[GH-ISSUE #12019] 长时间运行高负载运行模型容易卡住,需要手动停止模型才能继续 #70040

Open
opened 2026-05-04 20:08:29 -05:00 by GiteaMirror · 0 comments
Owner

Originally created by @geogesors on GitHub (Aug 22, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12019

What is the issue?

长时间运行高负载运行模型容易卡住,需要手动停止模型才能继续
(base) [dj@djjx docker]$ ollama stop gpt-oss:20b ollama才继续运行
Every 3.0s: nvidia-smi djjx: Fri Aug 22 08:49:26 2025

Fri Aug 22 08:49:26 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 570.124.04 Driver Version: 570.124.04 CUDA Version: 12.8 |
|-----------------------------------------+------------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA GeForce RTX 3090 Off | 00000000:03:00.0 On | N/A |
| 71% 65C P0 287W / 350W | 13807MiB / 24576MiB | 60% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 37189 G /usr/libexec/Xorg 173MiB |
| 0 N/A N/A 37576 G /usr/bin/gnome-shell 35MiB |
| 0 N/A N/A 57123 G ...der --variations-seed-version 78MiB |
| 0 N/A N/A 3074934 C /usr/local/bin/ollama 13482MiB |
+-----------------------------------------------------------------------------------------+

Relevant log output

8月 21 22:02:37 djjx ollama[471687]: [GIN] 2025/08/21 - 22:02:37 | 200 |  2.970703814s |      172.22.0.6 | POST     "/api/chat"
8月 21 22:02:39 djjx ollama[471687]: [GIN] 2025/08/21 - 22:02:39 | 200 |   2.45759075s |      172.22.0.6 | POST     "/api/chat"
8月 21 22:03:14 djjx ollama[471687]: [GIN] 2025/08/21 - 22:03:14 | 200 | 34.134935717s |      172.22.0.6 | POST     "/api/chat"
8月 21 22:03:24 djjx ollama[471687]: [GIN] 2025/08/21 - 22:03:24 | 200 | 10.492666631s |      172.22.0.6 | POST     "/api/chat"
8月 21 22:04:05 djjx ollama[471687]: [GIN] 2025/08/21 - 22:04:05 | 200 | 40.795863116s |      172.22.0.6 | POST     "/api/chat"
8月 21 22:04:45 djjx ollama[471687]: [GIN] 2025/08/21 - 22:04:45 | 200 | 39.964508371s |      172.22.0.6 | POST     "/api/chat"
8月 21 22:06:33 djjx ollama[471687]: [GIN] 2025/08/21 - 22:06:33 | 200 |         1m47s |      172.22.0.6 | POST     "/api/chat"
8月 21 22:06:46 djjx ollama[471687]: [GIN] 2025/08/21 - 22:06:46 | 200 | 13.194685419s |      172.22.0.6 | POST     "/api/chat"
8月 21 22:07:11 djjx ollama[471687]: [GIN] 2025/08/21 - 22:07:11 | 200 | 24.952486008s |      172.22.0.6 | POST     "/api/chat"
8月 21 22:07:41 djjx ollama[471687]: [GIN] 2025/08/21 - 22:07:41 | 200 |  30.44871662s |      172.22.0.6 | POST     "/api/chat"
8月 22 08:43:49 djjx ollama[471687]: [GIN] 2025/08/22 - 08:43:49 | 200 |     11h44m30s |      172.22.0.6 | POST     "/api/chat"
8月 22 08:44:06 djjx ollama[471687]: [GIN] 2025/08/22 - 08:44:06 | 200 |     11h44m26s |      172.22.0.6 | POST     "/api/chat"
8月 22 08:44:11 djjx ollama[471687]: [GIN] 2025/08/22 - 08:44:11 | 200 |     11h44m28s |      172.22.0.6 | POST     "/api/chat"
8月 22 08:44:36 djjx ollama[471687]: [GIN] 2025/08/22 - 08:44:36 | 200 |     11h44m36s |      172.22.0.6 | POST     "/api/chat"
8月 22 08:44:58 djjx ollama[471687]: [GIN] 2025/08/22 - 08:44:58 | 200 |     11h44m24s |      172.22.0.6 | POST     "/api/chat"

Every 3.0s: ollama ps                                                                                                                                                                                           djjx: Fri Aug 22 08:48:10 2025

NAME           ID              SIZE     PROCESSOR    CONTEXT    UNTIL
gpt-oss:20b    aa4295ac10c3    16 GB    100% GPU     16384	Forever

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @geogesors on GitHub (Aug 22, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12019 ### What is the issue? 长时间运行高负载运行模型容易卡住,需要手动停止模型才能继续 (base) [dj@djjx docker]$ ollama stop gpt-oss:20b ollama才继续运行 Every 3.0s: nvidia-smi djjx: Fri Aug 22 08:49:26 2025 Fri Aug 22 08:49:26 2025 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 570.124.04 Driver Version: 570.124.04 CUDA Version: 12.8 | |-----------------------------------------+------------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+========================+======================| | 0 NVIDIA GeForce RTX 3090 Off | 00000000:03:00.0 On | N/A | | 71% 65C P0 287W / 350W | 13807MiB / 24576MiB | 60% Default | | | | N/A | +-----------------------------------------+------------------------+----------------------+ +-----------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=========================================================================================| | 0 N/A N/A 37189 G /usr/libexec/Xorg 173MiB | | 0 N/A N/A 37576 G /usr/bin/gnome-shell 35MiB | | 0 N/A N/A 57123 G ...der --variations-seed-version 78MiB | | 0 N/A N/A 3074934 C /usr/local/bin/ollama 13482MiB | +-----------------------------------------------------------------------------------------+ ### Relevant log output ```shell 8月 21 22:02:37 djjx ollama[471687]: [GIN] 2025/08/21 - 22:02:37 | 200 | 2.970703814s | 172.22.0.6 | POST "/api/chat" 8月 21 22:02:39 djjx ollama[471687]: [GIN] 2025/08/21 - 22:02:39 | 200 | 2.45759075s | 172.22.0.6 | POST "/api/chat" 8月 21 22:03:14 djjx ollama[471687]: [GIN] 2025/08/21 - 22:03:14 | 200 | 34.134935717s | 172.22.0.6 | POST "/api/chat" 8月 21 22:03:24 djjx ollama[471687]: [GIN] 2025/08/21 - 22:03:24 | 200 | 10.492666631s | 172.22.0.6 | POST "/api/chat" 8月 21 22:04:05 djjx ollama[471687]: [GIN] 2025/08/21 - 22:04:05 | 200 | 40.795863116s | 172.22.0.6 | POST "/api/chat" 8月 21 22:04:45 djjx ollama[471687]: [GIN] 2025/08/21 - 22:04:45 | 200 | 39.964508371s | 172.22.0.6 | POST "/api/chat" 8月 21 22:06:33 djjx ollama[471687]: [GIN] 2025/08/21 - 22:06:33 | 200 | 1m47s | 172.22.0.6 | POST "/api/chat" 8月 21 22:06:46 djjx ollama[471687]: [GIN] 2025/08/21 - 22:06:46 | 200 | 13.194685419s | 172.22.0.6 | POST "/api/chat" 8月 21 22:07:11 djjx ollama[471687]: [GIN] 2025/08/21 - 22:07:11 | 200 | 24.952486008s | 172.22.0.6 | POST "/api/chat" 8月 21 22:07:41 djjx ollama[471687]: [GIN] 2025/08/21 - 22:07:41 | 200 | 30.44871662s | 172.22.0.6 | POST "/api/chat" 8月 22 08:43:49 djjx ollama[471687]: [GIN] 2025/08/22 - 08:43:49 | 200 | 11h44m30s | 172.22.0.6 | POST "/api/chat" 8月 22 08:44:06 djjx ollama[471687]: [GIN] 2025/08/22 - 08:44:06 | 200 | 11h44m26s | 172.22.0.6 | POST "/api/chat" 8月 22 08:44:11 djjx ollama[471687]: [GIN] 2025/08/22 - 08:44:11 | 200 | 11h44m28s | 172.22.0.6 | POST "/api/chat" 8月 22 08:44:36 djjx ollama[471687]: [GIN] 2025/08/22 - 08:44:36 | 200 | 11h44m36s | 172.22.0.6 | POST "/api/chat" 8月 22 08:44:58 djjx ollama[471687]: [GIN] 2025/08/22 - 08:44:58 | 200 | 11h44m24s | 172.22.0.6 | POST "/api/chat" Every 3.0s: ollama ps djjx: Fri Aug 22 08:48:10 2025 NAME ID SIZE PROCESSOR CONTEXT UNTIL gpt-oss:20b aa4295ac10c3 16 GB 100% GPU 16384 Forever ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-05-04 20:08:29 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#70040