github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #7779] Models get stuck in stopping state #82653

New Issue
Closed
opened 2026-05-09 15:27:28 -05:00 by GiteaMirror · 18 comments
GiteaMirror commented 2026-05-09 15:27:28 -05:00
Owner
Copy Link

Originally created by @ProjectMoon on GitHub (Nov 21, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7779

What is the issue?

Possibly similar or related to https://github.com/ollama/ollama/issues/6380. But I'm opening this issue because I don't see semaphore errors when this happens.

I notice that when I am issuing a lot of requests (code generation/fill in the middle tasks), the code completion model cannot stop, and then ollama gets locked up. Logs (without debug mode on) just show the model being loaded, and then any subsequent API requests. The service has to be restarted in order to get responses working again. I actually set up a 0.3.14 ollama instance proxied through OpenWebUI to get the code completion model working properly, as 0.3.14 does not have this problem.

OS

Linux

GPU

AMD

CPU

AMD

Ollama version

0.4.2,0.43-rc0

Originally created by @ProjectMoon on GitHub (Nov 21, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7779 ### What is the issue? Possibly similar or related to https://github.com/ollama/ollama/issues/6380. But I'm opening this issue because I don't see semaphore errors when this happens. I notice that when I am issuing a lot of requests (code generation/fill in the middle tasks), the code completion model cannot stop, and then ollama gets locked up. Logs (without debug mode on) just show the model being loaded, and then any subsequent API requests. The service has to be restarted in order to get responses working again. I actually set up a 0.3.14 ollama instance proxied through OpenWebUI to get the code completion model working properly, as 0.3.14 does not have this problem. ### OS Linux ### GPU AMD ### CPU AMD ### Ollama version 0.4.2,0.43-rc0
GiteaMirror added the topbug labels 2026-05-09 15:27:28 -05:00
GiteaMirror commented 2026-05-09 15:27:31 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 21, 2024):

Server logs may aid in debugging. Does it get in this state if OLLAMA_NUM_PARALLEL=1?

<!-- gh-comment-id:2491856607 --> @rick-github commented on GitHub (Nov 21, 2024): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) may aid in debugging. Does it get in this state if `OLLAMA_NUM_PARALLEL=1`?
GiteaMirror commented 2026-05-09 15:27:31 -05:00
Author
Owner
Copy Link

@jessegross commented on GitHub (Nov 21, 2024):

If you are able to turn on OLLAMA_DEBUG and run it with 0.4.3-rc0 (or later) that would greatly help. When it gets stuck is the system idle or is there 100% CPU/GPU usage?

<!-- gh-comment-id:2491989083 --> @jessegross commented on GitHub (Nov 21, 2024): If you are able to turn on OLLAMA_DEBUG and run it with 0.4.3-rc0 (or later) that would greatly help. When it gets stuck is the system idle or is there 100% CPU/GPU usage?
GiteaMirror commented 2026-05-09 15:27:32 -05:00
Author
Owner
Copy Link

@ProjectMoon commented on GitHub (Nov 21, 2024):

So I just spent a good 20 minutes messing with this trying to get it to lock up, but it hasn't happened with 0.4.3-rc0. Which is odd because I could've swore it was happening with it this morning. Even messing around with 0.4.2 a bit doesn't do it. I will continue using the code completion on 0.4.3-rc0 and see what happens. Not able to reproduce again yet, which is odd because I was able to reproduce it this morning, and last night...

<!-- gh-comment-id:2492129788 --> @ProjectMoon commented on GitHub (Nov 21, 2024): So I just spent a good 20 minutes messing with this trying to get it to lock up, but it hasn't happened with 0.4.3-rc0. Which is odd because I could've swore it was happening with it this morning. Even messing around with 0.4.2 a bit doesn't do it. I will continue using the code completion on 0.4.3-rc0 and see what happens. Not able to reproduce again yet, which is odd because I was able to reproduce it this morning, and last night...
GiteaMirror commented 2026-05-09 15:27:32 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 21, 2024):

Which model are you using? I have a stress test I can put it through with various versions of ollama to see if the problem can be triggered.

<!-- gh-comment-id:2492186804 --> @rick-github commented on GitHub (Nov 21, 2024): Which model are you using? I have a stress test I can put it through with various versions of ollama to see if the problem can be triggered.
GiteaMirror commented 2026-05-09 15:27:33 -05:00
Author
Owner
Copy Link

@ProjectMoon commented on GitHub (Nov 21, 2024):

I first noticed it yesterday when using ollama 0.4.2 with https://gitlab.com/daanturo/starhugger.el with StarCoder2 3b. But then I also noticed it with Qwen2.5 Coder Base 7b and 14b. So I am not sure it's really related to the model itself. But those are what I've been using. Would it help if I share a part of my Emacs configuration for Starhugger?

<!-- gh-comment-id:2492311265 --> @ProjectMoon commented on GitHub (Nov 21, 2024): I first noticed it yesterday when using ollama 0.4.2 with https://gitlab.com/daanturo/starhugger.el with StarCoder2 3b. But then I also noticed it with Qwen2.5 Coder Base 7b and 14b. So I am not sure it's really related to the model itself. But those are what I've been using. Would it help if I share a part of my Emacs configuration for Starhugger?
GiteaMirror commented 2026-05-09 15:27:35 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 21, 2024):

Yes, it would be helpful to have the Starhugger config so I can craft the fuzz input.

<!-- gh-comment-id:2492358120 --> @rick-github commented on GitHub (Nov 21, 2024): Yes, it would be helpful to have the Starhugger config so I can craft the fuzz input.
GiteaMirror commented 2026-05-09 15:27:36 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 21, 2024):

Is it starcoder2:3b-q4_K_M or a different quant?

<!-- gh-comment-id:2492365345 --> @rick-github commented on GitHub (Nov 21, 2024): Is it starcoder2:3b-q4_K_M or a different quant?
GiteaMirror commented 2026-05-09 15:27:36 -05:00
Author
Owner
Copy Link

@ProjectMoon commented on GitHub (Nov 21, 2024):

I think for 3b I used q6_K. For Starcoder2 7b I used q4_K_M.

Double checking shell history:

ollama pull starcoder2
ollama pull starcoder2:7b-q6_K

But this isn't the whole history, because I also pulled the 3b model too at some point.

<!-- gh-comment-id:2492379441 --> @ProjectMoon commented on GitHub (Nov 21, 2024): ~I think for 3b I used q6_K. For Starcoder2 7b I used q4_K_M.~ Double checking shell history: ``` ollama pull starcoder2 ollama pull starcoder2:7b-q6_K ``` But this isn't the whole history, because I also pulled the 3b model too at some point.
GiteaMirror commented 2026-05-09 15:27:37 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Nov 21, 2024):

So, I've got ollama loaded with starcoder2:3b-q6_K to go unresponsive. Or rather, the runner is, because I have starcoder2:3b-q4_K_M also loaded and that's responding to queries. What I observe with the q6_K runner is that it's still processing, but the sequences it's sending back to the ollama server are variations of the string "<|end_of_text|>", like " end_of_text|>" or "end_of_text\n". Notably, the template and Modelfile consider "<|end_of_text|>" as a stop token, but this sequence is not found in the tokenizer list. Instead, "<|endoftext|>" is listed, and that's found in tokenizer config at HF. Also, because it's generating thousands of these sequences and filling the context window, it's generating tons of "context limit hit - shifting" events, which is a performance bottleneck in 0.4.2 which was just fixed for 0.4.3. So at least part of this can be blamed on a poorly configured model. I'll see if adjusting the config improves performance, and then I'll look at the other starcoder2 quants and the qwen2.5-coder family to see if anything similar is happening there.

$ curl -s localhost:11434/api/show -d '{"model":"starcoder2:3b-q6_K","verbose":true}' | jq '.model_info."tokenizer.ggml.tokens"[0]'
"<|endoftext|>"
<!-- gh-comment-id:2492467274 --> @rick-github commented on GitHub (Nov 21, 2024): So, I've got ollama loaded with starcoder2:3b-q6_K to go unresponsive. Or rather, the runner is, because I have starcoder2:3b-q4_K_M also loaded and that's responding to queries. What I observe with the q6_K runner is that it's still processing, but the sequences it's sending back to the ollama server are variations of the string "<|end_of_text|>", like " end_of_text|>" or "end_of_text\n". Notably, the template and Modelfile consider "<|end_of_text|>" as a stop token, but this sequence is not found in the tokenizer list. Instead, "<|endoftext|>" is listed, and that's found in [tokenizer config](https://huggingface.co/bigcode/starcoder2-3b/blob/main/tokenizer_config.json) at HF. Also, because it's generating thousands of these sequences and filling the context window, it's generating tons of "context limit hit - shifting" events, which is a performance bottleneck in 0.4.2 which was just fixed for 0.4.3. So at least part of this can be blamed on a poorly configured model. I'll see if adjusting the config improves performance, and then I'll look at the other starcoder2 quants and the qwen2.5-coder family to see if anything similar is happening there. ```console $ curl -s localhost:11434/api/show -d '{"model":"starcoder2:3b-q6_K","verbose":true}' | jq '.model_info."tokenizer.ggml.tokens"[0]' "<|endoftext|>" ```
GiteaMirror commented 2026-05-09 15:27:38 -05:00
Author
Owner
Copy Link

@ProjectMoon commented on GitHub (Nov 22, 2024):

I definitely noticed it with StarCoder2 first, which is the recommended model for use with Starhugger. I switched to Qwen2.5 Coder though, and when I was trying to get ollama to lock up yesterday and failing, I was using Qwen. I swear that it also happened with Qwen2.5, but it seemed to not happen yesterday. 🤷 I am using Qwen Coder 2.5 7b base at Q5_K_M, if you want to try that one too.

Good to see it was possible to reproduce with StarCoder2 though!

<!-- gh-comment-id:2493335547 --> @ProjectMoon commented on GitHub (Nov 22, 2024): I definitely noticed it with StarCoder2 first, which is the recommended model for use with Starhugger. I switched to Qwen2.5 Coder though, and when I was trying to get ollama to lock up yesterday and failing, I was using Qwen. I swear that it also happened with Qwen2.5, but it seemed to not happen yesterday. :shrug: I am using Qwen Coder 2.5 7b base at Q5_K_M, if you want to try that one too. Good to see it was possible to reproduce with StarCoder2 though!
GiteaMirror commented 2026-05-09 15:27:38 -05:00
Author
Owner
Copy Link

@mbbyn commented on GitHub (Jun 12, 2025):

This one still happens very frequently for us over many ollama upgrades, the latest being 0.7.1

Image

Image

Linux [REDACTED] 4.18.0-513.24.1.el8_9.x86_64 #1 SMP Thu Mar 14 14:20:09 EDT 2024 x86_64 x86_64 x86_64 GNU/Linux
<!-- gh-comment-id:2966491528 --> @mbbyn commented on GitHub (Jun 12, 2025): This one still happens very frequently for us over many ollama upgrades, the latest being 0.7.1 ![Image](https://github.com/user-attachments/assets/933163c7-d4ee-4a5d-b005-63919f5b3b17) ![Image](https://github.com/user-attachments/assets/89bebd59-2b12-4558-878a-dbd5681498b7) ``` Linux [REDACTED] 4.18.0-513.24.1.el8_9.x86_64 #1 SMP Thu Mar 14 14:20:09 EDT 2024 x86_64 x86_64 x86_64 GNU/Linux ```
GiteaMirror commented 2026-05-09 15:27:39 -05:00
Author
Owner
Copy Link

@necedema commented on GitHub (Jul 7, 2025):

Aye, it still happens occasionally for me as well.

ollama ps
NAME             ID              SIZE     PROCESSOR          UNTIL    
qwen3:30b-a3b    0b28110b7a33    20 GB    66%/34% CPU/GPU    Stopping...

ollama --version
ollama version is 0.9.5

uname -a
Linux 6.15.3 #1-NixOS SMP PREEMPT_DYNAMIC Thu Jun 19 13:41:08 UTC 2025 x86_64 GNU/Linux

nvidia-smi
Sun Jul  6 17:09:48 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 575.64                 Driver Version: 575.64         CUDA Version: 12.9     |
|-----------------------------------------+------------------------+----------------------+
| 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 3070 ...    Off |   00000000:01:00.0  On |                  N/A |
| N/A   43C    P8             17W /  115W |    6927MiB /   8192MiB |     13%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A            2429    C+G   ...-sway-unwrapped-1.11/bin/sway        112MiB |
|    0   N/A  N/A            3081      G   alacritty                                43MiB |
|    0   N/A  N/A            3142      G   ...-140.0.1/bin/.firefox-wrapped        497MiB |
|    0   N/A  N/A            4812      G   ....0.2/bin/.thunderbird-wrapped        120MiB |
+-----------------------------------------------------------------------------------------+
<!-- gh-comment-id:3043168525 --> @necedema commented on GitHub (Jul 7, 2025): Aye, it still happens occasionally for me as well. ``` ollama ps NAME ID SIZE PROCESSOR UNTIL qwen3:30b-a3b 0b28110b7a33 20 GB 66%/34% CPU/GPU Stopping... ``` ``` ollama --version ollama version is 0.9.5 ``` ``` uname -a Linux 6.15.3 #1-NixOS SMP PREEMPT_DYNAMIC Thu Jun 19 13:41:08 UTC 2025 x86_64 GNU/Linux ``` ``` nvidia-smi Sun Jul 6 17:09:48 2025 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 575.64 Driver Version: 575.64 CUDA Version: 12.9 | |-----------------------------------------+------------------------+----------------------+ | 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 3070 ... Off | 00000000:01:00.0 On | N/A | | N/A 43C P8 17W / 115W | 6927MiB / 8192MiB | 13% Default | | | | N/A | +-----------------------------------------+------------------------+----------------------+ +-----------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=========================================================================================| | 0 N/A N/A 2429 C+G ...-sway-unwrapped-1.11/bin/sway 112MiB | | 0 N/A N/A 3081 G alacritty 43MiB | | 0 N/A N/A 3142 G ...-140.0.1/bin/.firefox-wrapped 497MiB | | 0 N/A N/A 4812 G ....0.2/bin/.thunderbird-wrapped 120MiB | +-----------------------------------------------------------------------------------------+ ```
GiteaMirror commented 2026-05-09 15:27:40 -05:00
Author
Owner
Copy Link

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

Server logs may aid in debugging.

<!-- gh-comment-id:3044095172 --> @rick-github commented on GitHub (Jul 7, 2025): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) may aid in debugging.
GiteaMirror commented 2026-05-09 15:27:41 -05:00
Author
Owner
Copy Link

@kripper commented on GitHub (Aug 7, 2025):

Still happenning here too:

]# ollama -v
ollama version is 0.11.3


]# ollama ps
NAME           ID              SIZE     PROCESSOR    CONTEXT    UNTIL
gpt-oss:20b    f2b8351c629c    15 GB    100% CPU     8192       Stopping...


]# ps aux|grep ollam
ollama   1408918  0.0  0.0 2834648 66728 ?       Ssl  02:08   0:05 /usr/local/bin/ollama serve
ollama   1409269 1182  4.1 21455356 16434092 ?   Sl   02:13 1464:45 /usr/local/bin/ollama runner --ollama-engine --model /usr/share/ollama/.ollama/models/blobs/sha256-b112e727c6f18875636c56a779790a590d705aec9e1c0eb5a97d51fc2a778583 --ctx-size 8192 --batch-size 512 --threads 12 --no-mmap --parallel 1 --port 45983
root     1509510  0.0  0.0  12220  1212 pts/7    S+   04:17   0:00 grep --color=auto ollam

Please reopen.

<!-- gh-comment-id:3163043451 --> @kripper commented on GitHub (Aug 7, 2025): Still happenning here too: ``` ]# ollama -v ollama version is 0.11.3 ``` ``` ]# ollama ps NAME ID SIZE PROCESSOR CONTEXT UNTIL gpt-oss:20b f2b8351c629c 15 GB 100% CPU 8192 Stopping... ``` ``` ]# ps aux|grep ollam ollama 1408918 0.0 0.0 2834648 66728 ? Ssl 02:08 0:05 /usr/local/bin/ollama serve ollama 1409269 1182 4.1 21455356 16434092 ? Sl 02:13 1464:45 /usr/local/bin/ollama runner --ollama-engine --model /usr/share/ollama/.ollama/models/blobs/sha256-b112e727c6f18875636c56a779790a590d705aec9e1c0eb5a97d51fc2a778583 --ctx-size 8192 --batch-size 512 --threads 12 --no-mmap --parallel 1 --port 45983 root 1509510 0.0 0.0 12220 1212 pts/7 S+ 04:17 0:00 grep --color=auto ollam ``` Please reopen.
GiteaMirror commented 2026-05-09 15:27:43 -05:00
Author
Owner
Copy Link

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

Server logs may aid in debugging.

<!-- gh-comment-id:3163047604 --> @rick-github commented on GitHub (Aug 7, 2025): [Server logs](https://github.com/ollama/ollama/blob/main/docs/troubleshooting.md#how-to-troubleshoot-issues) may aid in debugging.
GiteaMirror commented 2026-05-09 15:27:45 -05:00
Author
Owner
Copy Link

@kripper commented on GitHub (Aug 7, 2025):

Still happenning here too:

Aug 07 02:08:55 test.com systemd[1]: Started Ollama Service.
Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.200-04:00 level=INFO source=routes.go:1297 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/usr/share/ollama/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]"
Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.215-04:00 level=INFO source=images.go:477 msg="total blobs: 72"
Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.220-04:00 level=INFO source=images.go:484 msg="total unused blobs removed: 0"
Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.222-04:00 level=INFO source=routes.go:1350 msg="Listening on 127.0.0.1:11434 (version 0.11.3)"
Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.222-04:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.253-04:00 level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered"
Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.253-04:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="377.3 GiB" available="246.0 GiB"
Aug 07 02:09:04 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:04 | 200 |     222.996µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:09:23 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:23 | 200 |    9.133904ms |       127.0.0.1 | GET      "/api/tags"
Aug 07 02:09:24 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:24 | 200 |     233.505µs |       127.0.0.1 | GET      "/api/ps"
Aug 07 02:09:28 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:28 | 200 |      67.168µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:09:38 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:38 | 200 |     113.847µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:10:25 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:10:25 | 200 |      112.05µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:12:56 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:12:56 | 200 |   10.686386ms |       127.0.0.1 | GET      "/api/tags"
Aug 07 02:12:56 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:12:56 | 200 |       91.67µs |       127.0.0.1 | GET      "/api/ps"
Aug 07 02:13:00 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:00 | 200 |     127.103µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:13:01 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:01 | 200 |    9.368537ms |       127.0.0.1 | GET      "/api/tags"
Aug 07 02:13:01 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:01 | 200 |      84.869µs |       127.0.0.1 | GET      "/api/ps"
Aug 07 02:13:43 test.com ollama[1408918]: time=2025-08-07T02:13:43.934-04:00 level=INFO source=server.go:135 msg="system memory" total="377.3 GiB" free="246.0 GiB" free_swap="0 B"
Aug 07 02:13:43 test.com ollama[1408918]: time=2025-08-07T02:13:43.935-04:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=25 layers.offload=0 layers.split="" memory.available="[246.1 GiB]" memory.gpu_overhead="0 B" memory.required.full="14.5 GiB" memory.required.partial="0 B" memory.required.kv="300.0 MiB" memory.required.allocations="[14.5 GiB]" memory.weights.total="11.7 GiB" memory.weights.repeating="10.7 GiB" memory.weights.nonrepeating="1.1 GiB" memory.graph.full="2.0 GiB" memory.graph.partial="2.0 GiB"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.163-04:00 level=INFO source=server.go:438 msg="starting llama server" cmd="/usr/local/bin/ollama runner --ollama-engine --model /usr/share/ollama/.ollama/models/blobs/sha256-b112e727c6f18875636c56a779790a590d705aec9e1c0eb5a97d51fc2a778583 --ctx-size 8192 --batch-size 512 --threads 12 --no-mmap --parallel 1 --port 45983"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.165-04:00 level=INFO source=sched.go:481 msg="loaded runners" count=1
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.165-04:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.166-04:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server not responding"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.196-04:00 level=INFO source=runner.go:925 msg="starting ollama engine"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.198-04:00 level=INFO source=runner.go:983 msg="Server listening on 127.0.0.1:45983"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.411-04:00 level=INFO source=ggml.go:92 msg="" architecture=gptoss file_type=MXFP4 name="" description="" num_tensors=315 num_key_values=30
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.419-04:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model"
Aug 07 02:13:44 test.com ollama[1408918]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-sandybridge.so
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.455-04:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc)
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.458-04:00 level=INFO source=ggml.go:367 msg="offloading 0 repeating layers to GPU"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.458-04:00 level=INFO source=ggml.go:371 msg="offloading output layer to CPU"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.458-04:00 level=INFO source=ggml.go:378 msg="offloaded 0/25 layers to GPU"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.458-04:00 level=INFO source=ggml.go:381 msg="model weights" buffer=CPU size="12.8 GiB"
Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.723-04:00 level=INFO source=ggml.go:672 msg="compute graph" backend=CPU buffer_type=CPU size="2.0 GiB"
Aug 07 02:13:52 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:52 | 200 |   11.316961ms |       127.0.0.1 | GET      "/api/tags"
Aug 07 02:13:52 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:52 | 200 |     205.944µs |       127.0.0.1 | GET      "/api/ps"
Aug 07 02:14:16 test.com ollama[1408918]: time=2025-08-07T02:14:16.351-04:00 level=INFO source=server.go:637 msg="llama runner started in 32.19 seconds"
Aug 07 02:14:54 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:14:54 | 200 |      63.191µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:15:04 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:15:04 | 200 |      61.809µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:15:27 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:15:27 | 200 |      81.695µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:16:06 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:06 | 200 |      82.663µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:16:22 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:22 | 200 |      75.726µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:16:33 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:33 | 200 |      65.331µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:16:42 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:42 | 200 |      64.228µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:16:45 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:45 | 200 |       61.72µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:16:48 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:48 | 200 |       63.05µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:16:53 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:53 | 200 |      62.658µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:16:56 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:56 | 200 |      59.861µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:16:59 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:59 | 200 |      67.694µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:17:08 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:17:08 | 200 |      62.172µs |       127.0.0.1 | GET      "/api/version"
Aug 07 02:22:48 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:48 | 200 |    5.369596ms |       127.0.0.1 | GET      "/api/tags"
Aug 07 02:22:48 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:48 | 200 |     123.983µs |       127.0.0.1 | GET      "/api/ps"
Aug 07 02:22:53 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:53 | 200 |    3.309931ms |       127.0.0.1 | GET      "/api/tags"
Aug 07 02:22:53 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:53 | 200 |      79.883µs |       127.0.0.1 | GET      "/api/ps"
Aug 07 02:22:57 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:57 | 200 |      70.308µs |       127.0.0.1 | GET      "/api/version"
Aug 07 03:49:13 test.com ollama[1408918]: [GIN] 2025/08/07 - 03:49:13 | 200 |      1h35m30s |       127.0.0.1 | POST     "/api/chat"
Aug 07 04:13:06 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:13:06 | 200 |     6.06126ms |       127.0.0.1 | GET      "/api/tags"
Aug 07 04:13:06 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:13:06 | 200 |      106.86µs |       127.0.0.1 | GET      "/api/ps"
Aug 07 04:13:10 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:13:10 | 200 |      70.114µs |       127.0.0.1 | GET      "/api/version"
Aug 07 04:13:17 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:13:17 | 200 |      62.369µs |       127.0.0.1 | GET      "/api/version"
Aug 07 04:15:54 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:15:54 | 200 |       46.39µs |       127.0.0.1 | HEAD     "/"
Aug 07 04:15:54 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:15:54 | 200 |      63.465µs |       127.0.0.1 | GET      "/api/ps"
Aug 07 04:17:56 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:17:56 | 200 |      66.803µs |       127.0.0.1 | GET      "/api/version"
Aug 07 04:18:55 test.com systemd[1]: Stopping Ollama Service...
Aug 07 04:18:55 test.com ollama[1408918]: time=2025-08-07T04:18:55.820-04:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:45983/completion\": context canceled"
Aug 07 04:18:55 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:18:55 | 500 |        29m41s |       127.0.0.1 | POST     "/api/chat"
Aug 07 04:18:56 test.com systemd[1]: ollama.service: Succeeded.
Aug 07 04:18:56 test.com systemd[1]: Stopped Ollama Service.
<!-- gh-comment-id:3163112449 --> @kripper commented on GitHub (Aug 7, 2025): > Still happenning here too: ``` Aug 07 02:08:55 test.com systemd[1]: Started Ollama Service. Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.200-04:00 level=INFO source=routes.go:1297 msg="server config" env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_CONTEXT_LENGTH:4096 OLLAMA_DEBUG:INFO OLLAMA_FLASH_ATTENTION:false OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://127.0.0.1:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE: OLLAMA_LLM_LIBRARY: OLLAMA_LOAD_TIMEOUT:5m0s OLLAMA_MAX_LOADED_MODELS:0 OLLAMA_MAX_QUEUE:512 OLLAMA_MODELS:/usr/share/ollama/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NEW_ENGINE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:1 OLLAMA_ORIGINS:[http://localhost https://localhost http://localhost:* https://localhost:* http://127.0.0.1 https://127.0.0.1 http://127.0.0.1:* https://127.0.0.1:* http://0.0.0.0 https://0.0.0.0 http://0.0.0.0:* https://0.0.0.0:* app://* file://* tauri://* vscode-webview://* vscode-file://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]" Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.215-04:00 level=INFO source=images.go:477 msg="total blobs: 72" Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.220-04:00 level=INFO source=images.go:484 msg="total unused blobs removed: 0" Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.222-04:00 level=INFO source=routes.go:1350 msg="Listening on 127.0.0.1:11434 (version 0.11.3)" Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.222-04:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.253-04:00 level=INFO source=gpu.go:377 msg="no compatible GPUs were discovered" Aug 07 02:08:55 test.com ollama[1408918]: time=2025-08-07T02:08:55.253-04:00 level=INFO source=types.go:130 msg="inference compute" id=0 library=cpu variant="" compute="" driver=0.0 name="" total="377.3 GiB" available="246.0 GiB" Aug 07 02:09:04 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:04 | 200 | 222.996µs | 127.0.0.1 | GET "/api/version" Aug 07 02:09:23 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:23 | 200 | 9.133904ms | 127.0.0.1 | GET "/api/tags" Aug 07 02:09:24 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:24 | 200 | 233.505µs | 127.0.0.1 | GET "/api/ps" Aug 07 02:09:28 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:28 | 200 | 67.168µs | 127.0.0.1 | GET "/api/version" Aug 07 02:09:38 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:09:38 | 200 | 113.847µs | 127.0.0.1 | GET "/api/version" Aug 07 02:10:25 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:10:25 | 200 | 112.05µs | 127.0.0.1 | GET "/api/version" Aug 07 02:12:56 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:12:56 | 200 | 10.686386ms | 127.0.0.1 | GET "/api/tags" Aug 07 02:12:56 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:12:56 | 200 | 91.67µs | 127.0.0.1 | GET "/api/ps" Aug 07 02:13:00 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:00 | 200 | 127.103µs | 127.0.0.1 | GET "/api/version" Aug 07 02:13:01 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:01 | 200 | 9.368537ms | 127.0.0.1 | GET "/api/tags" Aug 07 02:13:01 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:01 | 200 | 84.869µs | 127.0.0.1 | GET "/api/ps" Aug 07 02:13:43 test.com ollama[1408918]: time=2025-08-07T02:13:43.934-04:00 level=INFO source=server.go:135 msg="system memory" total="377.3 GiB" free="246.0 GiB" free_swap="0 B" Aug 07 02:13:43 test.com ollama[1408918]: time=2025-08-07T02:13:43.935-04:00 level=INFO source=server.go:175 msg=offload library=cpu layers.requested=-1 layers.model=25 layers.offload=0 layers.split="" memory.available="[246.1 GiB]" memory.gpu_overhead="0 B" memory.required.full="14.5 GiB" memory.required.partial="0 B" memory.required.kv="300.0 MiB" memory.required.allocations="[14.5 GiB]" memory.weights.total="11.7 GiB" memory.weights.repeating="10.7 GiB" memory.weights.nonrepeating="1.1 GiB" memory.graph.full="2.0 GiB" memory.graph.partial="2.0 GiB" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.163-04:00 level=INFO source=server.go:438 msg="starting llama server" cmd="/usr/local/bin/ollama runner --ollama-engine --model /usr/share/ollama/.ollama/models/blobs/sha256-b112e727c6f18875636c56a779790a590d705aec9e1c0eb5a97d51fc2a778583 --ctx-size 8192 --batch-size 512 --threads 12 --no-mmap --parallel 1 --port 45983" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.165-04:00 level=INFO source=sched.go:481 msg="loaded runners" count=1 Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.165-04:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.166-04:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server not responding" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.196-04:00 level=INFO source=runner.go:925 msg="starting ollama engine" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.198-04:00 level=INFO source=runner.go:983 msg="Server listening on 127.0.0.1:45983" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.411-04:00 level=INFO source=ggml.go:92 msg="" architecture=gptoss file_type=MXFP4 name="" description="" num_tensors=315 num_key_values=30 Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.419-04:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server loading model" Aug 07 02:13:44 test.com ollama[1408918]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-sandybridge.so Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.455-04:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 compiler=cgo(gcc) Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.458-04:00 level=INFO source=ggml.go:367 msg="offloading 0 repeating layers to GPU" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.458-04:00 level=INFO source=ggml.go:371 msg="offloading output layer to CPU" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.458-04:00 level=INFO source=ggml.go:378 msg="offloaded 0/25 layers to GPU" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.458-04:00 level=INFO source=ggml.go:381 msg="model weights" buffer=CPU size="12.8 GiB" Aug 07 02:13:44 test.com ollama[1408918]: time=2025-08-07T02:13:44.723-04:00 level=INFO source=ggml.go:672 msg="compute graph" backend=CPU buffer_type=CPU size="2.0 GiB" Aug 07 02:13:52 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:52 | 200 | 11.316961ms | 127.0.0.1 | GET "/api/tags" Aug 07 02:13:52 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:13:52 | 200 | 205.944µs | 127.0.0.1 | GET "/api/ps" Aug 07 02:14:16 test.com ollama[1408918]: time=2025-08-07T02:14:16.351-04:00 level=INFO source=server.go:637 msg="llama runner started in 32.19 seconds" Aug 07 02:14:54 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:14:54 | 200 | 63.191µs | 127.0.0.1 | GET "/api/version" Aug 07 02:15:04 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:15:04 | 200 | 61.809µs | 127.0.0.1 | GET "/api/version" Aug 07 02:15:27 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:15:27 | 200 | 81.695µs | 127.0.0.1 | GET "/api/version" Aug 07 02:16:06 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:06 | 200 | 82.663µs | 127.0.0.1 | GET "/api/version" Aug 07 02:16:22 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:22 | 200 | 75.726µs | 127.0.0.1 | GET "/api/version" Aug 07 02:16:33 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:33 | 200 | 65.331µs | 127.0.0.1 | GET "/api/version" Aug 07 02:16:42 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:42 | 200 | 64.228µs | 127.0.0.1 | GET "/api/version" Aug 07 02:16:45 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:45 | 200 | 61.72µs | 127.0.0.1 | GET "/api/version" Aug 07 02:16:48 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:48 | 200 | 63.05µs | 127.0.0.1 | GET "/api/version" Aug 07 02:16:53 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:53 | 200 | 62.658µs | 127.0.0.1 | GET "/api/version" Aug 07 02:16:56 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:56 | 200 | 59.861µs | 127.0.0.1 | GET "/api/version" Aug 07 02:16:59 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:16:59 | 200 | 67.694µs | 127.0.0.1 | GET "/api/version" Aug 07 02:17:08 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:17:08 | 200 | 62.172µs | 127.0.0.1 | GET "/api/version" Aug 07 02:22:48 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:48 | 200 | 5.369596ms | 127.0.0.1 | GET "/api/tags" Aug 07 02:22:48 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:48 | 200 | 123.983µs | 127.0.0.1 | GET "/api/ps" Aug 07 02:22:53 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:53 | 200 | 3.309931ms | 127.0.0.1 | GET "/api/tags" Aug 07 02:22:53 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:53 | 200 | 79.883µs | 127.0.0.1 | GET "/api/ps" Aug 07 02:22:57 test.com ollama[1408918]: [GIN] 2025/08/07 - 02:22:57 | 200 | 70.308µs | 127.0.0.1 | GET "/api/version" Aug 07 03:49:13 test.com ollama[1408918]: [GIN] 2025/08/07 - 03:49:13 | 200 | 1h35m30s | 127.0.0.1 | POST "/api/chat" Aug 07 04:13:06 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:13:06 | 200 | 6.06126ms | 127.0.0.1 | GET "/api/tags" Aug 07 04:13:06 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:13:06 | 200 | 106.86µs | 127.0.0.1 | GET "/api/ps" Aug 07 04:13:10 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:13:10 | 200 | 70.114µs | 127.0.0.1 | GET "/api/version" Aug 07 04:13:17 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:13:17 | 200 | 62.369µs | 127.0.0.1 | GET "/api/version" Aug 07 04:15:54 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:15:54 | 200 | 46.39µs | 127.0.0.1 | HEAD "/" Aug 07 04:15:54 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:15:54 | 200 | 63.465µs | 127.0.0.1 | GET "/api/ps" Aug 07 04:17:56 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:17:56 | 200 | 66.803µs | 127.0.0.1 | GET "/api/version" Aug 07 04:18:55 test.com systemd[1]: Stopping Ollama Service... Aug 07 04:18:55 test.com ollama[1408918]: time=2025-08-07T04:18:55.820-04:00 level=ERROR source=server.go:807 msg="post predict" error="Post \"http://127.0.0.1:45983/completion\": context canceled" Aug 07 04:18:55 test.com ollama[1408918]: [GIN] 2025/08/07 - 04:18:55 | 500 | 29m41s | 127.0.0.1 | POST "/api/chat" Aug 07 04:18:56 test.com systemd[1]: ollama.service: Succeeded. Aug 07 04:18:56 test.com systemd[1]: Stopped Ollama Service. ```
GiteaMirror commented 2026-05-09 15:27:49 -05:00
Author
Owner
Copy Link

@kripper commented on GitHub (Aug 7, 2025):

context canceled...It seems like it didn't finish, but I can see the info icon on OpenWebUI, which normally is only displayed after finishing.
The generated output was also ending in garbage. I'm testing a refactoring task of 3807 prompt tokens. Maybe this model is too small for this task.

<!-- gh-comment-id:3163145920 --> @kripper commented on GitHub (Aug 7, 2025): `context canceled`...It seems like it didn't finish, but I can see the info icon on OpenWebUI, which normally is only displayed after finishing. The generated output was also ending in garbage. I'm testing a refactoring task of 3807 prompt tokens. Maybe this model is too small for this task.
GiteaMirror commented 2026-05-09 15:27:51 -05:00
Author
Owner
Copy Link

@jiwanprasadguragain-stack commented on GitHub (Apr 14, 2026):

I am facing the same issue. Please help.

I am serving the qwen3.5:9b Q4KM model using ollama serve.
OS: Windows 11

When i updated the ollama to the current version, it was working correctly for a while, but then again it started acting like this.
Even when I terminate the ollama process and restart, ollama ps results get stuck at stopping.
And, my application just keeps on waiting till timeout.

Image

server logs:
Image

PS: I just left the application running and after 20 minutes or so, the model has been loaded properly.

Image
<!-- gh-comment-id:4242151938 --> @jiwanprasadguragain-stack commented on GitHub (Apr 14, 2026): I am facing the same issue. Please help. I am serving the qwen3.5:9b Q4KM model using ollama serve. OS: Windows 11 When i updated the ollama to the current version, it was working correctly for a while, but then again it started acting like this. Even when I terminate the ollama process and restart, ollama ps results get stuck at stopping. And, my application just keeps on waiting till timeout. <img width="678" height="123" alt="Image" src="https://github.com/user-attachments/assets/092f4ee7-6fce-4d9b-9fba-dcf48f68bb5f" /> server logs: <img width="1178" height="668" alt="Image" src="https://github.com/user-attachments/assets/22fbb48a-2f47-4dbd-b485-d9b956a36aad" /> PS: I just left the application running and after 20 minutes or so, the model has been loaded properly. <img width="770" height="226" alt="Image" src="https://github.com/user-attachments/assets/53ccee8b-6fa1-4a25-b13e-65a3f946a449" />
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 top
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#82653
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?