A few more tweaks for an initial deploy

3 files changed, 77 insertions, 68 deletions

diff --git a/stator/management/commands/runstator.py b/stator/management/commands/runstator.py
index eaa2585..3030960 100644
--- a/stator/management/commands/runstator.py
+++ b/stator/management/commands/runstator.py
@@ -10,7 +10,7 @@ from stator.runner import StatorRunner
 
 
 class Command(BaseCommand):
-    help = "Runs a Stator runner for a short period"
+    help = "Runs a Stator runner"
 
     def add_arguments(self, parser):
         parser.add_argument(
@@ -20,9 +20,30 @@ class Command(BaseCommand):
             default=30,
             help="How many tasks to run at once",
         )
+        parser.add_argument(
+            "--liveness-file",
+            type=str,
+            default=None,
+            help="A file to touch at least every 30 seconds to say the runner is alive",
+        )
+        parser.add_argument(
+            "--schedule-interval",
+            "-s",
+            type=int,
+            default=30,
+            help="How often to run cleaning and scheduling",
+        )
         parser.add_argument("model_labels", nargs="*", type=str)
 
-    def handle(self, model_labels: List[str], concurrency: int, *args, **options):
+    def handle(
+        self,
+        model_labels: List[str],
+        concurrency: int,
+        liveness_file: str,
+        schedule_interval: int,
+        *args,
+        **options
+    ):
         # Cache system config
         Config.system = Config.load_system()
         # Resolve the models list into names
@@ -34,5 +55,10 @@ class Command(BaseCommand):
             models = StatorModel.subclasses
         print("Running for models: " + " ".join(m._meta.label_lower for m in models))
         # Run a runner
-        runner = StatorRunner(models, concurrency=concurrency)
+        runner = StatorRunner(
+            models,
+            concurrency=concurrency,
+            liveness_file=liveness_file,
+            schedule_interval=schedule_interval,
+        )
         async_to_sync(runner.run)()
diff --git a/stator/runner.py b/stator/runner.py
index bb1b009..d286bc1 100644
--- a/stator/runner.py
+++ b/stator/runner.py
@@ -3,7 +3,7 @@ import datetime
 import time
 import traceback
 import uuid
-from typing import List, Type
+from typing import List, Optional, Type
 
 from django.utils import timezone
 
@@ -13,7 +13,7 @@ from stator.models import StatorModel
 class StatorRunner:
     """
     Runs tasks on models that are looking for state changes.
-    Designed to run in a one-shot mode, living inside a request.
+    Designed to run for a determinate amount of time, and then exit.
     """
 
     def __init__(
@@ -21,57 +21,63 @@ class StatorRunner:
         models: List[Type[StatorModel]],
         concurrency: int = 50,
         concurrency_per_model: int = 10,
-        run_period: int = 60,
-        wait_period: int = 30,
+        liveness_file: Optional[str] = None,
+        schedule_interval: int = 30,
+        lock_expiry: int = 300,
     ):
         self.models = models
         self.runner_id = uuid.uuid4().hex
         self.concurrency = concurrency
         self.concurrency_per_model = concurrency_per_model
-        self.run_period = run_period
-        self.total_period = run_period + wait_period
+        self.liveness_file = liveness_file
+        self.schedule_interval = schedule_interval
+        self.lock_expiry = lock_expiry
 
     async def run(self):
-        start_time = time.monotonic()
         self.handled = 0
+        self.last_clean = time.monotonic() - self.schedule_interval
         self.tasks = []
-        # Clean up old locks
-        print("Running initial cleaning and scheduling")
-        initial_tasks = []
-        for model in self.models:
-            initial_tasks.append(model.atransition_clean_locks())
-            initial_tasks.append(model.atransition_schedule_due())
-        await asyncio.gather(*initial_tasks)
         # For the first time period, launch tasks
         print("Running main task loop")
-        while (time.monotonic() - start_time) < self.run_period:
-            self.remove_completed_tasks()
-            space_remaining = self.concurrency - len(self.tasks)
-            # Fetch new tasks
-            for model in self.models:
-                if space_remaining > 0:
-                    for instance in await model.atransition_get_with_lock(
-                        number=min(space_remaining, self.concurrency_per_model),
-                        lock_expiry=(
-                            timezone.now()
-                            + datetime.timedelta(seconds=(self.total_period * 2) + 60)
-                        ),
-                    ):
-                        self.tasks.append(
-                            asyncio.create_task(self.run_transition(instance))
-                        )
-                        self.handled += 1
-                        space_remaining -= 1
-            # Prevent busylooping
-            await asyncio.sleep(0.1)
-        # Then wait for tasks to finish
-        print("Waiting for tasks to complete")
-        while (time.monotonic() - start_time) < self.total_period:
-            self.remove_completed_tasks()
-            if not self.tasks:
-                break
-            # Prevent busylooping
-            await asyncio.sleep(1)
+        try:
+            while True:
+                # Do we need to do cleaning?
+                if (time.monotonic() - self.last_clean) >= self.schedule_interval:
+                    print(f"{self.handled} tasks processed so far")
+                    print("Running cleaning and scheduling")
+                    self.remove_completed_tasks()
+                    for model in self.models:
+                        asyncio.create_task(model.atransition_clean_locks())
+                        asyncio.create_task(model.atransition_schedule_due())
+                    self.last_clean = time.monotonic()
+                # Calculate space left for tasks
+                space_remaining = self.concurrency - len(self.tasks)
+                # Fetch new tasks
+                for model in self.models:
+                    if space_remaining > 0:
+                        for instance in await model.atransition_get_with_lock(
+                            number=min(space_remaining, self.concurrency_per_model),
+                            lock_expiry=(
+                                timezone.now()
+                                + datetime.timedelta(seconds=self.lock_expiry)
+                            ),
+                        ):
+                            self.tasks.append(
+                                asyncio.create_task(self.run_transition(instance))
+                            )
+                            self.handled += 1
+                            space_remaining -= 1
+                # Prevent busylooping
+                await asyncio.sleep(0.1)
+        except KeyboardInterrupt:
+            # Wait for tasks to finish
+            print("Waiting for tasks to complete")
+            while True:
+                self.remove_completed_tasks()
+                if not self.tasks:
+                    break
+                # Prevent busylooping
+                await asyncio.sleep(1)
         print("Complete")
         return self.handled
 
diff --git a/stator/views.py b/stator/views.py
deleted file mode 100644
index 9d2e154..0000000
--- a/stator/views.py
+++ /dev/null
@@ -1,23 +0,0 @@
-from django.conf import settings
-from django.http import HttpResponse, HttpResponseForbidden
-from django.views import View
-
-from stator.models import StatorModel
-from stator.runner import StatorRunner
-
-
-class RequestRunner(View):
-    """
-    Runs a Stator runner within a HTTP request. For when you're on something
-    serverless.
-    """
-
-    async def get(self, request):
-        # Check the token, if supplied
-        if settings.STATOR_TOKEN:
-            if request.GET.get("token") != settings.STATOR_TOKEN:
-                return HttpResponseForbidden()
-        # Run on all models
-        runner = StatorRunner(StatorModel.subclasses)
-        handled = await runner.run()
-        return HttpResponse(f"Handled {handled}")