Adds settings for worker retry time, with what should be sensible default

diff --git a/docs/configuration.rst b/docs/configuration.rst
index 261eab7d53e0e0377f1ff71d29ed4789658fbde0..9df921cd6e774528d7e5ef105567203fe8b42fab 100644 (file)
--- a/docs/configuration.rst
+++ b/docs/configuration.rst
@@ -540,6 +540,10 @@ PAPERLESS_WORKER_TIMEOUT=<num>
     large documents within the default 1800 seconds. So extending this timeout
     may prove to be useful on weak hardware setups.
 
+PAPERLESS_WORKER_RETRY=<num>
+    If PAPERLESS_WORKER_TIMEOUT has been configured, the retry time for a task can
+    also be configured.  By default, this value will be set to 10s more than the
+    worker timeout.  This value should never be set less than the worker timeout.
 
 PAPERLESS_TIME_ZONE=<timezone>
     Set the time zone here.

diff --git a/src/paperless/settings.py b/src/paperless/settings.py
index 0f9559831eb70c392e97a0788349d4359823e77e..022d66f5886b49af5af562915932e3c1d921257d 100644 (file)
--- a/src/paperless/settings.py
+++ b/src/paperless/settings.py
@@ -3,6 +3,7 @@ import math
 import multiprocessing
 import os
 import re
+from typing import Final
 
 from concurrent_log_handler.queue import setup_logging_queues
 from django.utils.translation import gettext_lazy as _
@@ -29,7 +30,7 @@ elif os.path.exists("/usr/local/etc/paperless.conf"):
 os.environ["OMP_THREAD_LIMIT"] = "1"
 
 
-def __get_boolean(key, default="NO"):
+def __get_boolean(key: str, default: str = "NO") -> bool:
     """
     Return a boolean value based on whatever the user has supplied in the
     environment based on whether the value "looks like" it's True or not.
@@ -37,6 +38,13 @@ def __get_boolean(key, default="NO"):
     return bool(os.getenv(key, default).lower() in ("yes", "y", "1", "t", "true"))
 
 
+def __get_int(key: str, default: int) -> int:
+    """
+    Return an integer value based on the environment variable or a default
+    """
+    return int(os.getenv(key, default))
+
+
 # NEVER RUN WITH DEBUG IN PRODUCTION.
 DEBUG = __get_boolean("PAPERLESS_DEBUG", "NO")
 
@@ -395,7 +403,7 @@ LOGGING = {
 # in total.
 
 
-def default_task_workers():
+def default_task_workers() -> int:
     # always leave one core open
     available_cores = max(multiprocessing.cpu_count(), 1)
     try:
@@ -406,20 +414,29 @@ def default_task_workers():
         return 1
 
 
-TASK_WORKERS = int(os.getenv("PAPERLESS_TASK_WORKERS", default_task_workers()))
+TASK_WORKERS = __get_int("PAPERLESS_TASK_WORKERS", default_task_workers())
+
+PAPERLESS_WORKER_TIMEOUT: Final[int] = __get_int("PAPERLESS_WORKER_TIMEOUT", 1800)
+
+# Per django-q docs, timeout must be smaller than retry
+# We default retry to 10s more than the timeout
+PAPERLESS_WORKER_RETRY: Final[int] = __get_int(
+    "PAPERLESS_WORKER_RETRY",
+    PAPERLESS_WORKER_TIMEOUT + 10,
+)
 
 Q_CLUSTER = {
     "name": "paperless",
     "catch_up": False,
     "recycle": 1,
-    "retry": 1800,
-    "timeout": int(os.getenv("PAPERLESS_WORKER_TIMEOUT", 1800)),
+    "retry": PAPERLESS_WORKER_RETRY,
+    "timeout": PAPERLESS_WORKER_TIMEOUT,
     "workers": TASK_WORKERS,
     "redis": os.getenv("PAPERLESS_REDIS", "redis://localhost:6379"),
 }
 
 
-def default_threads_per_worker(task_workers):
+def default_threads_per_worker(task_workers) -> int:
     # always leave one core open
     available_cores = max(multiprocessing.cpu_count(), 1)
     try: