github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-23 06:01:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #9209] timed out waiting for llama runner to start - progress 0.00 - #6001

New Issue
Closed
opened 2026-04-12 17:21:07 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-12 17:21:07 -05:00
Owner
Copy Link

Originally created by @forReason on GitHub (Feb 19, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/9209

What is the issue?

When I want to load a model using ollama, the GPU memory goes to 348mb and then stalls until I receive the timeout error.
Im running NVIDIA A40-48Q and the memory should be enough. I get to load Qwen 0.5b but the loading still seems way slower than expected. My home lab with an RTX2000 loads faster.

Im not exactly sure what the issue is. Any ideas what I can debug?

Relevant log output

-- Boot bb21c84b3f7f492ca36c727a2403af9f --
Feb 19 05:53:14 echdxtv1kig0001 systemd[1]: Started Ollama Service.
Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: 2025/02/19 05:53:14 routes.go:1186: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false 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:/home/aiadmin/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 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://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy:socks5h://127.0.0.1:1080 https_proxy:socks5h://127.0.0.1:1080 no_proxy:]"
Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:14.782Z level=INFO source=images.go:432 msg="total blobs: 16"
Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:14.783Z level=INFO source=images.go:439 msg="total unused blobs removed: 0"
Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:14.784Z level=INFO source=routes.go:1237 msg="Listening on 127.0.0.1:11434 (version 0.5.11)"
Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:14.788Z level=INFO source=gpu.go:217 msg="looking for compatible GPUs"
Feb 19 05:53:15 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:15.329Z level=INFO source=types.go:130 msg="inference compute" id=GPU-e1783cca-1df5-11b2-aad6-cefe98c46e11 library=cuda variant=v11 compute=8.6 driver=11.4 name="NVIDIA A40-48Q" total="47.7 GiB" available="43.9 GiB"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: [GIN] 2025/02/19 - 05:53:47 | 200 |      87.796µs |       127.0.0.1 | HEAD     "/"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: [GIN] 2025/02/19 - 05:53:47 | 200 |  113.645211ms |       127.0.0.1 | POST     "/api/show"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.445Z level=INFO source=sched.go:714 msg="new model will fit in available VRAM in single GPU, loading" model=/home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339 gpu=GPU-e1783cca-1df5-11b2-aad6-cefe98c46e11 parallel=4 available=47106818048 required="43.6 GiB"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.579Z level=INFO source=server.go:100 msg="system memory" total="251.7 GiB" free="249.4 GiB" free_swap="8.0 GiB"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.580Z level=INFO source=memory.go:356 msg="offload to cuda" layers.requested=-1 layers.model=81 layers.offload=81 layers.split="" memory.available="[43.9 GiB]" memory.gpu_overhead="0 B" memory.required.full="43.6 GiB" memory.required.partial="43.6 GiB" memory.required.kv="2.5 GiB" memory.required.allocations="[43.6 GiB]" memory.weights.total="40.7 GiB" memory.weights.repeating="39.9 GiB" memory.weights.nonrepeating="822.0 MiB" memory.graph.full="1.1 GiB" memory.graph.partial="1.1 GiB"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.580Z level=INFO source=server.go:380 msg="starting llama server" cmd="/usr/local/bin/ollama runner --model /home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339 --ctx-size 8192 --batch-size 512 --n-gpu-layers 81 --threads 8 --parallel 4 --port 37691"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.581Z level=INFO source=sched.go:449 msg="loaded runners" count=1
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.581Z level=INFO source=server.go:557 msg="waiting for llama runner to start responding"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.581Z level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.604Z level=INFO source=runner.go:936 msg="starting go runner"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.604Z level=INFO source=runner.go:937 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | cgo(gcc)" threads=8
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.604Z level=INFO source=runner.go:995 msg="Server listening on 127.0.0.1:37691"
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: ggml_cuda_init: GGML_CUDA_FORCE_MMQ:    no
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: ggml_cuda_init: found 1 CUDA devices:
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]:   Device 0: NVIDIA A40-48Q, compute capability 8.6, VMM: no
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: load_backend: loaded CUDA backend from /usr/local/lib/ollama/cuda_v11/libggml-cuda.so
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-icelake.so
Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.838Z level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model"
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_load_model_from_file: using device CUDA0 (NVIDIA A40-48Q) - 44834 MiB free
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: loaded meta data with 30 key-value pairs and 724 tensors from /home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339 (version GGUF V3 (latest))
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   0:                       general.architecture str              = llama
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   1:                               general.type str              = model
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   2:                               general.name str              = DeepSeek R1 Distill Llama 70B
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   3:                           general.basename str              = DeepSeek-R1-Distill-Llama
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   4:                         general.size_label str              = 70B
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   5:                          llama.block_count u32              = 80
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   6:                       llama.context_length u32              = 131072
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   7:                     llama.embedding_length u32              = 8192
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   8:                  llama.feed_forward_length u32              = 28672
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv   9:                 llama.attention.head_count u32              = 64
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  10:              llama.attention.head_count_kv u32              = 8
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  11:                       llama.rope.freq_base f32              = 500000.000000
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  12:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  13:                 llama.attention.key_length u32              = 128
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  14:               llama.attention.value_length u32              = 128
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  15:                          general.file_type u32              = 15
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  16:                           llama.vocab_size u32              = 128256
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  17:                 llama.rope.dimension_count u32              = 128
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  18:                       tokenizer.ggml.model str              = gpt2
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  19:                         tokenizer.ggml.pre str              = llama-bpe
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  20:                      tokenizer.ggml.tokens arr[str,128256]  = ["!", "\"", "#", "$", "%", "&", "'", ...
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  21:                  tokenizer.ggml.token_type arr[i32,128256]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  22:                      tokenizer.ggml.merges arr[str,280147]  = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  23:                tokenizer.ggml.bos_token_id u32              = 128000
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  24:                tokenizer.ggml.eos_token_id u32              = 128001
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  25:            tokenizer.ggml.padding_token_id u32              = 128001
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  26:               tokenizer.ggml.add_bos_token bool             = true
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  27:               tokenizer.ggml.add_eos_token bool             = false
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  28:                    tokenizer.chat_template str              = {% if not add_generation_prompt is de...
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv  29:               general.quantization_version u32              = 2
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - type  f32:  162 tensors
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - type q4_K:  441 tensors
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - type q5_K:   40 tensors
Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - type q6_K:   81 tensors
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_vocab: special tokens cache size = 256
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_vocab: token to piece cache size = 0.7999 MB
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: format           = GGUF V3 (latest)
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: arch             = llama
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: vocab type       = BPE
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_vocab          = 128256
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_merges         = 280147
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: vocab_only       = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_ctx_train      = 131072
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd           = 8192
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_layer          = 80
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_head           = 64
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_head_kv        = 8
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_rot            = 128
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_swa            = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd_head_k    = 128
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd_head_v    = 128
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_gqa            = 8
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd_k_gqa     = 1024
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd_v_gqa     = 1024
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_norm_eps       = 0.0e+00
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_norm_rms_eps   = 1.0e-05
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_clamp_kqv      = 0.0e+00
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_max_alibi_bias = 0.0e+00
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_logit_scale    = 0.0e+00
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_ff             = 28672
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_expert         = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_expert_used    = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: causal attn      = 1
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: pooling type     = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: rope type        = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: rope scaling     = linear
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: freq_base_train  = 500000.0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: freq_scale_train = 1
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_ctx_orig_yarn  = 131072
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: rope_finetuned   = unknown
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_d_conv       = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_d_inner      = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_d_state      = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_dt_rank      = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_dt_b_c_rms   = 0
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: model type       = 70B
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: model ftype      = Q4_K - Medium
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: model params     = 70.55 B
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: model size       = 39.59 GiB (4.82 BPW)
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: general.name     = DeepSeek R1 Distill Llama 70B
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: BOS token        = 128000 '<｜begin▁of▁sentence｜>'
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOS token        = 128001 '<｜end▁of▁sentence｜>'
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOT token        = 128009 '<|eot_id|>'
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOM token        = 128008 '<|eom_id|>'
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: PAD token        = 128001 '<｜end▁of▁sentence｜>'
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: LF token         = 128 'Ä'
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOG token        = 128001 '<｜end▁of▁sentence｜>'
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOG token        = 128008 '<|eom_id|>'
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOG token        = 128009 '<|eot_id|>'
Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: max token length = 256
Feb 19 05:58:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:58:47.704Z level=ERROR source=sched.go:455 msg="error loading llama server" error="timed out waiting for llama runner to start - progress 0.00 - "
Feb 19 05:58:47 echdxtv1kig0001 ollama[881]: [GIN] 2025/02/19 - 05:58:47 | 500 |          5m0s |       127.0.0.1 | POST     "/api/generate"
Feb 19 05:58:52 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:58:52.865Z level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.161645181 model=/home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339
Feb 19 05:58:53 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:58:53.115Z level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.411137856 model=/home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339
Feb 19 05:58:53 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:58:53.366Z level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.662011707 model=/home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.5.11

Originally created by @forReason on GitHub (Feb 19, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/9209 ### What is the issue? When I want to load a model using ollama, the GPU memory goes to 348mb and then stalls until I receive the timeout error. Im running NVIDIA A40-48Q and the memory should be enough. I get to load Qwen 0.5b but the loading still seems way slower than expected. My home lab with an RTX2000 loads faster. Im not exactly sure what the issue is. Any ideas what I can debug? ### Relevant log output ```shell -- Boot bb21c84b3f7f492ca36c727a2403af9f -- Feb 19 05:53:14 echdxtv1kig0001 systemd[1]: Started Ollama Service. Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: 2025/02/19 05:53:14 routes.go:1186: INFO server config env="map[CUDA_VISIBLE_DEVICES: GPU_DEVICE_ORDINAL: HIP_VISIBLE_DEVICES: HSA_OVERRIDE_GFX_VERSION: HTTPS_PROXY: HTTP_PROXY: NO_PROXY: OLLAMA_DEBUG:false 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:/home/aiadmin/.ollama/models OLLAMA_MULTIUSER_CACHE:false OLLAMA_NOHISTORY:false OLLAMA_NOPRUNE:false OLLAMA_NUM_PARALLEL:0 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://*] OLLAMA_SCHED_SPREAD:false ROCR_VISIBLE_DEVICES: http_proxy:socks5h://127.0.0.1:1080 https_proxy:socks5h://127.0.0.1:1080 no_proxy:]" Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:14.782Z level=INFO source=images.go:432 msg="total blobs: 16" Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:14.783Z level=INFO source=images.go:439 msg="total unused blobs removed: 0" Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:14.784Z level=INFO source=routes.go:1237 msg="Listening on 127.0.0.1:11434 (version 0.5.11)" Feb 19 05:53:14 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:14.788Z level=INFO source=gpu.go:217 msg="looking for compatible GPUs" Feb 19 05:53:15 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:15.329Z level=INFO source=types.go:130 msg="inference compute" id=GPU-e1783cca-1df5-11b2-aad6-cefe98c46e11 library=cuda variant=v11 compute=8.6 driver=11.4 name="NVIDIA A40-48Q" total="47.7 GiB" available="43.9 GiB" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: [GIN] 2025/02/19 - 05:53:47 | 200 | 87.796µs | 127.0.0.1 | HEAD "/" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: [GIN] 2025/02/19 - 05:53:47 | 200 | 113.645211ms | 127.0.0.1 | POST "/api/show" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.445Z level=INFO source=sched.go:714 msg="new model will fit in available VRAM in single GPU, loading" model=/home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339 gpu=GPU-e1783cca-1df5-11b2-aad6-cefe98c46e11 parallel=4 available=47106818048 required="43.6 GiB" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.579Z level=INFO source=server.go:100 msg="system memory" total="251.7 GiB" free="249.4 GiB" free_swap="8.0 GiB" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.580Z level=INFO source=memory.go:356 msg="offload to cuda" layers.requested=-1 layers.model=81 layers.offload=81 layers.split="" memory.available="[43.9 GiB]" memory.gpu_overhead="0 B" memory.required.full="43.6 GiB" memory.required.partial="43.6 GiB" memory.required.kv="2.5 GiB" memory.required.allocations="[43.6 GiB]" memory.weights.total="40.7 GiB" memory.weights.repeating="39.9 GiB" memory.weights.nonrepeating="822.0 MiB" memory.graph.full="1.1 GiB" memory.graph.partial="1.1 GiB" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.580Z level=INFO source=server.go:380 msg="starting llama server" cmd="/usr/local/bin/ollama runner --model /home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339 --ctx-size 8192 --batch-size 512 --n-gpu-layers 81 --threads 8 --parallel 4 --port 37691" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.581Z level=INFO source=sched.go:449 msg="loaded runners" count=1 Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.581Z level=INFO source=server.go:557 msg="waiting for llama runner to start responding" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.581Z level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server error" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.604Z level=INFO source=runner.go:936 msg="starting go runner" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.604Z level=INFO source=runner.go:937 msg=system info="CPU : LLAMAFILE = 1 | CPU : LLAMAFILE = 1 | cgo(gcc)" threads=8 Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.604Z level=INFO source=runner.go:995 msg="Server listening on 127.0.0.1:37691" Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: ggml_cuda_init: found 1 CUDA devices: Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: Device 0: NVIDIA A40-48Q, compute capability 8.6, VMM: no Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: load_backend: loaded CUDA backend from /usr/local/lib/ollama/cuda_v11/libggml-cuda.so Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: load_backend: loaded CPU backend from /usr/local/lib/ollama/libggml-cpu-icelake.so Feb 19 05:53:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:53:47.838Z level=INFO source=server.go:591 msg="waiting for server to become available" status="llm server loading model" Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_load_model_from_file: using device CUDA0 (NVIDIA A40-48Q) - 44834 MiB free Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: loaded meta data with 30 key-value pairs and 724 tensors from /home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339 (version GGUF V3 (latest)) Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output. Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 0: general.architecture str = llama Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 1: general.type str = model Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 2: general.name str = DeepSeek R1 Distill Llama 70B Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 3: general.basename str = DeepSeek-R1-Distill-Llama Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 4: general.size_label str = 70B Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 5: llama.block_count u32 = 80 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 6: llama.context_length u32 = 131072 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 7: llama.embedding_length u32 = 8192 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 8: llama.feed_forward_length u32 = 28672 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 9: llama.attention.head_count u32 = 64 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 10: llama.attention.head_count_kv u32 = 8 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 11: llama.rope.freq_base f32 = 500000.000000 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 12: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 13: llama.attention.key_length u32 = 128 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 14: llama.attention.value_length u32 = 128 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 15: general.file_type u32 = 15 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 16: llama.vocab_size u32 = 128256 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 17: llama.rope.dimension_count u32 = 128 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 18: tokenizer.ggml.model str = gpt2 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 19: tokenizer.ggml.pre str = llama-bpe Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 20: tokenizer.ggml.tokens arr[str,128256] = ["!", "\"", "#", "$", "%", "&", "'", ... Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 21: tokenizer.ggml.token_type arr[i32,128256] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ... Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 22: tokenizer.ggml.merges arr[str,280147] = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "... Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 23: tokenizer.ggml.bos_token_id u32 = 128000 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 24: tokenizer.ggml.eos_token_id u32 = 128001 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 25: tokenizer.ggml.padding_token_id u32 = 128001 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 26: tokenizer.ggml.add_bos_token bool = true Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 27: tokenizer.ggml.add_eos_token bool = false Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 28: tokenizer.chat_template str = {% if not add_generation_prompt is de... Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - kv 29: general.quantization_version u32 = 2 Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - type f32: 162 tensors Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - type q4_K: 441 tensors Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - type q5_K: 40 tensors Feb 19 05:53:51 echdxtv1kig0001 ollama[881]: llama_model_loader: - type q6_K: 81 tensors Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_vocab: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_vocab: special tokens cache size = 256 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_vocab: token to piece cache size = 0.7999 MB Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: format = GGUF V3 (latest) Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: arch = llama Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: vocab type = BPE Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_vocab = 128256 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_merges = 280147 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: vocab_only = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_ctx_train = 131072 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd = 8192 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_layer = 80 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_head = 64 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_head_kv = 8 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_rot = 128 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_swa = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd_head_k = 128 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd_head_v = 128 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_gqa = 8 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd_k_gqa = 1024 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_embd_v_gqa = 1024 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_norm_eps = 0.0e+00 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_norm_rms_eps = 1.0e-05 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_clamp_kqv = 0.0e+00 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_max_alibi_bias = 0.0e+00 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: f_logit_scale = 0.0e+00 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_ff = 28672 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_expert = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_expert_used = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: causal attn = 1 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: pooling type = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: rope type = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: rope scaling = linear Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: freq_base_train = 500000.0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: freq_scale_train = 1 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: n_ctx_orig_yarn = 131072 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: rope_finetuned = unknown Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_d_conv = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_d_inner = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_d_state = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_dt_rank = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: ssm_dt_b_c_rms = 0 Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: model type = 70B Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: model ftype = Q4_K - Medium Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: model params = 70.55 B Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: model size = 39.59 GiB (4.82 BPW) Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: general.name = DeepSeek R1 Distill Llama 70B Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: BOS token = 128000 '<｜begin▁of▁sentence｜>' Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOS token = 128001 '<｜end▁of▁sentence｜>' Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOT token = 128009 '<|eot_id|>' Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOM token = 128008 '<|eom_id|>' Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: PAD token = 128001 '<｜end▁of▁sentence｜>' Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: LF token = 128 'Ä' Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOG token = 128001 '<｜end▁of▁sentence｜>' Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOG token = 128008 '<|eom_id|>' Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: EOG token = 128009 '<|eot_id|>' Feb 19 05:53:52 echdxtv1kig0001 ollama[881]: llm_load_print_meta: max token length = 256 Feb 19 05:58:47 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:58:47.704Z level=ERROR source=sched.go:455 msg="error loading llama server" error="timed out waiting for llama runner to start - progress 0.00 - " Feb 19 05:58:47 echdxtv1kig0001 ollama[881]: [GIN] 2025/02/19 - 05:58:47 | 500 | 5m0s | 127.0.0.1 | POST "/api/generate" Feb 19 05:58:52 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:58:52.865Z level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.161645181 model=/home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339 Feb 19 05:58:53 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:58:53.115Z level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.411137856 model=/home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339 Feb 19 05:58:53 echdxtv1kig0001 ollama[881]: time=2025-02-19T05:58:53.366Z level=WARN source=sched.go:646 msg="gpu VRAM usage didn't recover within timeout" seconds=5.662011707 model=/home/aiadmin/.ollama/models/blobs/sha256-4cd576d9aa16961244012223abf01445567b061f1814b57dfef699e4cf8df339 ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.5.11
GiteaMirror added the bug label 2026-04-12 17:21:07 -05:00
GiteaMirror commented 2026-04-12 17:21:08 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Feb 19, 2025):

Set OLLAMA_LOAD_TIMEOUT=30m in the server environment.

<!-- gh-comment-id:2667958023 --> @rick-github commented on GitHub (Feb 19, 2025): Set `OLLAMA_LOAD_TIMEOUT=30m` in the server environment.
GiteaMirror commented 2026-04-12 17:21:08 -05:00
Author
Owner
Copy Link

@forReason commented on GitHub (Feb 19, 2025):

indeed.
long time nothing, then all of a sudden. Can this loading be improved somehow?

Update. I dont think the root cause is resolved. I am on powerful hardware and the model loading can take hours, when the same models on my consumer hardware at home take a couple of seconds to load...

<!-- gh-comment-id:2668562407 --> @forReason commented on GitHub (Feb 19, 2025): indeed. long time nothing, then all of a sudden. Can this loading be improved somehow? Update. I dont think the root cause is resolved. I am on powerful hardware and the model loading can take hours, when the same models on my consumer hardware at home take a couple of seconds to load...
GiteaMirror commented 2026-04-12 17:21:09 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Mar 4, 2025):

Cloud server? https://github.com/ollama/ollama/issues/9292#issuecomment-2676702560

<!-- gh-comment-id:2698237774 --> @rick-github commented on GitHub (Mar 4, 2025): Cloud server? https://github.com/ollama/ollama/issues/9292#issuecomment-2676702560
GiteaMirror commented 2026-04-12 17:21:09 -05:00
Author
Owner
Copy Link

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

@forReason also, what's your disk space look like?

<!-- gh-comment-id:2705314048 --> @mknwebsolutions commented on GitHub (Mar 7, 2025): @forReason also, what's your disk space look like?
GiteaMirror commented 2026-04-12 17:21:10 -05:00
Author
Owner
Copy Link

@forReason commented on GitHub (Mar 8, 2025):

Its on an ubuntu server. We discarded it. windows server loads fast. Reason unknown. its not an issue with ubuntu per se. Probably with the NAS storage-ubuntu connection at work

<!-- gh-comment-id:2708217276 --> @forReason commented on GitHub (Mar 8, 2025): Its on an ubuntu server. We discarded it. windows server loads fast. Reason unknown. its not an issue with ubuntu per se. Probably with the NAS storage-ubuntu connection at 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#6001
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?