skypilot-nightly 1.0.0.dev20250328__py3-none-any.whl → 1.0.0.dev20250329__py3-none-any.whl

sky/__init__.py

@@ -5,7 +5,7 @@ from typing import Optional
 import urllib.request
 
 # Replaced with the current commit when building the wheels.
-_SKYPILOT_COMMIT_SHA = '4dab9ac5b17fd6dc807b98dcbccecf481011835a'
+_SKYPILOT_COMMIT_SHA = '9e6cef22ae679a1fef4539cbfc02a6d5cdc405ed'
 
 
 def _get_git_commit():
@@ -35,7 +35,7 @@ def _get_git_commit():
 
 
 __commit__ = _get_git_commit()
-__version__ = '1.0.0.dev20250328'
+__version__ = '1.0.0.dev20250329'
 __root_dir__ = os.path.dirname(os.path.abspath(__file__))
 
 

sky/clouds/do.py

@@ -280,6 +280,8 @@ class DO(clouds.Cloud):
         return True, None
 
     def get_credential_file_mounts(self) -> Dict[str, str]:
+        if do_utils.CREDENTIALS_PATH is None:
+            return {}
         if not os.path.exists(os.path.expanduser(do_utils.CREDENTIALS_PATH)):
             return {}
         return {

sky/server/common.py

@@ -51,6 +51,11 @@ API_SERVER_CMD = '-m sky.server.server'
 API_SERVER_CLIENT_DIR = pathlib.Path('~/.sky/api_server/clients')
 RETRY_COUNT_ON_TIMEOUT = 3
 
+# The maximum time to wait for the API server to start, set to a conservative
+# value that unlikely to reach since the server might be just starting slowly
+# (e.g. in high contention env) and we will exit eagerly if server exit.
+WAIT_APISERVER_START_TIMEOUT_SEC = 60
+
 SKY_API_VERSION_WARNING = (
     f'{colorama.Fore.YELLOW}SkyPilot API server is too old: '
     f'v{{server_version}} (client version is v{{client_version}}). '
@@ -179,7 +184,8 @@ def _start_api_server(deploy: bool = False,
     server_url = get_server_url(host)
     assert server_url in AVAILABLE_LOCAL_API_SERVER_URLS, (
         f'server url {server_url} is not a local url')
-    with rich_utils.client_status('Starting SkyPilot API server'):
+    with rich_utils.client_status('Starting SkyPilot API server, '
+                                  f'view logs at {constants.API_SERVER_LOGS}'):
         logger.info(f'{colorama.Style.DIM}Failed to connect to '
                     f'SkyPilot API server at {server_url}. '
                     'Starting a local server.'
@@ -216,14 +222,16 @@ def _start_api_server(deploy: bool = False,
         # If this is called from a CLI invocation, we need
         # start_new_session=True so that SIGINT on the CLI will not also kill
         # the API server.
-        subprocess.Popen(cmd, shell=True, start_new_session=True)
+        proc = subprocess.Popen(cmd, shell=True, start_new_session=True)
 
-        # Wait for the server to start until timeout.
-        # Conservative upper time bound for starting the server based on
-        # profiling.
-        timeout_sec = 12
         start_time = time.time()
         while True:
+            # Check if process has exited
+            if proc.poll() is not None:
+                with ux_utils.print_exception_no_traceback():
+                    raise RuntimeError(
+                        'SkyPilot API server process exited unexpectedly.\n'
+                        f'View logs at: {constants.API_SERVER_LOGS}')
             api_server_info = get_api_server_status()
             assert api_server_info.status != ApiServerStatus.VERSION_MISMATCH, (
                 f'API server version mismatch when starting the server. '
@@ -231,7 +239,7 @@ def _start_api_server(deploy: bool = False,
                 f'Client version: {server_constants.API_VERSION}')
             if api_server_info.status == ApiServerStatus.HEALTHY:
                 break
-            elif time.time() - start_time >= timeout_sec:
+            elif time.time() - start_time >= WAIT_APISERVER_START_TIMEOUT_SEC:
                 with ux_utils.print_exception_no_traceback():
                     raise RuntimeError(
                         'Failed to start SkyPilot API server at '

sky/server/requests/executor.py

@@ -465,7 +465,7 @@ def start(deploy: bool) -> List[multiprocessing.Process]:
             target=mp_queue.start_queue_manager, args=(queue_names, port))
         queue_server.start()
         sub_procs.append(queue_server)
-        mp_queue.wait_for_queues_to_be_ready(queue_names, port=port)
+        mp_queue.wait_for_queues_to_be_ready(queue_names, queue_server, port)
 
     logger.info('Request queues created')
 

sky/server/requests/queues/mp_queue.py

@@ -1,4 +1,5 @@
 """Shared queues for multiprocessing."""
+import multiprocessing
 from multiprocessing import managers
 import queue
 import time
@@ -57,10 +58,13 @@ def get_queue(queue_name: str,
 
 
 def wait_for_queues_to_be_ready(queue_names: List[str],
+                                queue_server: multiprocessing.Process,
                                 port: int = DEFAULT_QUEUE_MANAGER_PORT) -> None:
     """Wait for the queues to be ready after queue manager is just started."""
     initial_time = time.time()
-    max_wait_time = 5
+    # Wait for queue manager to be ready. Exit eagerly if the manager process
+    # exits, just wait for a long timeout that is unlikely to reach otherwise.
+    max_wait_time = 60
     while queue_names:
         try:
             get_queue(queue_names[0], port)
@@ -70,6 +74,9 @@ def wait_for_queues_to_be_ready(queue_names: List[str],
             logger.info(f'Waiting for request queue, named {queue_names[0]!r}, '
                         f'to be ready...')
             time.sleep(0.2)
+            if not queue_server.is_alive():
+                raise RuntimeError(
+                    'Queue manager process exited unexpectedly.') from e
             if time.time() - initial_time > max_wait_time:
                 raise RuntimeError(
                     f'Request queue, named {queue_names[0]!r}, '

sky/server/server.py

@@ -1116,7 +1116,7 @@ if __name__ == '__main__':
 
     sub_procs = []
     try:
-        sub_procs = executor.start(cmd_args.deploy)
+        sub_procs = executor.start(deploy=cmd_args.deploy)
         logger.info(f'Starting SkyPilot API server, workers={num_workers}')
         # We don't support reload for now, since it may cause leakage of request
         # workers or interrupt running requests.

{skypilot_nightly-1.0.0.dev20250328.dist-info → skypilot_nightly-1.0.0.dev20250329.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: skypilot-nightly
-Version: 1.0.0.dev20250328
+Version: 1.0.0.dev20250329
 Summary: SkyPilot: An intercloud broker for the clouds
 Author: SkyPilot Team
 License: Apache 2.0
@@ -230,7 +230,7 @@ SkyPilot **is easy to use for AI users**:
 - Easy job management: queue, run, and auto-recover many jobs
 
 SkyPilot **unifies multiple clusters, clouds, and hardware**:
-- One interface to use reserved GPUs, Kubernetes clusters, or 15+ clouds
+- One interface to use reserved GPUs, Kubernetes clusters, or 16+ clouds
 - [Flexible provisioning](https://docs.skypilot.co/en/latest/examples/auto-failover.html) of GPUs, TPUs, CPUs, with auto-retry
 - [Team deployment](https://docs.skypilot.co/en/latest/reference/api-server/api-server.html) and resource sharing
 
@@ -262,6 +262,7 @@ VMware vSphere, Nebius.
 <p align="center">
   <img alt="SkyPilot" src="https://raw.githubusercontent.com/skypilot-org/skypilot/master/docs/source/images/cloud-logos-light.png" width=85%>
 </p>
+<!-- source xcf file: https://drive.google.com/drive/folders/1S_acjRsAD3T14qMeEnf6FFrIwHu_Gs_f?usp=drive_link -->
 
 
 ## Getting started

{skypilot_nightly-1.0.0.dev20250328.dist-info → skypilot_nightly-1.0.0.dev20250329.dist-info}/RECORD

@@ -1,4 +1,4 @@
-sky/__init__.py,sha256=4GWeKYvInaF3mKBoHHRbOaX6jSHOfnjY0yHaGLpRmPQ,6428
+sky/__init__.py,sha256=sMgwavp0z5P8piZvV57bnGkBJpsUPU6jlzZUqdzi3ZU,6428
 sky/admin_policy.py,sha256=hPo02f_A32gCqhUueF0QYy1fMSSKqRwYEg_9FxScN_s,3248
 sky/authentication.py,sha256=ND011K_-Ud1dVZF37A9KrwYir_ihJXcHc7iDWmuBc8Q,22872
 sky/check.py,sha256=iMGuM7yjUPRgDHc13Pf1_LSybBqIexq-6aXfnVcaI54,15898
@@ -51,7 +51,7 @@ sky/clouds/aws.py,sha256=-feXZ72UMUjuZz8dV4qMTjMMTC9pwfkQQT3KJDcIv5A,54633
 sky/clouds/azure.py,sha256=Zpo6ftWz_B30mX7N-An7JVO-8v7aU3f9cw1iH9phvwE,32251
 sky/clouds/cloud.py,sha256=OoSyFNYtby2Y0h2TpfMB_lEeolIZOQcfKgIn6AvRC68,36694
 sky/clouds/cudo.py,sha256=_UkLEtwJsfDMKlmJfML5W3rA8VArba4x8YGIdnvgZoM,13226
-sky/clouds/do.py,sha256=ZCJa3PTwnnO-mnX-97gD-7vSmo4tsXLLQU4KfnHogMo,11512
+sky/clouds/do.py,sha256=P38l4otp2AuDReUH9Ii621ht9s-NIyb7-R37jbtjHk8,11580
 sky/clouds/fluidstack.py,sha256=jIqW1MLe55MVME1PATZm8e6_FsiTnJawW7OdytPW0aM,12666
 sky/clouds/gcp.py,sha256=sUJ9LXUnMxYm6OYZ5P-z1dJHxgVILuC3OW3eFSTNCv8,56919
 sky/clouds/ibm.py,sha256=XtuPN8QgrwJdb1qb_b-7KwAE2tf_N9wh9eEfi2tcg-s,22013
@@ -229,20 +229,20 @@ sky/serve/server/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,
 sky/serve/server/core.py,sha256=pRvFadEIH_WTUkTtSmuFoPBP4JFq8Obt68ifi9DWuog,36865
 sky/serve/server/server.py,sha256=gQGVU9nHYdGbaLhGjIUNIYn4xwKjRASRJkiiTL5AI1Y,3283
 sky/server/__init__.py,sha256=MPPBqFzXz6Jv5QSk6td_IcvnfXfNErDZVcizu4MLRow,27
-sky/server/common.py,sha256=9gugDflAA0xKg18N0izgaBUezMbti3DdnSYSiH9jo1I,18651
+sky/server/common.py,sha256=6FQ-2X4AagshuXNxKxjgFRKT_mtgTGh7kG1GOJ2tZIM,19185
 sky/server/constants.py,sha256=_ZNrxYh8vmgbf3DmkGDduxjvO2y43ZSPTkH5rCNsVjU,770
-sky/server/server.py,sha256=62IysoY5jCbGi99xIsYrINFIuRgo-cKKIR8fXsKMuW0,44472
+sky/server/server.py,sha256=b0D1lxZgozC1ny7xDf1dVAMQaF5u386PT2QETa9dmIk,44479
 sky/server/stream_utils.py,sha256=4JMHgtoXPpCT8JwtqyUcDQ9IdZFir9om0JaCRr8rvbQ,5849
 sky/server/uvicorn.py,sha256=wajwPHJ3IEEP3GMNOCc0S81-1v2qT5F-ejUkLFVhUzk,2953
 sky/server/html/log.html,sha256=TSGZktua9Ysl_ysg3w60rjxAxhH61AJnsYDHdtqrjmI,6929
 sky/server/requests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sky/server/requests/event_loop.py,sha256=OhpPbuce65bbjpGRlcJa78AVnYSm08SzFKt70ypCUuQ,1211
-sky/server/requests/executor.py,sha256=BNJqkTQ3swYeRO5YVW-dTmobL2CYnDDf_m-kY7__n40,21684
+sky/server/requests/executor.py,sha256=wW-8s8APEakYRJKNiWIhHg2vrn2UUteseUTsqm9q3Fg,21693
 sky/server/requests/payloads.py,sha256=K9CzEqRLDdU2R7s3yxPDD2vC-h6beD_RPdSRpjoQzu4,16529
 sky/server/requests/preconditions.py,sha256=ipxIb_3JXG6S3-ymcOdqQNb7VDvoPqADxu9ZK7-nQWc,7179
 sky/server/requests/requests.py,sha256=9ovdQE-zv_Mvc6IbGATHVyQlOxSKjg_OankZbgDVGeE,21338
 sky/server/requests/queues/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-sky/server/requests/queues/mp_queue.py,sha256=_7AFas__0b1L8e7Bwy4lu0VYU18R85YwMlDHPhQCfh0,2998
+sky/server/requests/queues/mp_queue.py,sha256=jDqP4Jd28U3ibSFyMR1DF9I2OWZrPZqFJrG5S6RFpyw,3403
 sky/server/requests/serializers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sky/server/requests/serializers/decoders.py,sha256=0cpg80uAqkdK_LqcQPkpKswhcNUUztG9luDLm_0eUow,6811
 sky/server/requests/serializers/encoders.py,sha256=i4SAb5Oyp00CyMkyidbdA9dtxAzxZl40KTpL_x6pH0w,5679
@@ -349,9 +349,9 @@ sky/utils/kubernetes/k8s_gpu_labeler_setup.yaml,sha256=VLKT2KKimZu1GDg_4AIlIt488
 sky/utils/kubernetes/kubernetes_deploy_utils.py,sha256=HPVgNt-wbCVPd9dpDFiA7t2mzQLpjXHJ61eiwRbEr-c,10378
 sky/utils/kubernetes/rsync_helper.sh,sha256=h4YwrPFf9727CACnMJvF3EyK_0OeOYKKt4su_daKekw,1256
 sky/utils/kubernetes/ssh_jump_lifecycle_manager.py,sha256=Kq1MDygF2IxFmu9FXpCxqucXLmeUrvs6OtRij6XTQbo,6554
-skypilot_nightly-1.0.0.dev20250328.dist-info/licenses/LICENSE,sha256=emRJAvE7ngL6x0RhQvlns5wJzGI3NEQ_WMjNmd9TZc4,12170
-skypilot_nightly-1.0.0.dev20250328.dist-info/METADATA,sha256=Ean7z4vIG2Xc5ExZ6qUOl-9-84h1smKzaRcmvV4ec18,18657
-skypilot_nightly-1.0.0.dev20250328.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
-skypilot_nightly-1.0.0.dev20250328.dist-info/entry_points.txt,sha256=StA6HYpuHj-Y61L2Ze-hK2IcLWgLZcML5gJu8cs6nU4,36
-skypilot_nightly-1.0.0.dev20250328.dist-info/top_level.txt,sha256=qA8QuiNNb6Y1OF-pCUtPEr6sLEwy2xJX06Bd_CrtrHY,4
-skypilot_nightly-1.0.0.dev20250328.dist-info/RECORD,,
+skypilot_nightly-1.0.0.dev20250329.dist-info/licenses/LICENSE,sha256=emRJAvE7ngL6x0RhQvlns5wJzGI3NEQ_WMjNmd9TZc4,12170
+skypilot_nightly-1.0.0.dev20250329.dist-info/METADATA,sha256=wsP6fph4IygZSON28HbzRno8ygmvOgFBPxEfuruaPNQ,18771
+skypilot_nightly-1.0.0.dev20250329.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
+skypilot_nightly-1.0.0.dev20250329.dist-info/entry_points.txt,sha256=StA6HYpuHj-Y61L2Ze-hK2IcLWgLZcML5gJu8cs6nU4,36
+skypilot_nightly-1.0.0.dev20250329.dist-info/top_level.txt,sha256=qA8QuiNNb6Y1OF-pCUtPEr6sLEwy2xJX06Bd_CrtrHY,4
+skypilot_nightly-1.0.0.dev20250329.dist-info/RECORD,,