Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 14:13:35 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #3222] [FEAT]: Fetch operations logging #2072

New Issue
Closed
opened 2026-02-22 18:28:02 -05:00 by yindo · 9 comments
yindo commented 2026-02-22 18:28:02 -05:00
Owner
Copy Link
Copy Source

Originally created by @ChGen on GitHub (Feb 15, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3222

What would you like to see?

Hello,
Could you please add logging to fetch operations and model files search/access operations?
I'm getting "Fetch failed" error for agents for web search and other agent operations due to some network issues, I've downloaded some Xenova/all-MiniLM-L6-v2 files, according to AnythingLLM guides, manually, and enabled Electron logging, but I don't see anything in the CLI output related to accessing particular model files or fetching specific files/URLs. So, I can't debug it out easily.
Also, if there any suggestions how to build debug version or open/attach Web Console for debugging running AnythingLLM instance, I would be grateful.

Thanks.

Originally created by @ChGen on GitHub (Feb 15, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3222 ### What would you like to see? Hello, Could you please add logging to fetch operations and model files search/access operations? I'm getting "Fetch failed" error for agents for web search and other agent operations due to some network issues, I've downloaded some `Xenova/all-MiniLM-L6-v2` files, according to AnythingLLM guides, manually, and enabled Electron logging, but I don't see anything in the CLI output related to accessing particular model files or fetching specific files/URLs. So, I can't debug it out easily. Also, if there any suggestions how to build debug version or open/attach Web Console for debugging running AnythingLLM instance, I would be grateful. Thanks.
yindo added the enhancementfeature request labels 2026-02-22 18:28:02 -05:00
yindo closed this issue 2026-02-22 18:28:02 -05:00
yindo commented 2026-02-22 18:28:02 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 16, 2025):

Does this not help debug those issues?
https://docs.anythingllm.com/installation-desktop/debug

You should see more verbose errors in the logs for the application showing what is the issue when trying to access external assets or resources

@timothycarambat commented on GitHub (Feb 16, 2025): Does this not help debug those issues? https://docs.anythingllm.com/installation-desktop/debug You should see more verbose errors in the logs for the application showing what is the issue when trying to access external assets or resources
yindo commented 2026-02-22 18:28:02 -05:00
Author
Owner
Copy Link
Copy Source

@ChGen commented on GitHub (Feb 17, 2025):

@timothycarambat , yes I run it from the command line with "ELECTRON_ENABLE_LOGGING=1" env set and "--enable-logging" argument provided to the program. With all this set, I haven't seen logs related to the missing files and URL fetch operations.

@ChGen commented on GitHub (Feb 17, 2025): @timothycarambat , yes I run it from the command line with "ELECTRON_ENABLE_LOGGING=1" env set and "--enable-logging" argument provided to the program. With all this set, I haven't seen logs related to the missing files and URL fetch operations.
yindo commented 2026-02-22 18:28:03 -05:00
Author
Owner
Copy Link
Copy Source

@ChGen commented on GitHub (Feb 17, 2025):

@timothycarambat , I've found the root of the problem, it's 5min request timeout of fetch api in node_modules/ollama/dist/browser.cjs in async post. AnythingLLM app doesn't handle request timeout errors and there's no settings for allowed timeout too.
Maybe my test LLM is a bit big, but still this hidden timeout error is unexpected. If you want, I can create new separate issue on timeouts.

@ChGen commented on GitHub (Feb 17, 2025): @timothycarambat , I've found the root of the problem, it's 5min request timeout of fetch api in `node_modules/ollama/dist/browser.cjs` in `async post`. AnythingLLM app doesn't handle request timeout errors and there's no settings for allowed timeout too. Maybe my test LLM is a bit big, but still this hidden timeout error is unexpected. If you want, I can create new separate issue on timeouts.
yindo commented 2026-02-22 18:28:03 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 17, 2025):

Are you trying to run a very large model on your device and the time to first token is taking longer than 5 minutes? That is a very very long time to spend waiting for a first token!

@timothycarambat commented on GitHub (Feb 17, 2025): Are you trying to run a very large model on your device and the time to first token is taking longer than 5 minutes? That is a very very long time to spend waiting for a first token!
yindo commented 2026-02-22 18:28:03 -05:00
Author
Owner
Copy Link
Copy Source

@ChGen commented on GitHub (Feb 17, 2025):

@timothycarambat , I just have several LLM models in lm-studio, so I decided to reuse them and check which one supports "tools calling". Since not every model supports functions calling / tools calling and its unclear which one should be downloaded for this. For simple requests they answer on 30 to 60 t/s rate on my system.
So, I see two issues:

  • silent timeout error with ollama api and no clear response/error logging
  • no recommendations in the app for agentic model in the list of the models. Probably it worth adding some 3b model marked as "agent tools recommend", which would work good enough for newbies like me.
@ChGen commented on GitHub (Feb 17, 2025): @timothycarambat , I just have several LLM models in lm-studio, so I decided to reuse them and check which one supports "tools calling". Since not every model supports functions calling / tools calling and its unclear which one should be downloaded for this. For simple requests they answer on 30 to 60 t/s rate on my system. So, I see two issues: * silent timeout error with ollama api and no clear response/error logging * no recommendations in the app for agentic model in the list of the models. Probably it worth adding some 3b model marked as "agent tools recommend", which would work good enough for newbies like me.
yindo commented 2026-02-22 18:28:03 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 17, 2025):

no recommendations in the app for agentic model in the list of the models. Probably it worth adding some 3b model marked as "agent tools recommend", which would work good enough for newbies like me.

Its true not all models support tooling calling, but that isn't the case with AnythingLLM - we actually have our own way to get around that to where even models that dont support tool calls traditionally can still leverage them.

The downside is that can take longer for tool calls to parse, since most models tend to be chatty and the agent system prompt is larger the more tools you toggle on!

silent timeout error with ollama api and no clear response/error logging

This is for sure an issue, should be more descriptive

@timothycarambat commented on GitHub (Feb 17, 2025): > no recommendations in the app for agentic model in the list of the models. Probably it worth adding some 3b model marked as "agent tools recommend", which would work good enough for newbies like me. Its true not all models support tooling calling, but that isn't the case with AnythingLLM - we actually have our own way to get around that to where even models that dont support tool calls traditionally can still leverage them. The downside is that can take longer for tool calls to parse, since most models tend to be chatty and the agent system prompt is larger the more tools you toggle on! > silent timeout error with ollama api and no clear response/error logging This is for sure an issue, should be more descriptive
yindo commented 2026-02-22 18:28:03 -05:00
Author
Owner
Copy Link
Copy Source

@ChGen commented on GitHub (Feb 17, 2025):

@timothycarambat When I ask "@agent ..." for internet search (enabled all features in the settings first) I get from some models response that it doesn't support that and it tries to provide facts from its knowledge. Am I missing something in general?

@ChGen commented on GitHub (Feb 17, 2025): @timothycarambat When I ask "@agent ..." for internet search (enabled all features in the settings first) I get from some models response that it doesn't support that and it tries to provide facts from its knowledge. Am I missing something in general?
yindo commented 2026-02-22 18:28:03 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 17, 2025):

https://docs.anythingllm.com/agent-not-using-tools#llm-says-it-cannot-call-xyz-tool
TLDR; a hallunication

@timothycarambat commented on GitHub (Feb 17, 2025): https://docs.anythingllm.com/agent-not-using-tools#llm-says-it-cannot-call-xyz-tool TLDR; a hallunication
yindo commented 2026-02-22 18:28:03 -05:00
Author
Owner
Copy Link
Copy Source

@ChGen commented on GitHub (Feb 18, 2025):

Thank you for your responses, I'll try other and smaller models with single tool enabled.

@ChGen commented on GitHub (Feb 18, 2025): Thank you for your responses, I'll try other and smaller models with single tool enabled.
yindo changed title from [FEAT]: Fetch operations logging to [GH-ISSUE #3222] [FEAT]: Fetch operations logging 2026-06-05 14:44:21 -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 enhancement feature request
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#2072
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?