Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 22:23:50 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #4415] [BUG]: Crashing Xorg with OOM error #2812

New Issue
Closed
opened 2026-02-22 18:31:22 -05:00 by yindo · 4 comments
yindo commented 2026-02-22 18:31:22 -05:00
Owner
Copy Link
Copy Source

Originally created by @thelifeandtimes on GitHub (Sep 19, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/4415

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

I am running AnythingLLM on Ubuntu 20.04 with an RTX 2080. If I leave it running (which I need to do for my use case), it seems to eventually crash with an error something like the following:

(191853.2408873 Out of memory: Killed process 2107 (Xorg) total-vm:58130864kB, anon-rss:29317536kB, f11e-rss:8508kB, shmen-rss:64396kB, UID:1000 pgtables:61996kB com score_adj:0

The machine it is running on has 32GB RAM and is running AnythingLLM, ollama w/ llama3.2, and a single urbit process. The ollama and urbit processes don't have GUIs.

Expected Behavior would be for AnythingLLM to be able to run forever without issue. I need to be able to use the machine remotely (over Tailscale and similar virtual networking) and the GUI is convenient when I am by the machine running it, but it is not strictly necessary. If the best fix to this is to use a dockerized AnythingLLM instance or something, happy to do so.

Are there known steps to reproduce?

Run AnythingLLM Desktop with ./start and wait for it to crash Xorg. might be 24-48 hours or so. maybe less.

Originally created by @thelifeandtimes on GitHub (Sep 19, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/4415 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? I am running AnythingLLM on Ubuntu 20.04 with an RTX 2080. If I leave it running (which I need to do for my use case), it seems to eventually crash with an error something like the following: ``` (191853.2408873 Out of memory: Killed process 2107 (Xorg) total-vm:58130864kB, anon-rss:29317536kB, f11e-rss:8508kB, shmen-rss:64396kB, UID:1000 pgtables:61996kB com score_adj:0 ``` The machine it is running on has 32GB RAM and is running AnythingLLM, ollama w/ llama3.2, and a single urbit process. The ollama and urbit processes don't have GUIs. Expected Behavior would be for AnythingLLM to be able to run forever without issue. I need to be able to use the machine remotely (over Tailscale and similar virtual networking) and the GUI is convenient when I am by the machine running it, but it is not strictly necessary. If the best fix to this is to use a dockerized AnythingLLM instance or something, happy to do so. ### Are there known steps to reproduce? Run AnythingLLM Desktop with `./start` and wait for it to crash Xorg. might be 24-48 hours or so. maybe less.
yindo added the possible bug label 2026-02-22 18:31:22 -05:00
yindo closed this issue 2026-02-22 18:31:22 -05:00
yindo commented 2026-02-22 18:31:23 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Sep 19, 2025):

If the best fix to this is to use a dockerized AnythingLLM instance or something, happy to do so.

This would indeed be the best fix, as well as also being more ideal for your use case. We have seen weird electron memleaks on Ubuntu/Linux AppImage.

That being said, do you do anything with the app or can it still crash by just booting and waiting? I can pull this in and see if we can see a growing heap or something. If the app is just static on the homepage doing nothing and growing larger and larger then that is indeed an issue

@timothycarambat commented on GitHub (Sep 19, 2025): > If the best fix to this is to use a dockerized AnythingLLM instance or something, happy to do so. This would indeed be the best fix, as well as also being more ideal for your use case. We have seen weird electron memleaks on Ubuntu/Linux AppImage. That being said, do you _do anything_ with the app or can it still crash by just booting and waiting? I can pull this in and see if we can see a growing heap or something. If the app is just static on the homepage doing nothing and growing larger and larger then that is indeed an issue
yindo commented 2026-02-22 18:31:23 -05:00
Author
Owner
Copy Link
Copy Source

@thelifeandtimes commented on GitHub (Sep 19, 2025):

I had created a workspace, loaded it with custom documents, and after rebooting and restarting perhaps had 2-3 chat messages with it? Definitely not a heavy workload. Appimage memory leak would make sense. Will look into dockerized version.

@thelifeandtimes commented on GitHub (Sep 19, 2025): I had created a workspace, loaded it with custom documents, and after rebooting and restarting perhaps had 2-3 chat messages with it? Definitely not a heavy workload. Appimage memory leak would make sense. Will look into dockerized version.
yindo commented 2026-02-22 18:31:23 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Sep 19, 2025):

Thanks for that, will repro that and see what happens to attempt debug. Using the built-in LLM provider that comes by default?

@timothycarambat commented on GitHub (Sep 19, 2025): Thanks for that, will repro that and see what happens to attempt debug. Using the built-in LLM provider that comes by default?
yindo commented 2026-02-22 18:31:23 -05:00
Author
Owner
Copy Link
Copy Source

@thelifeandtimes commented on GitHub (Sep 19, 2025):

Using a locally hosted llama3.2 model via ollama following these steps:

  1. Install https://ollama.com/ and set up Llama 3.2

  2. Install https://anythingllm.com/ and set up with Ollama and Llama 3.2

  3. In Anything LLM, create a new workspace called "Privateer"

  4. Download .txt file corpus and add it to Privateer workspace.

  5. AnythingLLM > Settings > Admin > Enable Network Discovery: On

  6. AnythingLLM > Developer API > Generate New API Key. Copy this entire key, including the word "Bearer".

  7. Navigate to http://localhost:3001/api/docs/. If this page successfully loads, it should be good to go.

@thelifeandtimes commented on GitHub (Sep 19, 2025): Using a locally hosted llama3.2 model via ollama following these steps: 1. Install https://ollama.com/ and set up Llama 3.2 2. Install https://anythingllm.com/ and set up with Ollama and Llama 3.2 3. In Anything LLM, create a new workspace called "Privateer" 4. Download .txt file corpus and add it to Privateer workspace. 5. AnythingLLM > Settings > Admin > Enable Network Discovery: On 6. AnythingLLM > Developer API > Generate New API Key. Copy this entire key, including the word "Bearer". 7. Navigate to http://localhost:3001/api/docs/. If this page successfully loads, it should be good to go.
yindo changed title from [BUG]: Crashing Xorg with OOM error to [GH-ISSUE #4415] [BUG]: Crashing Xorg with OOM error 2026-06-05 14:48:42 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label possible bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#2812
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?