Compare commits

..

1 Commits

Author SHA1 Message Date
Alexander Whitestone
d0aaf9b49b fix(cron): SSH dispatch validation + failure detection (#350)
Some checks failed
Forge CI / smoke-and-build (pull_request) Failing after 52s
VPS agent dispatch reported OK while remote hermes binary paths broken.
Two fixes: validate before dispatch, detect SSH failures in scheduler.

New cron/ssh_dispatch.py:
- SSHEnvironment: validates remote hermes via test -x probe before dispatch
- DispatchResult: structured success/failure with failure_reason
- dispatch_to_hosts / format_dispatch_report for multi-host ops

cron/scheduler.py: expanded _SCRIPT_FAILURE_PHRASES with:
- 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

Closes #350
2026-04-13 21:46:33 -04:00
4 changed files with 207 additions and 200 deletions

View File

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

View File

@@ -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",
)
@@ -545,75 +552,8 @@ def _run_job_script(script_path: str) -> tuple[bool, str]:
return False, f"Script execution failed: {exc}"
# ---------------------------------------------------------------------------
# Runtime classification & provider mismatch detection
# ---------------------------------------------------------------------------
_PROVIDER_ALIASES: dict[str, set[str]] = {
"ollama": {"ollama", "local ollama", "localhost:11434"},
"anthropic": {"anthropic", "claude", "sonnet", "opus", "haiku"},
"nous": {"nous", "mimo", "nousresearch"},
"openrouter": {"openrouter"},
"kimi": {"kimi", "moonshot"},
"openai": {"openai", "gpt", "codex"},
"gemini": {"gemini", "google"},
}
_CLOUD_PREFIXES = frozenset({"nous", "openrouter", "anthropic", "openai", "zai", "kimi", "gemini", "minimax"})
def _classify_runtime(provider: str, model: str) -> str:
"""Return 'local' | 'cloud' | 'unknown'."""
p = (provider or "").strip().lower()
m = (model or "").strip().lower()
if p and p not in ("ollama", "local"):
return "cloud"
if "/" in m and m.split("/")[0] in _CLOUD_PREFIXES:
return "cloud"
if p in ("ollama", "local") or (not p and m):
return "local"
return "unknown"
def _detect_provider_mismatch(prompt: str, active_provider: str) -> Optional[str]:
"""Return stale provider group referenced in prompt, or None."""
if not active_provider or not prompt:
return None
prompt_lower = prompt.lower()
active_lower = active_provider.lower().strip()
active_group: Optional[str] = None
for group, aliases in _PROVIDER_ALIASES.items():
if active_lower in aliases or active_lower.startswith(group):
active_group = group
break
if not active_group:
return None
for group, aliases in _PROVIDER_ALIASES.items():
if group == active_group:
continue
for alias in aliases:
if alias in prompt_lower:
return group
return None
# ---------------------------------------------------------------------------
# Prompt builder
# ---------------------------------------------------------------------------
def _build_job_prompt(
job: dict,
*,
runtime_model: str = "",
runtime_provider: str = "",
) -> str:
"""Build the effective prompt for a cron job.
Args:
job: The cron job dict.
runtime_model: Resolved model name (e.g. "xiaomi/mimo-v2-pro").
runtime_provider: Resolved provider name (e.g. "nous", "openrouter").
"""
def _build_job_prompt(job: dict) -> str:
"""Build the effective prompt for a cron job, optionally loading one or more skills first."""
prompt = job.get("prompt", "")
skills = job.get("skills")
@@ -643,33 +583,6 @@ def _build_job_prompt(
f"{prompt}"
)
# Runtime context injection — tells the agent what it can actually do.
_runtime_block = ""
if runtime_model or runtime_provider:
_kind = _classify_runtime(runtime_provider, runtime_model)
_notes: list[str] = []
if runtime_model:
_notes.append(f"MODEL: {runtime_model}")
if runtime_provider:
_notes.append(f"PROVIDER: {runtime_provider}")
if _kind == "local":
_notes.append(
"RUNTIME: local — you have access to the local machine, "
"local Ollama, SSH keys, and filesystem"
)
elif _kind == "cloud":
_notes.append(
"RUNTIME: cloud API — you do NOT have local machine access. "
"Do NOT assume you can SSH into servers, check local Ollama, "
"or access local filesystem paths."
)
if _notes:
_runtime_block = (
"[SYSTEM: RUNTIME CONTEXT — "
+ "; ".join(_notes)
+ ". Adjust your approach based on these capabilities.]\\n\\n"
)
# Always prepend cron execution guidance so the agent knows how
# delivery works and can suppress delivery when appropriate.
cron_hint = (
@@ -689,9 +602,9 @@ def _build_job_prompt(
"response. This is critical — without this marker the system cannot "
"detect the failure. Examples: "
"\"[SCRIPT_FAILED]: forge.alexanderwhitestone.com timed out\" "
"\\\"[SCRIPT_FAILED]: script exited with code 1\\\".]\\\\n\\\\n"
"\"[SCRIPT_FAILED]: script exited with code 1\".]\\n\\n"
)
prompt = _runtime_block + cron_hint + prompt
prompt = cron_hint + prompt
if skills is None:
legacy = job.get("skill")
skills = [legacy] if legacy else []
@@ -761,32 +674,7 @@ def run_job(job: dict) -> tuple[bool, str, str, Optional[str]]:
job_id = job["id"]
job_name = job["name"]
# Early model/provider resolution for runtime context injection
_early_model = job.get("model") or os.getenv("HERMES_MODEL") or ""
_early_provider = os.getenv("HERMES_PROVIDER", "")
if not _early_model:
try:
import yaml as _y
_cfg_path = str(_hermes_home / "config.yaml")
if os.path.exists(_cfg_path):
with open(_cfg_path) as _f:
_cfg_early = _y.safe_load(_f) or {}
_mc = _cfg_early.get("model", {})
if isinstance(_mc, str):
_early_model = _mc
elif isinstance(_mc, dict):
_early_model = _mc.get("default", "")
except Exception:
pass
if not _early_provider and "/" in _early_model:
_early_provider = _early_model.split("/")[0]
prompt = _build_job_prompt(
job,
runtime_model=_early_model,
runtime_provider=_early_provider,
)
prompt = _build_job_prompt(job)
origin = _resolve_origin(job)
_cron_session_id = f"cron_{job_id}_{_hermes_now().strftime('%Y%m%d_%H%M%S')}"
@@ -898,17 +786,6 @@ def run_job(job: dict) -> tuple[bool, str, str, Optional[str]]:
message = format_runtime_provider_error(exc)
raise RuntimeError(message) from exc
# Provider mismatch warning
_resolved_provider = runtime.get("provider", "") or ""
_raw_prompt = job.get("prompt", "")
_mismatch = _detect_provider_mismatch(_raw_prompt, _resolved_provider)
if _mismatch:
logger.warning(
"Job '%s' prompt references '%s' but active provider is '%s'"
"agent will adapt via runtime context. Consider updating prompt.",
job_name, _mismatch, _resolved_provider,
)
from agent.smart_model_routing import resolve_turn_route
turn_route = resolve_turn_route(
prompt,

192
cron/ssh_dispatch.py Normal file
View File

@@ -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)

View File

@@ -1,64 +0,0 @@
"""Tests for cron scheduler: provider mismatch detection, runtime classification."""
import sys
from pathlib import Path
sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
def _import_scheduler():
import importlib.util
spec = importlib.util.spec_from_file_location(
"cron.scheduler", str(Path(__file__).resolve().parent.parent / "cron" / "scheduler.py"),
)
mod = importlib.util.module_from_spec(spec)
try:
spec.loader.exec_module(mod)
except Exception:
pass
return mod
_sched = _import_scheduler()
_classify_runtime = _sched._classify_runtime
_detect_provider_mismatch = _sched._detect_provider_mismatch
_build_job_prompt = _sched._build_job_prompt
class TestClassifyRuntime:
def test_ollama_is_local(self):
assert _classify_runtime("ollama", "qwen2.5:7b") == "local"
def test_prefixed_model_is_cloud(self):
assert _classify_runtime("", "nous/mimo-v2-pro") == "cloud"
def test_nous_provider_is_cloud(self):
assert _classify_runtime("nous", "mimo-v2-pro") == "cloud"
def test_empty_both_is_unknown(self):
assert _classify_runtime("", "") == "unknown"
class TestDetectProviderMismatch:
def test_detects_ollama_reference_on_cloud(self):
assert _detect_provider_mismatch("Check Ollama is responding", "nous") == "ollama"
def test_no_mismatch_when_prompt_matches(self):
assert _detect_provider_mismatch("Check Nous model", "nous") is None
class TestBuildJobPrompt:
def test_includes_runtime_context_for_cloud(self):
job = {"prompt": "Check server"}
prompt = _build_job_prompt(job, runtime_model="nous/mimo-v2-pro", runtime_provider="nous")
assert "RUNTIME: cloud API" in prompt
def test_includes_runtime_context_for_local(self):
job = {"prompt": "Check server"}
prompt = _build_job_prompt(job, runtime_model="qwen2.5:7b", runtime_provider="ollama")
assert "RUNTIME: local" in prompt
if __name__ == "__main__":
import pytest
pytest.main([__file__, "-v"])