Pauses in opencode execution. #3124

Activity

1. github-actions commented on Oct 13, 2025

   

   github-actionsBot

   on Oct 13, 2025 – with GitHub Actions

   Contributor

   More actions

   This issue might be a duplicate of existing issues. Please check:

   • Performance noticeably drops with very long history #3094: Performance noticeably drops with very long history - describes performance issues and UI becoming jumpy during long operations
   • Opencode Silent on Reaching Provider Quota #3106: Opencode Silent on Reaching Provider Quota - mentions remaining in "generating" state without clear indication of completion
   • rg hangs indefinitely in 0.14.7 in a linux/amd64 container; works on 0.14.6 #3089: rg hangs indefinitely in 0.14.7 - reports hanging behavior in recent version

   Feel free to ignore if none of these address your specific case.
2. OpeOginni commented on Oct 13, 2025

   

   OpeOginni

   on Oct 13, 2025

   Contributor

   More actions

   What model and what type of operation example causes the issue?
3. thawkins commented on Oct 13, 2025

   

   thawkins

   on Oct 13, 2025 via email

   Author

   More actions

   I was using grok-fast-code-1 I was running an implementation pass on a phase of development after building a plan from a specification markdown file. I also been seeing some json errors coming back from the model I belive, I will try to capture one and

   …

   On Mon, Oct 13, 2025, 17:36 OpeOginni ***@***.***> wrote: *OpeOginni* left a comment (anomalyco/opencode#3124) <#3124 (comment)> What model and what type of operation example causes the issue? — Reply to this email directly, view it on GitHub <#3124 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAADRSMHGSBBPYFMT34ZTFT3XN6CBAVCNFSM6AAAAACJAGB2GGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTGOJWHA4TAOJSHE> . You are receiving this because you were mentioned.Message ID: ***@***.***>
4. OpeOginni commented on Oct 13, 2025

   

   OpeOginni

   on Oct 13, 2025

   Last edited by OpeOginni

   Contributor

   More actions

   This could not be a fix, but its still best practice, you can add your large lock file to be ignored by OpenCode, that way the model doesnt mistakenly use it as context, and spend forever on it.

   Create and add this to a opencode.json file in the root of the project

   {
       "$schema": "https://opencode.ai/config.json",
       "watcher": {
         "ignore": ["Cargo.lock"]
       }
     }
5. thawkins commented on Oct 13, 2025

   

   thawkins

   on Oct 13, 2025 via email

   Author

   More actions

   Is it worth ignoring the target folder too?

   …

   On Mon, Oct 13, 2025, 17:49 OpeOginni ***@***.***> wrote: *OpeOginni* left a comment (anomalyco/opencode#3124) <#3124 (comment)> Could not be the fix, but its still best practice, you can add your large lock file to be ignored by OpenCode, that way the model doesnt mistakenly use it as context, and spend forever on it. Create and add this to a opencode.json file in the root of the project { "$schema": "https://opencode.ai/config.json", "watcher": { "ignore": ["Cargo.lock"] } } — Reply to this email directly, view it on GitHub <#3124 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAADRSODHJHBWBKYLONJ7CL3XN7TRAVCNFSM6AAAAACJAGB2GGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTGOJWHE2TEMRXG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>
6. OpeOginni commented on Oct 13, 2025

   

   OpeOginni

   on Oct 13, 2025

   Contributor

   More actions

   @thawkins Depends on you, if it adds no important context to your agent, but its large enough to confuse it, then its best to add it as well.
7. rekram1-node commented on Oct 13, 2025

   

   rekram1-node

   on Oct 13, 2025

   Last edited by rekram1-node

   Contributor

   More actions

   @OpeOginni @thawkins the file watcher shouldn't be related here

   Firstly it is only enabled if you have set OPENCODE_EXPERIMENTAL_WATCHER to true and most people don't have that

   Also it doesn't have anything to do with context given to llm

   So no need to do this:

   Create and add this to a opencode.json file in the root of the project
   {
   "$schema": "https://opencode.ai/config.json",
   "watcher": {
   "ignore": ["Cargo.lock"]
   }
   }

   👍React with 👍1Reacted by OpeOginni
8. rekram1-node

   assigned

   thdxr

   on Oct 13, 2025
9. rekram1-node commented on Oct 13, 2025

   

   rekram1-node

   on Oct 13, 2025

   Contributor

   More actions

   @thawkins in regards to

   When performing large operations, the tool often seems to return to the "enter send" ready mode, and sit there for 30 seconds to a minute, it then returns to the "working .... " state. its often very hard to determine if the tool has finished working.

   Are you saying when the llm is making edits and stuff? Just trying to confirm what you mean by "large operations" here.

   There is a chance it could involve the rust lsp if you see long hangs on file edits to rust files which would probably be fixed by:

   {
     "$schema": "https://opencode.ai/config.json",
     "lsp": {
       "rust": {
         "disabled": true
       }
     }
   }
   

   But if it is just in general hangs and not limited to file edits then ignore this suggestion since there is no need to add that config needlessly
10. thdxr commented on Mar 25, 2026

    

    thdxr

    on Mar 25, 2026

    Member

    More actions

    To stay organized issues are automatically closed after 90 days of no activity. If the issue is still relevant please open a new one.
11. thdxr

    closed this as completedon Mar 25, 2026