convex down ?

[astral abyss]

getting this message for 114 function call fails : "Your request couldn't be completed. Try again later."
what does it mean ?

[ruby rune]

Hi @astral abyss. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:

---

That message is Convex’s generic “internal error” surface. It does not by itself mean your code is wrong; it means the request failed inside Convex (or on the network path) and Convex couldn’t complete it.

From the docs:

• Internal Convex errors are one of the four error categories, along with application, developer, and read/write limit errors.[Error handling]
• “If there are problems on our end, we'll automatically retry your queries and mutations until the problem is resolved and your queries and mutations succeed.”[Error handling]
• The user‑visible message in many of these cases is exactly what you’re seeing: “Your request couldn’t be completed. Try again later.” This same wording shows up in several support threads when there were transient backend or infrastructure issues.[Airbyte 500s][Locked dev env]

A few concrete implications:

1. It might be a transient Convex issue. There are examples where users hit 500s with that exact message and it started working again without code changes once Convex recovered.[Unable to connect]
2. It can also be caused by hitting limits inside a function, like:
   • Memory limits or other resource limits in actions (for heavy PDF processing, etc.). Users have seen that same message when their action appears to exhaust RAM.[10MB PDF thread]
   • Read/write limits on queries or mutations (those are described separately but still show up as internal failures if Convex aborts the run).[Read/write limits]
3. Sometimes it’s network‑level, e.g., TCP connection failures or 500s from the Convex HTTP API. Those also end up looking like “network error” / “your request couldn’t be completed” in the CLI.[Network test fail][BadDeployKey loop]

Given you see this on “114 function call fails,” possibilities are:

• A Convex incident or hiccup in your deployment region.
• A systemic issue in your code (e.g. a particular action that always hits memory or read/write limits).
• A network problem between you and your Convex deployment.

Because the docs and threads don’t give a single canonical cause for this exact text, I can’t be more specific without details of the failing functions. The recommended next steps based on the sources:

1. Check Convex status / try again later. Internal errors are automatically retried for queries and mutations; if it clears up after some time, it was likely a transient Convex issue.[Error handling]
2. Look at your Convex logs for those functions to see if they hit memory or read/write limits, or other exceptions. (The docs mention using log streaming or exception reporting for diagnosis, though they don’t show this specific message.)[Error handling]
3. If they’re actions, remember Convex won’t auto‑retry; your caller needs to handle errors, and you may need to split or offload heavy work if you’re hitting resource limits.[Actions error handling]
4. If this persists across many calls and projects, it might be like the “locked dev environment” or Airbyte 500 cases and may require Convex support to inspect your deployment.[Locked dev env][Airbyte 500s]

If you can share whether these 114 failures are actions vs queries/mutations, and roughly what the functions do (e.g., big file processing, large scans, many writes), I can map that more directly to the documented limits and behaviors.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon: