github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #11778] gptoss-20b at 0.11.3 has invalid ggml type 39 #69866

New Issue
Closed
opened 2026-05-04 19:38:24 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-05-04 19:38:24 -05:00
Owner
Copy Link

Originally created by @Kinardus on GitHub (Aug 7, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/11778

What is the issue?

Good day.
I have been using ollama in a secure perimeter (without internet access) for a long time: I always updated according to the manual, downloaded GGUF models from hf.com on another machine and uploaded them via ollama create -f ./Modelfile. It worked always, but not with gpt-oss-20b. I get an error (log attached). I tried both GGUF from unsloth and from bartowski, and original weights safetensors. I see that it works for others - hence the question: what am I doing wrong?

Debian 12
ollama-linux-amd64.tgz
2x4060Ti
Models that i tryed from hf.com:

  • bartowski/openai_gpt-oss-20b-GGUF (Q8_0, BF16)
  • lmstudio-community/gpt-oss-20b-GGUF (MXFP4)
  • openai/gpt-oss-20b (safetensors)
  • unsloth/gpt-oss-20b-GGUF (Q8_0, BF16, F16)

Another nuance: I have a separate machine on Windows - there, ollama pull worked perfectly and the model starts. I even tried to transfer the blob data to Linux - the error is unchanged.

Relevant log output

Aug 07 11:18:18 llms01 systemd[1]: Started ollama.service - Ollama Service.
Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.880+03:00 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:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE:q8_0 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:true 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:true ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]"
Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.881+03:00 level=INFO source=images.go:477 msg="total blobs: 14"
Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.882+03:00 level=INFO source=images.go:484 msg="total unused blobs removed: 0"
Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.882+03:00 level=INFO source=routes.go:1350 msg="Listening on [::]:11434 (version 0.11.3)"
Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.883+03:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
Aug 07 11:18:19 llms01 ollama[1536]: time=2025-08-07T11:18:19.211+03:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-22e769d8-4d23-ead0-425d-77bf7ef299cf library=cuda variant=v12 compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4060 Ti" total="15.6 GiB" available="15.5 GiB"
Aug 07 11:18:19 llms01 ollama[1536]: time=2025-08-07T11:18:19.211+03:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-ce93c9d2-77e5-5fbe-91ee-a28768fb9bef library=cuda variant=v12 compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4060 Ti" total="15.6 GiB" available="15.5 GiB"
Aug 07 11:18:21 llms01 ollama[1536]: [GIN] 2025/08/07 - 11:18:21 | 200 |      95.101µs |       127.0.0.1 | HEAD     "/"
Aug 07 11:18:21 llms01 ollama[1536]: [GIN] 2025/08/07 - 11:18:21 | 200 |   49.595413ms |       127.0.0.1 | POST     "/api/show"
Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.183+03:00 level=INFO source=sched.go:802 msg="new model will fit in available VRAM, loading" model=/usr/share/ollama/.ollama/models/blobs/sha256-db9d08d2105a0cd9a6b03556595de60656c95df47780a43d0d5e51a2d51f826c library=cuda parallel=1 required="2.5 GiB"
Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.357+03:00 level=INFO source=server.go:135 msg="system memory" total="62.5 GiB" free="61.1 GiB" free_swap="977.0 MiB"
Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.530+03:00 level=INFO source=server.go:175 msg=offload library=cuda layers.requested=-1 layers.model=25 layers.offload=25 layers.split=13,12 memory.available="[15.5 GiB 15.5 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.5 GiB" memory.required.partial="2.5 GiB" memory.required.kv="96.0 MiB" memory.required.allocations="[1.6 GiB 1005.6 MiB]" memory.weights.total="1.2 GiB" memory.weights.repeating="680.5 MiB" memory.weights.nonrepeating="586.8 MiB" memory.graph.full="128.0 MiB" memory.graph.partial="128.0 MiB"
Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.530+03:00 level=INFO source=server.go:218 msg="enabling flash attention"
Aug 07 11:18:22 llms01 ollama[1536]: gguf_init_from_file_impl: tensor 'blk.0.ffn_down_exps.weight' has invalid ggml type 39 (NONE)
Aug 07 11:18:22 llms01 ollama[1536]: gguf_init_from_file_impl: failed to read tensor info
Aug 07 11:18:22 llms01 ollama[1536]: llama_model_load: error loading model: llama_model_loader: failed to load model from /usr/share/ollama/.ollama/models/blobs/sha256-db9d08d2105a0cd9a6b03556595de60656c95df47780a43d0d5e51a2d51f826c
Aug 07 11:18:22 llms01 ollama[1536]: llama_model_load_from_file_impl: failed to load model
Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.587+03:00 level=INFO source=sched.go:453 msg="NewLlamaServer failed" model=/usr/share/ollama/.ollama/models/blobs/sha256-db9d08d2105a0cd9a6b03556595de60656c95df47780a43d0d5e51a2d51f826c error="unable to load model: /usr/share/ollama/.ollama/models/blobs/sha256-db9d08d2105a0cd9a6b03556595de60656c95df47780a43d0d5e51a2d51f826c"
Aug 07 11:18:22 llms01 ollama[1536]: [GIN] 2025/08/07 - 11:18:22 | 500 |  832.548147ms |       127.0.0.1 | POST     "/api/generate"

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.11.3

Originally created by @Kinardus on GitHub (Aug 7, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/11778 ### What is the issue? Good day. I have been using ollama in a secure perimeter (without internet access) for a long time: I always updated according to the manual, downloaded GGUF models from hf.com on another machine and uploaded them via `ollama create -f ./Modelfile`. It worked always, but not with _gpt-oss-20b._ I get an error (log attached). I tried both GGUF from _unsloth_ and from _bartowski_, and _original weights safetensors_. I see that it works for others - hence the question: what am I doing wrong? Debian 12 ollama-linux-amd64.tgz 2x4060Ti Models that i tryed from hf.com: - bartowski/openai_gpt-oss-20b-GGUF (Q8_0, BF16) - lmstudio-community/gpt-oss-20b-GGUF (MXFP4) - openai/gpt-oss-20b (safetensors) - unsloth/gpt-oss-20b-GGUF (Q8_0, BF16, F16) Another nuance: I have a separate machine on Windows - there, `ollama pull` worked perfectly and the model starts. I even tried to transfer the blob data to Linux - the error is unchanged. ### Relevant log output ```shell Aug 07 11:18:18 llms01 systemd[1]: Started ollama.service - Ollama Service. Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.880+03:00 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:true OLLAMA_GPU_OVERHEAD:0 OLLAMA_HOST:http://0.0.0.0:11434 OLLAMA_INTEL_GPU:false OLLAMA_KEEP_ALIVE:5m0s OLLAMA_KV_CACHE_TYPE:q8_0 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:true 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:true ROCR_VISIBLE_DEVICES: http_proxy: https_proxy: no_proxy:]" Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.881+03:00 level=INFO source=images.go:477 msg="total blobs: 14" Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.882+03:00 level=INFO source=images.go:484 msg="total unused blobs removed: 0" Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.882+03:00 level=INFO source=routes.go:1350 msg="Listening on [::]:11434 (version 0.11.3)" Aug 07 11:18:18 llms01 ollama[1536]: time=2025-08-07T11:18:18.883+03:00 level=INFO source=gpu.go:217 msg="looking for compatible GPUs" Aug 07 11:18:19 llms01 ollama[1536]: time=2025-08-07T11:18:19.211+03:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-22e769d8-4d23-ead0-425d-77bf7ef299cf library=cuda variant=v12 compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4060 Ti" total="15.6 GiB" available="15.5 GiB" Aug 07 11:18:19 llms01 ollama[1536]: time=2025-08-07T11:18:19.211+03:00 level=INFO source=types.go:130 msg="inference compute" id=GPU-ce93c9d2-77e5-5fbe-91ee-a28768fb9bef library=cuda variant=v12 compute=8.9 driver=12.8 name="NVIDIA GeForce RTX 4060 Ti" total="15.6 GiB" available="15.5 GiB" Aug 07 11:18:21 llms01 ollama[1536]: [GIN] 2025/08/07 - 11:18:21 | 200 | 95.101µs | 127.0.0.1 | HEAD "/" Aug 07 11:18:21 llms01 ollama[1536]: [GIN] 2025/08/07 - 11:18:21 | 200 | 49.595413ms | 127.0.0.1 | POST "/api/show" Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.183+03:00 level=INFO source=sched.go:802 msg="new model will fit in available VRAM, loading" model=/usr/share/ollama/.ollama/models/blobs/sha256-db9d08d2105a0cd9a6b03556595de60656c95df47780a43d0d5e51a2d51f826c library=cuda parallel=1 required="2.5 GiB" Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.357+03:00 level=INFO source=server.go:135 msg="system memory" total="62.5 GiB" free="61.1 GiB" free_swap="977.0 MiB" Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.530+03:00 level=INFO source=server.go:175 msg=offload library=cuda layers.requested=-1 layers.model=25 layers.offload=25 layers.split=13,12 memory.available="[15.5 GiB 15.5 GiB]" memory.gpu_overhead="0 B" memory.required.full="2.5 GiB" memory.required.partial="2.5 GiB" memory.required.kv="96.0 MiB" memory.required.allocations="[1.6 GiB 1005.6 MiB]" memory.weights.total="1.2 GiB" memory.weights.repeating="680.5 MiB" memory.weights.nonrepeating="586.8 MiB" memory.graph.full="128.0 MiB" memory.graph.partial="128.0 MiB" Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.530+03:00 level=INFO source=server.go:218 msg="enabling flash attention" Aug 07 11:18:22 llms01 ollama[1536]: gguf_init_from_file_impl: tensor 'blk.0.ffn_down_exps.weight' has invalid ggml type 39 (NONE) Aug 07 11:18:22 llms01 ollama[1536]: gguf_init_from_file_impl: failed to read tensor info Aug 07 11:18:22 llms01 ollama[1536]: llama_model_load: error loading model: llama_model_loader: failed to load model from /usr/share/ollama/.ollama/models/blobs/sha256-db9d08d2105a0cd9a6b03556595de60656c95df47780a43d0d5e51a2d51f826c Aug 07 11:18:22 llms01 ollama[1536]: llama_model_load_from_file_impl: failed to load model Aug 07 11:18:22 llms01 ollama[1536]: time=2025-08-07T11:18:22.587+03:00 level=INFO source=sched.go:453 msg="NewLlamaServer failed" model=/usr/share/ollama/.ollama/models/blobs/sha256-db9d08d2105a0cd9a6b03556595de60656c95df47780a43d0d5e51a2d51f826c error="unable to load model: /usr/share/ollama/.ollama/models/blobs/sha256-db9d08d2105a0cd9a6b03556595de60656c95df47780a43d0d5e51a2d51f826c" Aug 07 11:18:22 llms01 ollama[1536]: [GIN] 2025/08/07 - 11:18:22 | 500 | 832.548147ms | 127.0.0.1 | POST "/api/generate" ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.11.3
GiteaMirror added the bug label 2026-05-04 19:38:24 -05:00
GiteaMirror commented 2026-05-04 19:38:29 -05:00
Author
Owner
Copy Link

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

Can you show the log of failure when you tried the blob data from the Windows machine?

<!-- gh-comment-id:3163180777 --> @rick-github commented on GitHub (Aug 7, 2025): Can you show the log of failure when you tried the blob data from the Windows machine?
GiteaMirror commented 2026-05-04 19:38:29 -05:00
Author
Owner
Copy Link

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

some here on macos

time=2025-08-07T17:39:56.175+08:00 level=INFO source=sched.go:786 msg="new model will fit in available VRAM in single GPU, loading" model=/Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7 gpu=0 parallel=1 available=52428800000 required="3.3 GiB"
time=2025-08-07T17:39:56.175+08:00 level=INFO source=server.go:135 msg="system memory" total="64.0 GiB" free="24.6 GiB" free_swap="0 B"
time=2025-08-07T17:39:56.175+08:00 level=INFO source=server.go:175 msg=offload library=metal layers.requested=-1 layers.model=25 layers.offload=25 layers.split="" memory.available="[48.8 GiB]" memory.gpu_overhead="0 B" memory.required.full="3.3 GiB" memory.required.partial="3.3 GiB" memory.required.kv="192.0 MiB" memory.required.allocations="[3.3 GiB]" memory.weights.total="2.3 GiB" memory.weights.repeating="1.2 GiB" memory.weights.nonrepeating="1.1 GiB" memory.graph.full="256.0 MiB" memory.graph.partial="256.0 MiB"
llama_model_load_from_file_impl: using device Metal (Apple M1 Max) - 49999 MiB free
gguf_init_from_file_impl: tensor 'blk.0.ffn_down_exps.weight' has invalid ggml type 39 (NONE)
gguf_init_from_file_impl: failed to read tensor info
llama_model_load: error loading model: llama_model_loader: failed to load model from /Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7

llama_model_load_from_file_impl: failed to load model
time=2025-08-07T17:39:56.233+08:00 level=INFO source=sched.go:453 msg="NewLlamaServer failed" model=/Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7 error="unable to load model: /Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7"
[GIN] 2025/08/07 - 17:39:56 | 500 | 147.638667ms | 127.0.0.1 | POST "/api/chat"

<!-- gh-comment-id:3163338297 --> @phonzia commented on GitHub (Aug 7, 2025): some here on macos > time=2025-08-07T17:39:56.175+08:00 level=INFO source=sched.go:786 msg="new model will fit in available VRAM in single GPU, loading" model=/Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7 gpu=0 parallel=1 available=52428800000 required="3.3 GiB" time=2025-08-07T17:39:56.175+08:00 level=INFO source=server.go:135 msg="system memory" total="64.0 GiB" free="24.6 GiB" free_swap="0 B" time=2025-08-07T17:39:56.175+08:00 level=INFO source=server.go:175 msg=offload library=metal layers.requested=-1 layers.model=25 layers.offload=25 layers.split="" memory.available="[48.8 GiB]" memory.gpu_overhead="0 B" memory.required.full="3.3 GiB" memory.required.partial="3.3 GiB" memory.required.kv="192.0 MiB" memory.required.allocations="[3.3 GiB]" memory.weights.total="2.3 GiB" memory.weights.repeating="1.2 GiB" memory.weights.nonrepeating="1.1 GiB" memory.graph.full="256.0 MiB" memory.graph.partial="256.0 MiB" llama_model_load_from_file_impl: using device Metal (Apple M1 Max) - 49999 MiB free gguf_init_from_file_impl: tensor 'blk.0.ffn_down_exps.weight' has invalid ggml type 39 (NONE) gguf_init_from_file_impl: failed to read tensor info llama_model_load: error loading model: llama_model_loader: failed to load model from /Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7 llama_model_load_from_file_impl: failed to load model time=2025-08-07T17:39:56.233+08:00 level=INFO source=sched.go:453 msg="NewLlamaServer failed" model=/Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7 error="unable to load model: /Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7" [GIN] 2025/08/07 - 17:39:56 | 500 | 147.638667ms | 127.0.0.1 | POST "/api/chat"
GiteaMirror commented 2026-05-04 19:38:30 -05:00
Author
Owner
Copy Link

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

Can you show the log of failure when you tried the blob data from the Windows machine?

No, I can't. I just transferred them again and the model started. Very strange, because yesterday I spent the whole day on this. BUT:

  1. Linux rebooted at night
  2. This morning in the ollama settings I enabled flash attention.
    I don`t know what helped. Blob from windows worked now, gguf & safetensors - no.
    This fresh log from safetensors:

Aug 07 13:33:30 llms01 ollama[1536]: [GIN] 2025/08/07 - 13:33:30 | 200 | 32.961µs | 127.0.0.1 | HEAD "/"
Aug 07 13:33:30 llms01 ollama[1536]: [GIN] 2025/08/07 - 13:33:30 | 200 | 56.24965ms | 127.0.0.1 | POST "/api/show"
Aug 07 13:33:30 llms01 ollama[1536]: time=2025-08-07T13:33:30.466+03:00 level=WARN source=memory.go:129 msg="model missing blk.0 layer size"
Aug 07 13:33:30 llms01 ollama[1536]: time=2025-08-07T13:33:30.848+03:00 level=INFO source=server.go:135 msg="system memory" total="62.5 GiB" free="40.8 GiB" free_swap="801.0 MiB"
Aug 07 13:33:30 llms01 ollama[1536]: time=2025-08-07T13:33:30.848+03:00 level=WARN source=memory.go:129 msg="model missing blk.0 layer size"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.042+03:00 level=INFO source=server.go:175 msg=offload library=cuda layers.requested=-1 layers.model=25 layers.offload=24 layers.split=12,12 memory.available="[15.5 GiB 15.5 GiB]" memory.gpu_overhead="0 B" memory.required.full="5.1 GiB" memory.required.partial="5.1 GiB" memory.required.kv="300.0 MiB" memory.required.allocations="[2.6 GiB 2.6 GiB]" memory.weights.total="0 B" memory.weights.repeating="0 B" memory.weights.nonrepeating="0 B" memory.graph.full="2.0 GiB" memory.graph.partial="2.0 GiB"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.042+03:00 level=WARN source=server.go:211 msg="flash attention enabled but not supported by model"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.042+03:00 level=WARN source=server.go:229 msg="quantized kv cache requested but flash attention disabled" type=q8_0
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.068+03:00 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/sha256-30e548d97202721de7d681ad630fcf9754cc7ae20623cce9c6bae8f063fc2edb --ctx-size 8192 --batch-size 512 --n-gpu-layers 24 --threads 8 --parallel 1 --tensor-split 12,12 --port 40159"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.069+03:00 level=INFO source=sched.go:481 msg="loaded runners" count=1
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.069+03:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.069+03:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server not responding"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.078+03:00 level=INFO source=runner.go:925 msg="starting ollama engine"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.079+03:00 level=INFO source=runner.go:983 msg="Server listening on 127.0.0.1:40159"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.111+03:00 level=INFO source=ggml.go:92 msg="" architecture=gptoss file_type=MXFP4 name="" description="" num_tensors=459 num_key_values=32
Aug 07 13:33:31 llms01 ollama[1536]: ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
Aug 07 13:33:31 llms01 ollama[1536]: ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
Aug 07 13:33:31 llms01 ollama[1536]: ggml_cuda_init: found 2 CUDA devices:
Aug 07 13:33:31 llms01 ollama[1536]: Device 0: NVIDIA GeForce RTX 4060 Ti, compute capability 8.9, VMM: yes
Aug 07 13:33:31 llms01 ollama[1536]: Device 1: NVIDIA GeForce RTX 4060 Ti, compute capability 8.9, VMM: yes
Aug 07 13:33:31 llms01 ollama[1536]: load_backend: loaded CUDA backend from /usr/local/lib/ollama/libggml-cuda.so
Aug 07 13:33:31 llms01 ollama[1536]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-alderlake.so
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.211+03:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 CUDA.1.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.1.USE_GRAPHS=1 CUDA.1.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc)
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:367 msg="offloading 24 repeating layers to GPU"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:371 msg="offloading output layer to CPU"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:378 msg="offloaded 24/25 layers to GPU"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:381 msg="model weights" buffer=CUDA0 size="10.1 GiB"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:381 msg="model weights" buffer=CUDA1 size="10.1 GiB"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:381 msg="model weights" buffer=CPU size="2.2 GiB"
Aug 07 13:33:31 llms01 ollama[1536]: panic: runtime error: invalid memory address or nil pointer dereference
Aug 07 13:33:31 llms01 ollama[1536]: [signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x55cd99aee726]
Aug 07 13:33:31 llms01 ollama[1536]: goroutine 42 [running]:
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/ml/nn.(*Embedding).Forward(...)
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/ml/nn/embedding.go:10
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/model/models/gptoss.(*Transformer).Forward(0xc0001936b0, {0x55cd9a9a8a90, 0xc0018305c0}, {{0x55cd9a9b36e8, 0xc00182f218}, {0x0, 0x0, 0x0}, {0xc001834800, 0x200, ...}, ...})
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/model/models/gptoss/model.go:32 +0x46
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner.(*Server).reserveWorstCaseGraph(0xc000114a20)
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner/runner.go:821 +0xac5
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner.(*Server).initModel(0xc000114a20, {0x7ffec69d1c81?, 0x0?}, {0x8, 0x0, 0x18, {0xc0001fd828, 0x2, 0x2}, 0x0}, ...)
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner/runner.go:865 +0x270
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner.(*Server).load(0xc000114a20, {0x55cd9a9a0790, 0xc0004f03c0}, {0x7ffec69d1c81?, 0x0?}, {0x8, 0x0, 0x18, {0xc0001fd828, 0x2, ...}, ...}, ...)
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner/runner.go:878 +0xb8
Aug 07 13:33:31 llms01 ollama[1536]: created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1
Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner/runner.go:959 +0xa11
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.343+03:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 2"
Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.569+03:00 level=ERROR source=sched.go:487 msg="error loading llama server" error="llama runner process has terminated: error:invalid memory address or nil pointer dereference\n[signal"
Aug 07 13:33:31 llms01 ollama[1536]: [GIN] 2025/08/07 - 13:33:31 | 500 | 1.40920778s | 127.0.0.1 | POST "/api/generate"
Aug 07 13:33:36 llms01 ollama[1536]: time=2025-08-07T13:33:36.770+03:00 level=WARN source=sched.go:685 msg="gpu VRAM usage didn't recover within timeout" seconds=5.200268118 runner.size="5.1 GiB" runner.vram="5.1 GiB" runner.parallel=1 runner.pid=85827 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-30e548d97202721de7d681ad630fcf9754cc7ae20623cce9c6bae8f063fc2edb
Aug 07 13:33:37 llms01 ollama[1536]: time=2025-08-07T13:33:37.020+03:00 level=WARN source=sched.go:685 msg="gpu VRAM usage didn't recover within timeout" seconds=5.450466782 runner.size="5.1 GiB" runner.vram="5.1 GiB" runner.parallel=1 runner.pid=85827 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-30e548d97202721de7d681ad630fcf9754cc7ae20623cce9c6bae8f063fc2edb
Aug 07 13:33:37 llms01 ollama[1536]: time=2025-08-07T13:33:37.270+03:00 level=WARN source=sched.go:685 msg="gpu VRAM usage didn't recover within timeout" seconds=5.7002688070000005 runner.size="5.1 GiB" runner.vram="5.1 GiB" runner.parallel=1 runner.pid=85827 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-30e548d97202721de7d681ad630fcf9754cc7ae20623cce9c6bae8f063fc2edb

<!-- gh-comment-id:3163516652 --> @Kinardus commented on GitHub (Aug 7, 2025): > Can you show the log of failure when you tried the blob data from the Windows machine? No, I can't. I just transferred them again and the model started. Very strange, because yesterday I spent the whole day on this. BUT: 1. Linux rebooted at night 2. This morning in the ollama settings I enabled flash attention. I don`t know what helped. Blob from windows worked now, gguf & safetensors - no. This fresh log from safetensors: > Aug 07 13:33:30 llms01 ollama[1536]: [GIN] 2025/08/07 - 13:33:30 | 200 | 32.961µs | 127.0.0.1 | HEAD "/" > Aug 07 13:33:30 llms01 ollama[1536]: [GIN] 2025/08/07 - 13:33:30 | 200 | 56.24965ms | 127.0.0.1 | POST "/api/show" > Aug 07 13:33:30 llms01 ollama[1536]: time=2025-08-07T13:33:30.466+03:00 level=WARN source=memory.go:129 msg="model missing blk.0 layer size" > Aug 07 13:33:30 llms01 ollama[1536]: time=2025-08-07T13:33:30.848+03:00 level=INFO source=server.go:135 msg="system memory" total="62.5 GiB" free="40.8 GiB" free_swap="801.0 MiB" > Aug 07 13:33:30 llms01 ollama[1536]: time=2025-08-07T13:33:30.848+03:00 level=WARN source=memory.go:129 msg="model missing blk.0 layer size" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.042+03:00 level=INFO source=server.go:175 msg=offload library=cuda layers.requested=-1 layers.model=25 layers.offload=24 layers.split=12,12 memory.available="[15.5 GiB 15.5 GiB]" memory.gpu_overhead="0 B" memory.required.full="5.1 GiB" memory.required.partial="5.1 GiB" memory.required.kv="300.0 MiB" memory.required.allocations="[2.6 GiB 2.6 GiB]" memory.weights.total="0 B" memory.weights.repeating="0 B" memory.weights.nonrepeating="0 B" memory.graph.full="2.0 GiB" memory.graph.partial="2.0 GiB" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.042+03:00 level=WARN source=server.go:211 msg="flash attention enabled but not supported by model" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.042+03:00 level=WARN source=server.go:229 msg="quantized kv cache requested but flash attention disabled" type=q8_0 > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.068+03:00 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/sha256-30e548d97202721de7d681ad630fcf9754cc7ae20623cce9c6bae8f063fc2edb --ctx-size 8192 --batch-size 512 --n-gpu-layers 24 --threads 8 --parallel 1 --tensor-split 12,12 --port 40159" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.069+03:00 level=INFO source=sched.go:481 msg="loaded runners" count=1 > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.069+03:00 level=INFO source=server.go:598 msg="waiting for llama runner to start responding" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.069+03:00 level=INFO source=server.go:632 msg="waiting for server to become available" status="llm server not responding" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.078+03:00 level=INFO source=runner.go:925 msg="starting ollama engine" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.079+03:00 level=INFO source=runner.go:983 msg="Server listening on 127.0.0.1:40159" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.111+03:00 level=INFO source=ggml.go:92 msg="" architecture=gptoss file_type=MXFP4 name="" description="" num_tensors=459 num_key_values=32 > Aug 07 13:33:31 llms01 ollama[1536]: ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no > Aug 07 13:33:31 llms01 ollama[1536]: ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no > Aug 07 13:33:31 llms01 ollama[1536]: ggml_cuda_init: found 2 CUDA devices: > Aug 07 13:33:31 llms01 ollama[1536]: Device 0: NVIDIA GeForce RTX 4060 Ti, compute capability 8.9, VMM: yes > Aug 07 13:33:31 llms01 ollama[1536]: Device 1: NVIDIA GeForce RTX 4060 Ti, compute capability 8.9, VMM: yes > Aug 07 13:33:31 llms01 ollama[1536]: load_backend: loaded CUDA backend from /usr/local/lib/ollama/libggml-cuda.so > Aug 07 13:33:31 llms01 ollama[1536]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-alderlake.so > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.211+03:00 level=INFO source=ggml.go:104 msg=system CPU.0.SSE3=1 CPU.0.SSSE3=1 CPU.0.AVX=1 CPU.0.AVX_VNNI=1 CPU.0.AVX2=1 CPU.0.F16C=1 CPU.0.FMA=1 CPU.0.BMI2=1 CPU.0.LLAMAFILE=1 CPU.1.LLAMAFILE=1 CUDA.0.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.0.USE_GRAPHS=1 CUDA.0.PEER_MAX_BATCH_SIZE=128 CUDA.1.ARCHS=500,600,610,700,750,800,860,870,890,900,1200 CUDA.1.USE_GRAPHS=1 CUDA.1.PEER_MAX_BATCH_SIZE=128 compiler=cgo(gcc) > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:367 msg="offloading 24 repeating layers to GPU" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:371 msg="offloading output layer to CPU" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:378 msg="offloaded 24/25 layers to GPU" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:381 msg="model weights" buffer=CUDA0 size="10.1 GiB" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:381 msg="model weights" buffer=CUDA1 size="10.1 GiB" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.272+03:00 level=INFO source=ggml.go:381 msg="model weights" buffer=CPU size="2.2 GiB" > Aug 07 13:33:31 llms01 ollama[1536]: panic: runtime error: invalid memory address or nil pointer dereference > Aug 07 13:33:31 llms01 ollama[1536]: [signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x55cd99aee726] > Aug 07 13:33:31 llms01 ollama[1536]: goroutine 42 [running]: > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/ml/nn.(*Embedding).Forward(...) > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/ml/nn/embedding.go:10 > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/model/models/gptoss.(*Transformer).Forward(0xc0001936b0, {0x55cd9a9a8a90, 0xc0018305c0}, {{0x55cd9a9b36e8, 0xc00182f218}, {0x0, 0x0, 0x0}, {0xc001834800, 0x200, ...}, ...}) > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/model/models/gptoss/model.go:32 +0x46 > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner.(*Server).reserveWorstCaseGraph(0xc000114a20) > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner/runner.go:821 +0xac5 > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner.(*Server).initModel(0xc000114a20, {0x7ffec69d1c81?, 0x0?}, {0x8, 0x0, 0x18, {0xc0001fd828, 0x2, 0x2}, 0x0}, ...) > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner/runner.go:865 +0x270 > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner.(*Server).load(0xc000114a20, {0x55cd9a9a0790, 0xc0004f03c0}, {0x7ffec69d1c81?, 0x0?}, {0x8, 0x0, 0x18, {0xc0001fd828, 0x2, ...}, ...}, ...) > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner/runner.go:878 +0xb8 > Aug 07 13:33:31 llms01 ollama[1536]: created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1 > Aug 07 13:33:31 llms01 ollama[1536]: github.com/ollama/ollama/runner/ollamarunner/runner.go:959 +0xa11 > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.343+03:00 level=ERROR source=server.go:464 msg="llama runner terminated" error="exit status 2" > Aug 07 13:33:31 llms01 ollama[1536]: time=2025-08-07T13:33:31.569+03:00 level=ERROR source=sched.go:487 msg="error loading llama server" error="llama runner process has terminated: error:invalid memory address or nil pointer dereference\n[signal" > Aug 07 13:33:31 llms01 ollama[1536]: [GIN] 2025/08/07 - 13:33:31 | 500 | 1.40920778s | 127.0.0.1 | POST "/api/generate" > Aug 07 13:33:36 llms01 ollama[1536]: time=2025-08-07T13:33:36.770+03:00 level=WARN source=sched.go:685 msg="gpu VRAM usage didn't recover within timeout" seconds=5.200268118 runner.size="5.1 GiB" runner.vram="5.1 GiB" runner.parallel=1 runner.pid=85827 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-30e548d97202721de7d681ad630fcf9754cc7ae20623cce9c6bae8f063fc2edb > Aug 07 13:33:37 llms01 ollama[1536]: time=2025-08-07T13:33:37.020+03:00 level=WARN source=sched.go:685 msg="gpu VRAM usage didn't recover within timeout" seconds=5.450466782 runner.size="5.1 GiB" runner.vram="5.1 GiB" runner.parallel=1 runner.pid=85827 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-30e548d97202721de7d681ad630fcf9754cc7ae20623cce9c6bae8f063fc2edb > Aug 07 13:33:37 llms01 ollama[1536]: time=2025-08-07T13:33:37.270+03:00 level=WARN source=sched.go:685 msg="gpu VRAM usage didn't recover within timeout" seconds=5.7002688070000005 runner.size="5.1 GiB" runner.vram="5.1 GiB" runner.parallel=1 runner.pid=85827 runner.model=/usr/share/ollama/.ollama/models/blobs/sha256-30e548d97202721de7d681ad630fcf9754cc7ae20623cce9c6bae8f063fc2edb
GiteaMirror commented 2026-05-04 19:38:31 -05:00
Author
Owner
Copy Link

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

@phonzia

model=/Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7

This model is not from the ollama library.

@Kinardus

msg="flash attention enabled but not supported by model"

Flash attention settings wouldn't change anything.

The likely problem is the model is not from the ollama library, so may have incompatibilities. #11714

<!-- gh-comment-id:3163634930 --> @rick-github commented on GitHub (Aug 7, 2025): @phonzia ``` model=/Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7 ``` This model is not from the ollama library. @Kinardus ``` msg="flash attention enabled but not supported by model" ``` Flash attention settings wouldn't change anything. The likely problem is the model is not from the ollama library, so may have incompatibilities. #11714
GiteaMirror commented 2026-05-04 19:38:32 -05:00
Author
Owner
Copy Link

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

@phonzia

model=/Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7

This model is not from the ollama library.

@Kinardus

msg="flash attention enabled but not supported by model"

Flash attention settings wouldn't change anything.

The likely problem is the model is not from the ollama library, so may have incompatibilities. #11714

Thank you very much for your work!

<!-- gh-comment-id:3163718540 --> @Kinardus commented on GitHub (Aug 7, 2025): > [@phonzia](https://github.com/phonzia) > > ``` > model=/Users/dx2880/.ollama/models/blobs/sha256-fcbc7ec4c2d1527c3da84b7049e59dc5af065876169216ec518bceab841e73f7 > ``` > > This model is not from the ollama library. > > [@Kinardus](https://github.com/Kinardus) > > ``` > msg="flash attention enabled but not supported by model" > ``` > > Flash attention settings wouldn't change anything. > > The likely problem is the model is not from the ollama library, so may have incompatibilities. [#11714](https://github.com/ollama/ollama/issues/11714) Thank you very much for your work!
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
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#69866
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?