Skip to content

fix: Replace asyncio.gather() with aiotools.PersistentTaskGroup() for kernel creation tasks #1129

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 11 commits into
base: main
Choose a base branch
from
Open

fix: Replace asyncio.gather() with aiotools.PersistentTaskGroup() for kernel creation tasks #1129

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
4c05199
fix: Add `BaseException` to the checklist
rapsealk Mar 6, 2023
e979b0b
docs: Update news fragment
rapsealk Mar 6, 2023
a961c56
docs: Rename the news fragment with the PR number
rapsealk Mar 6, 2023
50180fa
refactor: Replace `asyncio.gather()` with `aiotools.PersistentTaskGroup`
rapsealk Mar 8, 2023
86dcab4
Update 1129.fix.md
rapsealk Mar 9, 2023
d165f9e
Merge branch 'main' into fix/filter-and-raise-asyncio-cancelled-error
rapsealk Mar 15, 2023
5a223a8
deps: Bump aiotools from 1.5.9 to 1.6.x
rapsealk Mar 15, 2023
355d8b5
fix: Generate lockfile
rapsealk Mar 15, 2023
fb7692f
Merge branch 'main' into fix/filter-and-raise-asyncio-cancelled-error
rapsealk Mar 15, 2023
2cc6a7b
refactor: Replace `asyncio.gather()` with `aiotools.as_completed_safe()`
rapsealk Mar 15, 2023
584a6fa
fix: Reduce spaces for consistency
rapsealk Mar 15, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions changes/1129.fix.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
Replace `asyncio.gather()` with `aiotools.PersistentTaskGroup()` for kernel creation tasks
61 changes: 31 additions & 30 deletions python.lock
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,7 @@
// "aiomonitor-ng~=0.7.0",
// "aioresponses>=0.7.3",
// "aiosqlite~=0.17.0",
// "aiotools~=1.5.9",
// "aiotools~=1.6.0",
// "aiotusclient~=0.1.4",
// "alembic~=1.8.1",
// "appdirs~=1.4.4",
Expand Down Expand Up @@ -434,24 +434,25 @@
"artifacts": [
{
"algorithm": "sha256",
"hash": "3a9946f047cfb33e1fca0d39845c3d6d10aeda69a6602c06a8f7dcb67c12171d",
"url": "https://files.pythonhosted.org/packages/e0/6b/8b7b1c5dfcb1ab49ac0baa0e59f1edb437a05b3fe618b5c5ec931e15ec56/aiotools-1.5.9-py3-none-any.whl"
"hash": "a3d04d32952d3633f8a4f392da4a035c079efb22ff50ef4108a2937b0f31d9b2",
"url": "https://files.pythonhosted.org/packages/dc/ce/5f68a65ddb32f445284a1e85412a3a2e43c49b5a98316d28ffce07c214d2/aiotools-1.6.0-py3-none-any.whl"
},
{
"algorithm": "sha256",
"hash": "44e7b5f50cef692e177ad37231f0d6fa058465ea93c8759da7ee06a2b5cfac97",
"url": "https://files.pythonhosted.org/packages/83/b9/915ad8fa49f8834841bd7ca387db3c374ba8d953e949ef2f185e86ffd652/aiotools-1.5.9.tar.gz"
"hash": "50f0337eb3aa3e02c9264e2853c1fbdd11382690a1a0340d789d32dc0d5f6779",
"url": "https://files.pythonhosted.org/packages/89/75/10776c11516f7c4920944dad34b437cb436ad4db84fa7e8f226d1cd601a6/aiotools-1.6.0.tar.gz"
}
],
"project_name": "aiotools",
"requires_dists": [
"async-timeout~=4.0.2; extra == \"test\"",
"codecov; extra == \"test\"",
"flake8>=4.0.1; extra == \"lint\"",
"mypy>=0.920; extra == \"typecheck\"",
"pytest-asyncio~=0.16.0; extra == \"test\"",
"pytest-asyncio~=0.20.3; extra == \"test\"",
"pytest-cov; extra == \"test\"",
"pytest-mock; extra == \"test\"",
"pytest~=6.2.5; extra == \"test\"",
"pytest~=7.2.2; extra == \"test\"",
"setuptools>=51.2.0; extra == \"build\"",
"sphinx-rtd-theme~=1.0; extra == \"docs\"",
"sphinx~=4.3; extra == \"docs\"",
Expand All @@ -461,7 +462,7 @@
"wheel>=0.37.0; extra == \"build\""
],
"requires_python": ">=3.6",
"version": "1.5.9"
"version": "1.6.0"
},
{
"artifacts": [
Expand Down Expand Up @@ -855,36 +856,36 @@
"artifacts": [
{
"algorithm": "sha256",
"hash": "5374c409d0153e5da4ccde3a0f7e3efac81ce7c60e4e18ffc9acc87c67e0c0a3",
"url": "https://files.pythonhosted.org/packages/4a/ab/92d35739748e7cf94f315a408f3861cc075cecc0a40f2b455972639b6500/boto3-1.26.88-py3-none-any.whl"
"hash": "3ce2225a61832d69831d669d912424ea3863268ca1cfa2a82203bb90952acefa",
"url": "https://files.pythonhosted.org/packages/bd/3b/b59a11ea95f73b78ccd6474909d3bc05b6aebf10983421c60e6f37d00755/boto3-1.26.91-py3-none-any.whl"
},
{
"algorithm": "sha256",
"hash": "f4839565feeaaca4ef775b74a8df996b79b61d988db62c90571bcc948a9f24c5",
"url": "https://files.pythonhosted.org/packages/5a/f0/5e9144029ac56e7dc966a92b2d2181a81b57ab7965500d213cd3f9220780/boto3-1.26.88.tar.gz"
"hash": "278d896e9090a976f41ec68da5c572bc4e5b7cb1e515f1898fee8cb2fadfb50d",
"url": "https://files.pythonhosted.org/packages/0d/f2/b64fa41a705e01c4325e1537306d1dc910b6692e6d1626567a512a924969/boto3-1.26.91.tar.gz"
}
],
"project_name": "boto3",
"requires_dists": [
"botocore<1.30.0,>=1.29.88",
"botocore<1.30.0,>=1.29.91",
"botocore[crt]<2.0a0,>=1.21.0; extra == \"crt\"",
"jmespath<2.0.0,>=0.7.1",
"s3transfer<0.7.0,>=0.6.0"
],
"requires_python": ">=3.7",
"version": "1.26.88"
"version": "1.26.91"
},
{
"artifacts": [
{
"algorithm": "sha256",
"hash": "c320275d18ee140005ace4d04e9e720d5b1e30625cfde5267b1fc5133fd4d44d",
"url": "https://files.pythonhosted.org/packages/72/33/aeed379b1b44ccb6dea8cc4d427fe045ea69a1104f999efe1d4788c09b66/botocore-1.29.88-py3-none-any.whl"
"hash": "4ed6a488aee1b42367eace71f7d0993dda05b02eebd7dcdd78db5c9ce3d80da5",
"url": "https://files.pythonhosted.org/packages/5a/32/92642d3f17b11d3498caf644abd41b43f44b87ff02de0c215b31c0aa555e/botocore-1.29.91-py3-none-any.whl"
},
{
"algorithm": "sha256",
"hash": "48ed2af0a10e7c15a3445eda62f665b106a2aba4ffa1ce730d7083022fd9596f",
"url": "https://files.pythonhosted.org/packages/26/b7/ba1676a574200bfc0a8f99e41daaf4b454917eaf7eeb09c8512bf36193d4/botocore-1.29.88.tar.gz"
"hash": "a8a800a2a945da807758cace539fc5b5ec1d5082ce363799d3a3870c2c4ed6fc",
"url": "https://files.pythonhosted.org/packages/df/8d/9f006c2647f65a6278b7ec19257b1e9e9411f29e6423c08767c71ca01974/botocore-1.29.91.tar.gz"
}
],
"project_name": "botocore",
Expand All @@ -895,7 +896,7 @@
"urllib3<1.27,>=1.25.4"
],
"requires_python": ">=3.7",
"version": "1.29.88"
"version": "1.29.91"
},
{
"artifacts": [
Expand Down Expand Up @@ -2618,13 +2619,13 @@
"artifacts": [
{
"algorithm": "sha256",
"hash": "13b08a53ed71021350c9e300d4ea8668438fb0046ab3937ac9a29913a1a1350a",
"url": "https://files.pythonhosted.org/packages/ca/de/a33823fe54d52ea72fdae011115d737a2642d441c93b68ed17455a328e4c/platformdirs-3.1.0-py3-none-any.whl"
"hash": "e5986afb596e4bb5bde29a79ac9061aa955b94fca2399b7aaac4090860920dd8",
"url": "https://files.pythonhosted.org/packages/7b/e1/593f693096c50411a2bf9571f66bc3be9d0f79a4a50e95aab581458b0e3c/platformdirs-3.1.1-py3-none-any.whl"
},
{
"algorithm": "sha256",
"hash": "accc3665857288317f32c7bebb5a8e482ba717b474f3fc1d18ca7f9214be0cef",
"url": "https://files.pythonhosted.org/packages/8f/5f/01180534cebac14f3a792bf2f74fc99d34531c950c308fdebd9721e85550/platformdirs-3.1.0.tar.gz"
"hash": "024996549ee88ec1a9aa99ff7f8fc819bb59e2c3477b410d90a16d32d6e707aa",
"url": "https://files.pythonhosted.org/packages/79/c4/f98a05535344f79699bbd494e56ac9efc986b7a253fe9f4dba7414a7f505/platformdirs-3.1.1.tar.gz"
}
],
"project_name": "platformdirs",
Expand All @@ -2641,7 +2642,7 @@
"typing-extensions>=4.4; python_version < \"3.8\""
],
"requires_python": ">=3.7",
"version": "3.1.0"
"version": "3.1.1"
},
{
"artifacts": [
Expand Down Expand Up @@ -3944,13 +3945,13 @@
"artifacts": [
{
"algorithm": "sha256",
"hash": "75edcdc2f7d85b137124a6c3c9fc3933cdeaa12ecb9a6a959f22797a0feca7e1",
"url": "https://files.pythonhosted.org/packages/fe/ca/466766e20b767ddb9b951202542310cba37ea5f2d792dae7589f1741af58/urllib3-1.26.14-py2.py3-none-any.whl"
"hash": "aa751d169e23c7479ce47a0cb0da579e3ede798f994f5816a74e4f4500dcea42",
"url": "https://files.pythonhosted.org/packages/7b/f5/890a0baca17a61c1f92f72b81d3c31523c99bec609e60c292ea55b387ae8/urllib3-1.26.15-py2.py3-none-any.whl"
},
{
"algorithm": "sha256",
"hash": "076907bf8fd355cde77728471316625a4d2f7e713c125f51953bb5b3eecf4f72",
"url": "https://files.pythonhosted.org/packages/c5/52/fe421fb7364aa738b3506a2d99e4f3a56e079c0a798e9f4fa5e14c60922f/urllib3-1.26.14.tar.gz"
"hash": "8a388717b9476f934a21484e8c8e61875ab60644d29b9b39e11e4b9dc1c6b305",
"url": "https://files.pythonhosted.org/packages/21/79/6372d8c0d0641b4072889f3ff84f279b738cd8595b64c8e0496d4e848122/urllib3-1.26.15.tar.gz"
}
],
"project_name": "urllib3",
Expand All @@ -3967,7 +3968,7 @@
"urllib3-secure-extra; extra == \"secure\""
],
"requires_python": "!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,!=3.5.*,>=2.7",
"version": "1.26.14"
"version": "1.26.15"
},
{
"artifacts": [
Expand Down Expand Up @@ -4201,7 +4202,7 @@
"aiomonitor-ng~=0.7.0",
"aioresponses>=0.7.3",
"aiosqlite~=0.17.0",
"aiotools~=1.5.9",
"aiotools~=1.6.0",
"aiotusclient~=0.1.4",
"alembic~=1.8.1",
"appdirs~=1.4.4",
Expand Down
2 changes: 1 addition & 1 deletion requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@ aiodns>=3.0
aiomonitor-ng~=0.7.0
aioresponses>=0.7.3
aiosqlite~=0.17.0
aiotools~=1.5.9
aiotools~=1.6.0
aiotusclient~=0.1.4
alembic~=1.8.1
appdirs~=1.4.4
Expand Down
11 changes: 8 additions & 3 deletions src/ai/backend/agent/server.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -357,7 +357,6 @@ async def create_kernels(
):
cluster_info = cast(ClusterInfo, raw_cluster_info)
session_id = SessionId(UUID(raw_session_id))
raw_results = []
coros = []
throttle_sema = asyncio.Semaphore(self.local_config["agent"]["kernel-creation-concurrency"])
for raw_kernel_id, raw_config in zip(raw_kernel_ids, raw_configs):
Expand All @@ -377,8 +376,14 @@ async def create_kernels(
throttle_sema=throttle_sema,
)
)
results = await asyncio.gather(*coros, return_exceptions=True)
errors = [*filter(lambda item: isinstance(item, Exception), results)]
results = []
errors = []
async with aclosing(aiotools.as_completed_safe(coros)) as ag:
async for result in ag:
try:
results.append(await result)
except (BaseException, Exception) as e:
errors.append(e)
if errors:
# Raise up the first error.
if len(errors) == 1:
Expand Down