github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #9577] keepalive parameter not take into account #68302

New Issue
Closed
opened 2026-05-04 13:11:06 -05:00 by GiteaMirror · 8 comments
GiteaMirror commented 2026-05-04 13:11:06 -05:00
Owner
Copy Link

Originally created by @nicho2 on GitHub (Mar 7, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9577

What is the issue?

Hello,

I send a request with keep_alive = 8m

but the ollama release the model after 5m

time start = 2025-03-07T12:30:09
time_release= 2025-03-07T12:35:00

but in the log , there is duration=8m0s

Image

Relevant log output

time=2025-03-07T12:30:09.671Z level=DEBUG source=routes.go:1501 msg="chat request" images=0 prompt="<|start_header_id|>system<|end_header_id|>\n\n\n<i-----------------2<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n"

time=2025-03-07T12:30:09.752Z level=DEBUG source=cache.go:104 msg="loading cache slot" id=0 cache=0 prompt=71199 used=0 remaining=71199

time=2025-03-07T12:35:00.529Z level=DEBUG source=sched.go:467 msg="context for request finished"

time=2025-03-07T12:35:00.529Z level=DEBUG source=sched.go:340 msg="runner with non-zero duration has gone idle, adding timer" modelPath=/root/.ollama/models/blobs/sha256-4824460d29f2058aaf6e1118a63a7a197a09bed509f0e7d4e2efb1ee273b447d duration=8m0s

time=2025-03-07T12:35:00.530Z level=DEBUG source=sched.go:358 msg="after processing request finished event" modelPath=/root/.ollama/models/blobs/sha256-4824460d29f2058aaf6e1118a63a7a197a09bed509f0e7d4e2efb1ee273b447d refCount=0

[GIN] 2025/03/07 - 12:35:00 | 200 |          5m0s |      172.21.0.1 | POST     "/api/chat"

OS

Docker

GPU

Nvidia

CPU

Intel

Ollama version

0.5.13

Originally created by @nicho2 on GitHub (Mar 7, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9577 ### What is the issue? Hello, I send a request with keep_alive = 8m but the ollama release the model after 5m time start = 2025-03-07T12:30:09 time_release= 2025-03-07T12:35:00 but in the log , there is duration=8m0s ![Image](https://github.com/user-attachments/assets/e2873b61-4932-4bb0-97e4-f168e1971208) ### Relevant log output ```shell time=2025-03-07T12:30:09.671Z level=DEBUG source=routes.go:1501 msg="chat request" images=0 prompt="<|start_header_id|>system<|end_header_id|>\n\n\n<i-----------------2<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n" time=2025-03-07T12:30:09.752Z level=DEBUG source=cache.go:104 msg="loading cache slot" id=0 cache=0 prompt=71199 used=0 remaining=71199 time=2025-03-07T12:35:00.529Z level=DEBUG source=sched.go:467 msg="context for request finished" time=2025-03-07T12:35:00.529Z level=DEBUG source=sched.go:340 msg="runner with non-zero duration has gone idle, adding timer" modelPath=/root/.ollama/models/blobs/sha256-4824460d29f2058aaf6e1118a63a7a197a09bed509f0e7d4e2efb1ee273b447d duration=8m0s time=2025-03-07T12:35:00.530Z level=DEBUG source=sched.go:358 msg="after processing request finished event" modelPath=/root/.ollama/models/blobs/sha256-4824460d29f2058aaf6e1118a63a7a197a09bed509f0e7d4e2efb1ee273b447d refCount=0 [GIN] 2025/03/07 - 12:35:00 | 200 | 5m0s | 172.21.0.1 | POST "/api/chat" ``` ### OS Docker ### GPU Nvidia ### CPU Intel ### Ollama version 0.5.13
GiteaMirror added the bugneeds more info labels 2026-05-04 13:11:07 -05:00
GiteaMirror commented 2026-05-04 13:11:11 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 7, 2025):

The model hasn't been unloaded. The completion took 5 minutes, and then the 8 minute timeout started. If another request is not received, the model will be unloaded at 12:43.

<!-- gh-comment-id:2706401840 --> @rick-github commented on GitHub (Mar 7, 2025): The model hasn't been unloaded. The completion took 5 minutes, and then the 8 minute timeout started. If another request is not received, the model will be unloaded at 12:43.
GiteaMirror commented 2026-05-04 13:11:13 -05:00
Author
Owner
Copy Link

@nicho2 commented on GitHub (Mar 7, 2025):

The completion took 5 minutes, and i haven't an answer (llm).

just ollama send:
HTTP/1.1 200 OK
Content-Type: application/x-ndjson
Date: Fri, 07 Mar 2025 13:25:38 GMT
Transfer-Encoding: chunked

{"error":"POST predict: Post "http://127.0.0.1:35039/completion": context canceled"}

<!-- gh-comment-id:2706460496 --> @nicho2 commented on GitHub (Mar 7, 2025): The completion took 5 minutes, and i haven't an answer (llm). just ollama send: HTTP/1.1 200 OK Content-Type: application/x-ndjson Date: Fri, 07 Mar 2025 13:25:38 GMT Transfer-Encoding: chunked {"error":"POST predict: Post \"http://127.0.0.1:35039/completion\": context canceled"}
GiteaMirror commented 2026-05-04 13:11:17 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 7, 2025):

Does your client have a 5 minute timeout?

<!-- gh-comment-id:2706473808 --> @rick-github commented on GitHub (Mar 7, 2025): Does your client have a 5 minute timeout?
GiteaMirror commented 2026-05-04 13:11:20 -05:00
Author
Owner
Copy Link

@nicho2 commented on GitHub (Mar 7, 2025):

No , i don't see

is-it possible is linking with OLLAMA_LOAD_TIMEOUT ? :
OLLAMA_LOAD_TIMEOUT How long to allow model loads to stall before giving up (default "5m")

<!-- gh-comment-id:2706498509 --> @nicho2 commented on GitHub (Mar 7, 2025): No , i don't see is-it possible is linking with OLLAMA_LOAD_TIMEOUT ? : OLLAMA_LOAD_TIMEOUT How long to allow model loads to stall before giving up (default "5m")
GiteaMirror commented 2026-05-04 13:11:23 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 7, 2025):

No, the model is already loaded.

<!-- gh-comment-id:2706503599 --> @rick-github commented on GitHub (Mar 7, 2025): No, the model is already loaded.
GiteaMirror commented 2026-05-04 13:11:24 -05:00
Author
Owner
Copy Link

@nicho2 commented on GitHub (Mar 7, 2025):

sometimes when i regenerate from the chat , i have the answer immediatly and the cache in not null

time=2025-03-07T13:49:10.358Z level=DEBUG source=cache.go:104 msg="loading cache slot" id=0 cache=107963 prompt=95938 used=95888 remaining=50

<!-- gh-comment-id:2706532610 --> @nicho2 commented on GitHub (Mar 7, 2025): sometimes when i regenerate from the chat , i have the answer immediatly and the cache in not null time=2025-03-07T13:49:10.358Z level=DEBUG source=cache.go:104 msg="loading cache slot" id=0 cache=107963 prompt=95938 used=95888 remaining=50
GiteaMirror commented 2026-05-04 13:11:29 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 7, 2025):

Yes, that's the point of a cache.

<!-- gh-comment-id:2706537258 --> @rick-github commented on GitHub (Mar 7, 2025): Yes, that's the point of a cache.
GiteaMirror commented 2026-05-04 13:11:34 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Mar 12, 2025):

OK, I'm not quite sure how to help here. @nicho2 it seems like Ollama is behaving as intended?

<!-- gh-comment-id:2716162807 --> @pdevine commented on GitHub (Mar 12, 2025): OK, I'm not quite sure how to help here. @nicho2 it seems like Ollama is behaving as intended?
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug needs more info
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#68302
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?