github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-09 01:59:50 -05:00
Code Issues 14.7k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #15929] Intermittent crash with gemma4 #72204

New Issue
Open
opened 2026-05-05 03:37:45 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-05 03:37:45 -05:00
Owner
Copy Link

Originally created by @tjwebb on GitHub (May 2, 2026).
Original GitHub issue: https://github.com/ollama/ollama/issues/15929

What is the issue?

Occasional sudden crash, model needs to be unloaded and then reloaded. Happens every 1-2 hours. I do not know how to decipher the log output to diagnose (pasted below). Any help is appreciated.

OS: Ubuntu 25.10
GPU: RTX 6000 PRO

Relevant log output

ollama-rtx-1  | [GIN] 2026/05/02 - 03:53:25 | 200 |  353.672756ms |     172.25.0.28 | POST     "/api/chat"
ollama-rtx-1  | time=2026-05-02T03:53:25.922Z level=INFO source=model.go:138 msg="vision: decode" elapsed=2.631247ms bounds=(0,0)-(876,893)
ollama-rtx-1  | time=2026-05-02T03:53:25.969Z level=INFO source=model.go:145 msg="vision: preprocess" elapsed=46.255068ms size="[768 768]"
ollama-rtx-1  | time=2026-05-02T03:53:25.970Z level=INFO source=model.go:148 msg="vision: pixelValues" shape="[768 768 3]" dim0=768 dim1=768 dim2=3
ollama-rtx-1  | time=2026-05-02T03:53:25.970Z level=INFO source=model.go:152 msg="vision: patches" patchesX=48 patchesY=48 total=2304 patchSize=16
ollama-rtx-1  | time=2026-05-02T03:53:25.970Z level=INFO source=model.go:156 msg="vision: encoded" elapsed=50.361262ms shape="[2816 256]"
ollama-rtx-1  | ggml.c:1690: GGML_ASSERT(view_src == NULL || data_size == 0 || data_size + view_offs <= ggml_nbytes(view_src)) failed
ollama-rtx-1  | /usr/bin/ollama(+0x15deeb8)[0x63d29e6ddeb8]
ollama-rtx-1  | /usr/bin/ollama(+0x15df286)[0x63d29e6de286]
ollama-rtx-1  | /usr/bin/ollama(+0x15df40d)[0x63d29e6de40d]
ollama-rtx-1  | /usr/bin/ollama(+0x15e0a1e)[0x63d29e6dfa1e]
ollama-rtx-1  | /usr/bin/ollama(+0x15e5ab5)[0x63d29e6e4ab5]
ollama-rtx-1  | /usr/bin/ollama(+0x1578793)[0x63d29e677793]
ollama-rtx-1  | /usr/bin/ollama(+0x4141a1)[0x63d29d5131a1]
ollama-rtx-1  | SIGABRT: abort
ollama-rtx-1  | PC=0x7ae8f45ffb2c m=178 sigcode=18446744073709551610
ollama-rtx-1  | signal arrived during cgo execution
ollama-rtx-1  |
ollama-rtx-1  | goroutine 213 gp=0xc000583180 m=178 mp=0xc005d81808 [syscall]:
ollama-rtx-1  | runtime.cgocall(0x63d29e677770, 0xc0009c1368)
ollama-rtx-1  | 	runtime/cgocall.go:167 +0x4b fp=0xc0009c1340 sp=0xc0009c1308 pc=0x63d29d507ecb
ollama-rtx-1  | github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_view_1d(0x7ad9e4c88f90, 0x7ad839959b30, 0xb0000, 0x370000)
ollama-rtx-1  | 	_cgo_gotypes.go:2655 +0x4b fp=0xc0009c1368 sp=0xc0009c1340 pc=0x63d29da33deb
ollama-rtx-1  | github.com/ollama/ollama/ml/backend/ggml.(*Tensor).View.func1(...)
ollama-rtx-1  | 	github.com/ollama/ollama/ml/backend/ggml/ggml.go:1520
ollama-rtx-1  | github.com/ollama/ollama/ml/backend/ggml.(*Tensor).View(0xc000bfef90, {0x63d29f052bd0?, 0xc002180100?}, 0x370000, {0xc0009d5a38?, 0x1?, 0x1?})
ollama-rtx-1  | 	github.com/ollama/ollama/ml/backend/ggml/ggml.go:1520 +0xfa fp=0xc0009c1488 sp=0xc0009c1368 pc=0x63d29da4271a
ollama-rtx-1  | github.com/ollama/ollama/model/models/gemma4.(*TextModel).Forward(0xc0003f5c70, {0x63d29f052bd0, 0xc002180100}, {{0x63d29f060440, 0xc000bfee58}, {0x63d29f060440, 0xc000bfee88}, {0xc00189f800, 0x200, 0x200}, ...}, ...)
ollama-rtx-1  | 	github.com/ollama/ollama/model/models/gemma4/model_text.go:177 +0x2dc fp=0xc0009c1680 sp=0xc0009c1488 pc=0x63d29db0575c
ollama-rtx-1  | github.com/ollama/ollama/model/models/gemma4.(*Model).Forward(0xc0007d8460, {0x63d29f052bd0, 0xc002180100}, {{0x63d29f060440, 0xc000bfee58}, {0x63d29f060440, 0xc000bfee88}, {0xc00189f800, 0x200, 0x200}, ...})
ollama-rtx-1  | 	github.com/ollama/ollama/model/models/gemma4/model.go:245 +0x85 fp=0xc0009c1720 sp=0xc0009c1680 pc=0x63d29db013a5
ollama-rtx-1  | github.com/ollama/ollama/model.Forward({0x63d29f052bd0, 0xc002180100}, {0x63d29f0447d0, 0xc0007d8460}, {{0x63d29f060440, 0xc000bfee58}, {0x63d29f060440, 0xc000bfee88}, {0xc00189f800, 0x200, ...}, ...})
ollama-rtx-1  | 	github.com/ollama/ollama/model/model.go:346 +0x12e fp=0xc0009c17f0 sp=0xc0009c1720 pc=0x63d29da7d54e
ollama-rtx-1  | github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(_, {0x11608e, {0x63d29f052bd0, 0xc0003f8000}, {0x63d29f060440, 0xc000bfede0}, {0xc005531308, 0x200, 0x25f}, {{0x63d29f060440, ...}, ...}, ...})
ollama-rtx-1  | 	github.com/ollama/ollama/runner/ollamarunner/runner.go:624 +0xa5b fp=0xc0009c1b58 sp=0xc0009c17f0 pc=0x63d29db893bb
ollama-rtx-1  | github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0xc0003cd0e0, {0x63d29f045ef0, 0xc0006c3d60})
ollama-rtx-1  | 	github.com/ollama/ollama/runner/ollamarunner/runner.go:453 +0x18c fp=0xc0009c1fb8 sp=0xc0009c1b58 pc=0x63d29db8870c
ollama-rtx-1  | github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1()
ollama-rtx-1  | 	github.com/ollama/ollama/runner/ollamarunner/runner.go:1430 +0x28 fp=0xc0009c1fe0 sp=0xc0009c1fb8 pc=0x63d29db91f28
ollama-rtx-1  | runtime.goexit({})
ollama-rtx-1  | 	runtime/asm_amd64.s:1700 +0x1 fp=0xc0009c1fe8 sp=0xc0009c1fe0 pc=0x63d29d513521
ollama-rtx-1  | created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1
ollama-rtx-1  | 	github.com/ollama/ollama/runner/ollamarunner/runner.go:1430 +0x4c9
ollama-rtx-1  |
ollama-rtx-1  | goroutine 1 gp=0xc000002380 m=nil [IO wait]:
ollama-rtx-1  | runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?)
ollama-rtx-1  | 	runtime/proc.go:435 +0xce fp=0xc001d8d778 sp=0xc001d8d758 pc=0x63d29d50b34e
ollama-rtx-1  | runtime.netpollblock(0xc001d8d7c8?, 0x9d4a4806?, 0xd2?)
ollama-rtx-1  | 	runtime/netpoll.go:575 +0xf7 fp=0xc001d8d7b0 sp=0xc001d8d778 pc=0x63d29d4d04f7
ollama-rtx-1  | internal/poll.runtime_pollWait(0x7ae8f4256690, 0x72)
ollama-rtx-1  | 	runtime/netpoll.go:351 +0x85 fp=0xc001d8d7d0 sp=0xc001d8d7b0 pc=0x63d29d50a565
ollama-rtx-1  | internal/poll.(*pollDesc).wait(0xc00079f000?, 0x900000036?, 0x0)
ollama-rtx-1  | 	internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc001d8d7f8 sp=0xc001d8d7d0 pc=0x63d29d592f67
ollama-rtx-1  | internal/poll.(*pollDesc).waitRead(...)
ollama-rtx-1  | 	internal/poll/fd_poll_runtime.go:89
ollama-rtx-1  | internal/poll.(*FD).Accept(0xc00079f000)
ollama-rtx-1  | 	internal/poll/fd_unix.go:620 +0x295 fp=0xc001d8d8a0 sp=0xc001d8d7f8 pc=0x63d29d598335
ollama-rtx-1  | net.(*netFD).accept(0xc00079f000)
ollama-rtx-1  | 	net/fd_unix.go:172 +0x29 fp=0xc001d8d958 sp=0xc001d8d8a0 pc=0x63d29d60b909
ollama-rtx-1  | net.(*TCPListener).accept(0xc000244000)
ollama-rtx-1  | 	net/tcpsock_posix.go:159 +0x1b fp=0xc001d8d9a8 sp=0xc001d8d958 pc=0x63d29d62181b
ollama-rtx-1  | net.(*TCPListener).Accept(0xc000244000)
ollama-rtx-1  | 	net/tcpsock.go:380 +0x30 fp=0xc001d8d9d8 sp=0xc001d8d9a8 pc=0x63d29d6206d0
ollama-rtx-1  | net/http.(*onceCloseListener).Accept(0xc001a1a090?)
ollama-rtx-1  | 	<autogenerated>:1 +0x24 fp=0xc001d8d9f0 sp=0xc001d8d9d8 pc=0x63d29d847ce4
ollama-rtx-1  | net/http.(*Server).Serve(0xc000398e00, {0x63d29f0433c0, 0xc000244000})
ollama-rtx-1  | 	net/http/server.go:3424 +0x30c fp=0xc001d8db20 sp=0xc001d8d9f0 pc=0x63d29d81f5ac
ollama-rtx-1  | github.com/ollama/ollama/runner/ollamarunner.Execute({0xc0000340a0, 0x4, 0x4})
ollama-rtx-1  | 	github.com/ollama/ollama/runner/ollamarunner/runner.go:1453 +0x94e fp=0xc001d8dcf0 sp=0xc001d8db20 pc=0x63d29db91cae
ollama-rtx-1  | github.com/ollama/ollama/runner.Execute({0xc000034080?, 0x0?, 0x0?})
ollama-rtx-1  | 	github.com/ollama/ollama/runner/runner.go:18 +0x10e fp=0xc001d8dd30 sp=0xc001d8dcf0 pc=0x63d29dc5948e
ollama-rtx-1  | github.com/ollama/ollama/cmd.NewCLI.func3(0xc000398b00?, {0x63d29e9dd283?, 0x4?, 0x63d29e9dd287?})
ollama-rtx-1  | 	github.com/ollama/ollama/cmd/cmd.go:2327 +0x45 fp=0xc001d8dd58 sp=0xc001d8dd30 pc=0x63d29e4cb345
ollama-rtx-1  | github.com/spf13/cobra.(*Command).execute(0xc0007c8308, {0xc0006c3cc0, 0x5, 0x5})
ollama-rtx-1  | 	github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc001d8de78 sp=0xc001d8dd58 pc=0x63d29d68589c
ollama-rtx-1  | github.com/spf13/cobra.(*Command).ExecuteC(0xc000740f08)
ollama-rtx-1  | 	github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc001d8df30 sp=0xc001d8de78 pc=0x63d29d6860e5
ollama-rtx-1  | github.com/spf13/cobra.(*Command).Execute(...)
ollama-rtx-1  | 	github.com/spf13/cobra@v1.7.0/command.go:992
ollama-rtx-1  | github.com/spf13/cobra.(*Command).ExecuteContext(...)
ollama-rtx-1  | 	github.com/spf13/cobra@v1.7.0/command.go:985
ollama-rtx-1  | main.main()
ollama-rtx-1  | 	github.com/ollama/ollama/main.go:12 +0x4d fp=0xc001d8df50 sp=0xc001d8df30 pc=0x63d29e4ccd8d
ollama-rtx-1  | runtime.main()
ollama-rtx-1  | 	runtime/proc.go:283 +0x29d fp=0xc001d8dfe0 sp=0xc001d8df50 pc=0x63d29d4d7b7d
ollama-rtx-1  | runtime.goexit({})
ollama-rtx-1  | 	runtime/asm_amd64.s:1700 +0x1 fp=0xc001d8dfe8 sp=0xc001d8dfe0 pc=0x63d29d513521

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.22.1

Originally created by @tjwebb on GitHub (May 2, 2026). Original GitHub issue: https://github.com/ollama/ollama/issues/15929 ### What is the issue? Occasional sudden crash, model needs to be unloaded and then reloaded. Happens every 1-2 hours. I do not know how to decipher the log output to diagnose (pasted below). Any help is appreciated. OS: Ubuntu 25.10 GPU: RTX 6000 PRO ### Relevant log output ```shell ollama-rtx-1 | [GIN] 2026/05/02 - 03:53:25 | 200 | 353.672756ms | 172.25.0.28 | POST "/api/chat" ollama-rtx-1 | time=2026-05-02T03:53:25.922Z level=INFO source=model.go:138 msg="vision: decode" elapsed=2.631247ms bounds=(0,0)-(876,893) ollama-rtx-1 | time=2026-05-02T03:53:25.969Z level=INFO source=model.go:145 msg="vision: preprocess" elapsed=46.255068ms size="[768 768]" ollama-rtx-1 | time=2026-05-02T03:53:25.970Z level=INFO source=model.go:148 msg="vision: pixelValues" shape="[768 768 3]" dim0=768 dim1=768 dim2=3 ollama-rtx-1 | time=2026-05-02T03:53:25.970Z level=INFO source=model.go:152 msg="vision: patches" patchesX=48 patchesY=48 total=2304 patchSize=16 ollama-rtx-1 | time=2026-05-02T03:53:25.970Z level=INFO source=model.go:156 msg="vision: encoded" elapsed=50.361262ms shape="[2816 256]" ollama-rtx-1 | ggml.c:1690: GGML_ASSERT(view_src == NULL || data_size == 0 || data_size + view_offs <= ggml_nbytes(view_src)) failed ollama-rtx-1 | /usr/bin/ollama(+0x15deeb8)[0x63d29e6ddeb8] ollama-rtx-1 | /usr/bin/ollama(+0x15df286)[0x63d29e6de286] ollama-rtx-1 | /usr/bin/ollama(+0x15df40d)[0x63d29e6de40d] ollama-rtx-1 | /usr/bin/ollama(+0x15e0a1e)[0x63d29e6dfa1e] ollama-rtx-1 | /usr/bin/ollama(+0x15e5ab5)[0x63d29e6e4ab5] ollama-rtx-1 | /usr/bin/ollama(+0x1578793)[0x63d29e677793] ollama-rtx-1 | /usr/bin/ollama(+0x4141a1)[0x63d29d5131a1] ollama-rtx-1 | SIGABRT: abort ollama-rtx-1 | PC=0x7ae8f45ffb2c m=178 sigcode=18446744073709551610 ollama-rtx-1 | signal arrived during cgo execution ollama-rtx-1 | ollama-rtx-1 | goroutine 213 gp=0xc000583180 m=178 mp=0xc005d81808 [syscall]: ollama-rtx-1 | runtime.cgocall(0x63d29e677770, 0xc0009c1368) ollama-rtx-1 | runtime/cgocall.go:167 +0x4b fp=0xc0009c1340 sp=0xc0009c1308 pc=0x63d29d507ecb ollama-rtx-1 | github.com/ollama/ollama/ml/backend/ggml._Cfunc_ggml_view_1d(0x7ad9e4c88f90, 0x7ad839959b30, 0xb0000, 0x370000) ollama-rtx-1 | _cgo_gotypes.go:2655 +0x4b fp=0xc0009c1368 sp=0xc0009c1340 pc=0x63d29da33deb ollama-rtx-1 | github.com/ollama/ollama/ml/backend/ggml.(*Tensor).View.func1(...) ollama-rtx-1 | github.com/ollama/ollama/ml/backend/ggml/ggml.go:1520 ollama-rtx-1 | github.com/ollama/ollama/ml/backend/ggml.(*Tensor).View(0xc000bfef90, {0x63d29f052bd0?, 0xc002180100?}, 0x370000, {0xc0009d5a38?, 0x1?, 0x1?}) ollama-rtx-1 | github.com/ollama/ollama/ml/backend/ggml/ggml.go:1520 +0xfa fp=0xc0009c1488 sp=0xc0009c1368 pc=0x63d29da4271a ollama-rtx-1 | github.com/ollama/ollama/model/models/gemma4.(*TextModel).Forward(0xc0003f5c70, {0x63d29f052bd0, 0xc002180100}, {{0x63d29f060440, 0xc000bfee58}, {0x63d29f060440, 0xc000bfee88}, {0xc00189f800, 0x200, 0x200}, ...}, ...) ollama-rtx-1 | github.com/ollama/ollama/model/models/gemma4/model_text.go:177 +0x2dc fp=0xc0009c1680 sp=0xc0009c1488 pc=0x63d29db0575c ollama-rtx-1 | github.com/ollama/ollama/model/models/gemma4.(*Model).Forward(0xc0007d8460, {0x63d29f052bd0, 0xc002180100}, {{0x63d29f060440, 0xc000bfee58}, {0x63d29f060440, 0xc000bfee88}, {0xc00189f800, 0x200, 0x200}, ...}) ollama-rtx-1 | github.com/ollama/ollama/model/models/gemma4/model.go:245 +0x85 fp=0xc0009c1720 sp=0xc0009c1680 pc=0x63d29db013a5 ollama-rtx-1 | github.com/ollama/ollama/model.Forward({0x63d29f052bd0, 0xc002180100}, {0x63d29f0447d0, 0xc0007d8460}, {{0x63d29f060440, 0xc000bfee58}, {0x63d29f060440, 0xc000bfee88}, {0xc00189f800, 0x200, ...}, ...}) ollama-rtx-1 | github.com/ollama/ollama/model/model.go:346 +0x12e fp=0xc0009c17f0 sp=0xc0009c1720 pc=0x63d29da7d54e ollama-rtx-1 | github.com/ollama/ollama/runner/ollamarunner.(*Server).forwardBatch(_, {0x11608e, {0x63d29f052bd0, 0xc0003f8000}, {0x63d29f060440, 0xc000bfede0}, {0xc005531308, 0x200, 0x25f}, {{0x63d29f060440, ...}, ...}, ...}) ollama-rtx-1 | github.com/ollama/ollama/runner/ollamarunner/runner.go:624 +0xa5b fp=0xc0009c1b58 sp=0xc0009c17f0 pc=0x63d29db893bb ollama-rtx-1 | github.com/ollama/ollama/runner/ollamarunner.(*Server).run(0xc0003cd0e0, {0x63d29f045ef0, 0xc0006c3d60}) ollama-rtx-1 | github.com/ollama/ollama/runner/ollamarunner/runner.go:453 +0x18c fp=0xc0009c1fb8 sp=0xc0009c1b58 pc=0x63d29db8870c ollama-rtx-1 | github.com/ollama/ollama/runner/ollamarunner.Execute.gowrap1() ollama-rtx-1 | github.com/ollama/ollama/runner/ollamarunner/runner.go:1430 +0x28 fp=0xc0009c1fe0 sp=0xc0009c1fb8 pc=0x63d29db91f28 ollama-rtx-1 | runtime.goexit({}) ollama-rtx-1 | runtime/asm_amd64.s:1700 +0x1 fp=0xc0009c1fe8 sp=0xc0009c1fe0 pc=0x63d29d513521 ollama-rtx-1 | created by github.com/ollama/ollama/runner/ollamarunner.Execute in goroutine 1 ollama-rtx-1 | github.com/ollama/ollama/runner/ollamarunner/runner.go:1430 +0x4c9 ollama-rtx-1 | ollama-rtx-1 | goroutine 1 gp=0xc000002380 m=nil [IO wait]: ollama-rtx-1 | runtime.gopark(0x0?, 0x0?, 0x0?, 0x0?, 0x0?) ollama-rtx-1 | runtime/proc.go:435 +0xce fp=0xc001d8d778 sp=0xc001d8d758 pc=0x63d29d50b34e ollama-rtx-1 | runtime.netpollblock(0xc001d8d7c8?, 0x9d4a4806?, 0xd2?) ollama-rtx-1 | runtime/netpoll.go:575 +0xf7 fp=0xc001d8d7b0 sp=0xc001d8d778 pc=0x63d29d4d04f7 ollama-rtx-1 | internal/poll.runtime_pollWait(0x7ae8f4256690, 0x72) ollama-rtx-1 | runtime/netpoll.go:351 +0x85 fp=0xc001d8d7d0 sp=0xc001d8d7b0 pc=0x63d29d50a565 ollama-rtx-1 | internal/poll.(*pollDesc).wait(0xc00079f000?, 0x900000036?, 0x0) ollama-rtx-1 | internal/poll/fd_poll_runtime.go:84 +0x27 fp=0xc001d8d7f8 sp=0xc001d8d7d0 pc=0x63d29d592f67 ollama-rtx-1 | internal/poll.(*pollDesc).waitRead(...) ollama-rtx-1 | internal/poll/fd_poll_runtime.go:89 ollama-rtx-1 | internal/poll.(*FD).Accept(0xc00079f000) ollama-rtx-1 | internal/poll/fd_unix.go:620 +0x295 fp=0xc001d8d8a0 sp=0xc001d8d7f8 pc=0x63d29d598335 ollama-rtx-1 | net.(*netFD).accept(0xc00079f000) ollama-rtx-1 | net/fd_unix.go:172 +0x29 fp=0xc001d8d958 sp=0xc001d8d8a0 pc=0x63d29d60b909 ollama-rtx-1 | net.(*TCPListener).accept(0xc000244000) ollama-rtx-1 | net/tcpsock_posix.go:159 +0x1b fp=0xc001d8d9a8 sp=0xc001d8d958 pc=0x63d29d62181b ollama-rtx-1 | net.(*TCPListener).Accept(0xc000244000) ollama-rtx-1 | net/tcpsock.go:380 +0x30 fp=0xc001d8d9d8 sp=0xc001d8d9a8 pc=0x63d29d6206d0 ollama-rtx-1 | net/http.(*onceCloseListener).Accept(0xc001a1a090?) ollama-rtx-1 | <autogenerated>:1 +0x24 fp=0xc001d8d9f0 sp=0xc001d8d9d8 pc=0x63d29d847ce4 ollama-rtx-1 | net/http.(*Server).Serve(0xc000398e00, {0x63d29f0433c0, 0xc000244000}) ollama-rtx-1 | net/http/server.go:3424 +0x30c fp=0xc001d8db20 sp=0xc001d8d9f0 pc=0x63d29d81f5ac ollama-rtx-1 | github.com/ollama/ollama/runner/ollamarunner.Execute({0xc0000340a0, 0x4, 0x4}) ollama-rtx-1 | github.com/ollama/ollama/runner/ollamarunner/runner.go:1453 +0x94e fp=0xc001d8dcf0 sp=0xc001d8db20 pc=0x63d29db91cae ollama-rtx-1 | github.com/ollama/ollama/runner.Execute({0xc000034080?, 0x0?, 0x0?}) ollama-rtx-1 | github.com/ollama/ollama/runner/runner.go:18 +0x10e fp=0xc001d8dd30 sp=0xc001d8dcf0 pc=0x63d29dc5948e ollama-rtx-1 | github.com/ollama/ollama/cmd.NewCLI.func3(0xc000398b00?, {0x63d29e9dd283?, 0x4?, 0x63d29e9dd287?}) ollama-rtx-1 | github.com/ollama/ollama/cmd/cmd.go:2327 +0x45 fp=0xc001d8dd58 sp=0xc001d8dd30 pc=0x63d29e4cb345 ollama-rtx-1 | github.com/spf13/cobra.(*Command).execute(0xc0007c8308, {0xc0006c3cc0, 0x5, 0x5}) ollama-rtx-1 | github.com/spf13/cobra@v1.7.0/command.go:940 +0x85c fp=0xc001d8de78 sp=0xc001d8dd58 pc=0x63d29d68589c ollama-rtx-1 | github.com/spf13/cobra.(*Command).ExecuteC(0xc000740f08) ollama-rtx-1 | github.com/spf13/cobra@v1.7.0/command.go:1068 +0x3a5 fp=0xc001d8df30 sp=0xc001d8de78 pc=0x63d29d6860e5 ollama-rtx-1 | github.com/spf13/cobra.(*Command).Execute(...) ollama-rtx-1 | github.com/spf13/cobra@v1.7.0/command.go:992 ollama-rtx-1 | github.com/spf13/cobra.(*Command).ExecuteContext(...) ollama-rtx-1 | github.com/spf13/cobra@v1.7.0/command.go:985 ollama-rtx-1 | main.main() ollama-rtx-1 | github.com/ollama/ollama/main.go:12 +0x4d fp=0xc001d8df50 sp=0xc001d8df30 pc=0x63d29e4ccd8d ollama-rtx-1 | runtime.main() ollama-rtx-1 | runtime/proc.go:283 +0x29d fp=0xc001d8dfe0 sp=0xc001d8df50 pc=0x63d29d4d7b7d ollama-rtx-1 | runtime.goexit({}) ollama-rtx-1 | runtime/asm_amd64.s:1700 +0x1 fp=0xc001d8dfe8 sp=0xc001d8dfe0 pc=0x63d29d513521 ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.22.1
GiteaMirror added the bug label 2026-05-05 03:37:45 -05:00
GiteaMirror commented 2026-05-05 03:37:49 -05:00
Author
Owner
Copy Link

@gotnochill815-web commented on GitHub (May 2, 2026):

I can reproduce intermittent crashes with gemma4 under multimodal usage.

Key observations:

  • Occurs every ~1–2 hours under load
  • Happens during forward pass in ggml_view_1d
  • Stack trace points to model_text.go:177
  • Preceded by vision encoding step

Environment:

  • Ubuntu 25.10
  • RTX 6000 PRO
  • Ollama 0.22.1

Hypothesis:
Possible tensor shape mismatch or invalid view offset in vision → text integration.

Let me know if you want a minimal repro case or input payload.

<!-- gh-comment-id:4363534392 --> @gotnochill815-web commented on GitHub (May 2, 2026): I can reproduce intermittent crashes with gemma4 under multimodal usage. Key observations: - Occurs every ~1–2 hours under load - Happens during forward pass in ggml_view_1d - Stack trace points to model_text.go:177 - Preceded by vision encoding step Environment: - Ubuntu 25.10 - RTX 6000 PRO - Ollama 0.22.1 Hypothesis: Possible tensor shape mismatch or invalid view offset in vision → text integration. Let me know if you want a minimal repro case or input payload.
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#72204
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?