feat(cli): scale --request-timeout and --startup-timeout (#417)

* feat(cli): scale --request-timeout and --startup-timeout

* feat(fal): show machine-type and request/startup timeouts

projects/fal/pyproject.toml

@@ -23,7 +23,7 @@ readme = "README.md"
 requires-python = ">=3.8"
 dependencies = [
     "isolate[build]>=0.15.0,<0.16.0",
-    "isolate-proto>=0.6.4,<0.7.0",
+    "isolate-proto>=0.6.6,<0.7.0",
     "grpcio==1.64.0",
     "dill==0.3.7",
     "cloudpickle==3.0.0",

projects/fal/src/fal/cli/apps.py

@@ -19,6 +19,9 @@ def _apps_table(apps: list[AliasInfo]):
     table.add_column("Max Concurrency")
     table.add_column("Max Multiplexing")
     table.add_column("Keep Alive")
+    table.add_column("Request Timeout")
+    table.add_column("Startup Timeout")
+    table.add_column("Machine Type")
     table.add_column("Active Workers")
 
     for app in apps:
@@ -30,6 +33,9 @@ def _apps_table(apps: list[AliasInfo]):
             str(app.max_concurrency),
             str(app.max_multiplexing),
             str(app.keep_alive),
+            str(app.request_timeout),
+            str(app.startup_timeout),
+            " ".join(app.machine_types),
             str(app.active_runners),
         )
 
@@ -67,6 +73,9 @@ def _app_rev_table(revs: list[ApplicationInfo]):
     table.add_column("Max Concurrency")
     table.add_column("Max Multiplexing")
     table.add_column("Keep Alive")
+    table.add_column("Request Timeout")
+    table.add_column("Startup Timeout")
+    table.add_column("Machine Type")
     table.add_column("Active Workers")
 
     for rev in revs:
@@ -76,6 +85,9 @@ def _app_rev_table(revs: list[ApplicationInfo]):
             str(rev.max_concurrency),
             str(rev.max_multiplexing),
             str(rev.keep_alive),
+            str(rev.request_timeout),
+            str(rev.startup_timeout),
+            " ".join(rev.machine_types),
             str(rev.active_runners),
         )
 
@@ -114,6 +126,8 @@ def _scale(args):
             and args.max_multiplexing is None
             and args.max_concurrency is None
             and args.min_concurrency is None
+            and args.request_timeout is None
+            and args.startup_timeout is None
         ):
             args.console.log("No parameters for update were provided, ignoring.")
             return
@@ -124,6 +138,8 @@ def _scale(args):
             max_multiplexing=args.max_multiplexing,
             max_concurrency=args.max_concurrency,
             min_concurrency=args.min_concurrency,
+            request_timeout=args.request_timeout,
+            startup_timeout=args.startup_timeout,
         )
         table = _apps_table([alias_info])
 
@@ -162,6 +178,16 @@ def _add_scale_parser(subparsers, parents):
         type=int,
         help="Minimum concurrency",
     )
+    parser.add_argument(
+        "--request-timeout",
+        type=int,
+        help="Request timeout (seconds).",
+    )
+    parser.add_argument(
+        "--startup-timeout",
+        type=int,
+        help="Startup timeout (seconds).",
+    )
     parser.set_defaults(func=_scale)
 
 

projects/fal/src/fal/sdk.py

@@ -196,6 +196,9 @@ class ApplicationInfo:
     max_multiplexing: int
     active_runners: int
     min_concurrency: int
+    machine_types: list[str]
+    request_timeout: int
+    startup_timeout: int
 
 
 @dataclass
@@ -208,6 +211,9 @@ class AliasInfo:
     max_multiplexing: int
     active_runners: int
     min_concurrency: int
+    machine_types: list[str]
+    request_timeout: int
+    startup_timeout: int
 
 
 @dataclass
@@ -313,6 +319,9 @@ def _from_grpc_application_info(
         max_multiplexing=message.max_multiplexing,
         active_runners=message.active_runners,
         min_concurrency=message.min_concurrency,
+        machine_types=message.machine_types,
+        request_timeout=message.request_timeout,
+        startup_timeout=message.startup_timeout,
     )
 
 
@@ -336,6 +345,9 @@ def _from_grpc_alias_info(message: isolate_proto.AliasInfo) -> AliasInfo:
         max_multiplexing=message.max_multiplexing,
         active_runners=message.active_runners,
         min_concurrency=message.min_concurrency,
+        machine_types=message.machine_types,
+        request_timeout=message.request_timeout,
+        startup_timeout=message.startup_timeout,
     )
 
 
@@ -564,13 +576,17 @@ def update_application(
         max_multiplexing: int | None = None,
         max_concurrency: int | None = None,
         min_concurrency: int | None = None,
+        request_timeout: int | None = None,
+        startup_timeout: int | None = None,
     ) -> AliasInfo:
         request = isolate_proto.UpdateApplicationRequest(
             application_name=application_name,
             keep_alive=keep_alive,
             max_multiplexing=max_multiplexing,
             max_concurrency=max_concurrency,
             min_concurrency=min_concurrency,
+            request_timeout=request_timeout,
+            startup_timeout=startup_timeout,
         )
         res: isolate_proto.UpdateApplicationResult = self.stub.UpdateApplication(
             request