve_hermes_paths(self) -> list[str]:
+        if self.hermes_path:
+            return [self.hermes_path]
+        return [t.format(agent=self.agent) if "{agent}" in t else t for t in _DEFAULT_HERMES_PATHS]
+
+    def validate_remote_hermes_path(self) -> str:
+        """Probe remote host for a working hermes binary. Returns path or raises."""
+        if self._validated_path:
+            return self._validated_path
+        for path in self._resolve_hermes_paths():
+            try:
+                result = subprocess.run(
+                    self._ssh_base_cmd() + [f"test -x {path} && echo OK || echo MISSING"],
+                    capture_output=True, text=True, timeout=self.timeout,
+                )
+                if result.returncode == 0 and "OK" in (result.stdout or ""):
+                    logger.info("SSH %s: hermes validated at %s", self.host, path)
+                    self._validated_path = path
+                    return path
+            except subprocess.TimeoutExpired:
+                logger.warning("SSH %s: timeout probing %s", self.host, path)
+            except Exception as exc:
+                logger.debug("SSH %s: probe %s failed: %s", self.host, path, exc)
+        raise RuntimeError(
+            f"No working hermes binary found on {self.host}. "
+            f"Checked: {', '.join(self._resolve_hermes_paths())}."
+        )
+
+    def execute_command(self, remote_cmd: str) -> DispatchResult:
+        """Execute a command on the remote host."""
+        t0 = time.monotonic()
+        try:
+            result = subprocess.run(
+                self._ssh_base_cmd() + [remote_cmd],
+                capture_output=True, text=True, timeout=self.timeout,
+            )
+            elapsed = int((time.monotonic() - t0) * 1000)
+            stderr = (result.stderr or "").strip()
+            stdout = (result.stdout or "").strip()
+            if result.returncode != 0:
+                return DispatchResult(
+                    success=False, host=self.host, command=remote_cmd,
+                    exit_code=result.returncode, stdout=stdout, stderr=stderr,
+                    error=stderr.split("\n")[0] if stderr else f"exit code {result.returncode}",
+                    duration_ms=elapsed,
+                )
+            return DispatchResult(success=True, host=self.host, command=remote_cmd,
+                                  exit_code=0, stdout=stdout, stderr=stderr, duration_ms=elapsed)
+        except subprocess.TimeoutExpired:
+            return DispatchResult(success=False, host=self.host, command=remote_cmd,
+                                  error=f"SSH timed out after {self.timeout}s",
+                                  duration_ms=int((time.monotonic() - t0) * 1000))
+        except Exception as exc:
+            return DispatchResult(success=False, host=self.host, command=remote_cmd,
+                                  error=str(exc), duration_ms=int((time.monotonic() - t0) * 1000))
+
+    def dispatch(self, hermes_args: str, validate: bool = True) -> DispatchResult:
+        """Dispatch a hermes command. Only success=True if command actually ran."""
+        if validate:
+            try:
+                hermes_path = self.validate_remote_hermes_path()
+            except RuntimeError as exc:
+                return DispatchResult(success=False, host=self.host,
+                                      command=f"hermes {hermes_args}",
+                                      error=str(exc), hermes_path="(not found)")
+        else:
+            hermes_path = self.hermes_path or "hermes"
+        result = self.execute_command(f"{hermes_path} {hermes_args}")
+        result.hermes_path = hermes_path
+        return result
+
+
+def dispatch_to_hosts(hosts: list[str], hermes_args: str, **kwargs) -> dict[str, DispatchResult]:
+    """Dispatch to multiple hosts. Returns host -> DispatchResult."""
+    results: dict[str, DispatchResult] = {}
+    for host in hosts:
+        ssh = SSHEnvironment(host=host, **kwargs)
+        results[host] = ssh.dispatch(hermes_args)
+    return results
+
+
+def format_dispatch_report(results: dict[str, DispatchResult]) -> str:
+    """Format dispatch results as a human-readable report."""
+    ok = [r for r in results.values() if r.success]
+    failed = [r for r in results.values() if not r.success]
+    lines = [f"Dispatch report: {len(ok)} OK, {len(failed)} failed", ""]
+    for host, r in results.items():
+        s = "OK" if r.success else f"FAILED -- {r.failure_reason}"
+        lines.append(f"  {host}: {s}" + (f" ({r.duration_ms}ms)" if r.duration_ms else ""))
+    if failed:
+        lines += ["", "Failed dispatches:"]
+        for host, r in results.items():
+            if not r.success:
+                lines.append(f"  {host}: {r.failure_reason}")
+    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__":