### What version of Codex is running? codex-cli 0.80.0 ### What subscription do you have? Azure OpenAI API ### Which model were you using? _No response_ ### What platform is your computer? _No response_ ### What issue are you seeing? ```json { "error": { "message": "Invalid 'input[2].id': 'review:rollout:user'. Expected an ID that contains letters, numbers, underscores, or dashes, but this value contained additional characters.", "type": "invalid_request_error", "param":

Report Azure API Bug [Invalid 'input[2].id']

Resolved 💬 7 comments Opened Jan 13, 2026 by LeXwDeX Closed Jan 17, 2026

💡 Likely answer: A maintainer (etraut-openai, contributor) responded on this thread — see the highlighted reply below.

What version of Codex is running?

codex-cli 0.80.0

What subscription do you have?

Azure OpenAI API

Which model were you using?

_No response_

What platform is your computer?

_No response_

What issue are you seeing?

 {
  "error": {
    "message": "Invalid 'input[2].id': 'review:rollout:user'. Expected an ID that contains letters, numbers, underscores, or dashes, but this
value contained additional characters.",
    "type": "invalid_request_error",
    "param": "input[2].id",
    "code": "invalid_value"
  }
}

What steps can reproduce the bug?

Open CODEX-CLI.

In a project, trigger a “Code review.”

After the review output finishes, continue in the same session with a follow-up prompt / next task (e.g., “upgrade this session’s work to the latest framework” / “continue with an upgrade plan based on the review”).

When CODEX sends the next request, it immediately errors out and stops (instead of continuing to generate output).

What is the expected behavior?

_No response_

Additional information

_No response_

View original on GitHub ↗

7 Comments

etraut-openai contributor · 6 months ago

Could you please post your config settings for your model provider? Are you using the "responses" wire_api?

LeXwDeX · 6 months ago

Could you please post your config settings for your model provider? Are you using the "responses" wire_api?

Hi，this is my config

model = "gpt-5.1-codex-max"
model_provider = "azure"
model_reasoning_effort = "xhigh"

[model_providers.azure]
name = "Azure OpenAI"
base_url = "https://mydomain.openai.azure.com/openai/v1"
env_key = "AZURE_OPENAI_API_KEY"
wire_api = "responses"

[mcp_servers.sequential-thinking]
command = "npx"
args = ["-y", "@modelcontextprotocol/server-sequential-thinking"]

[mcp_servers.chrome-devtools]
command = "npx"
args = ["-y", "chrome-devtools-mcp@latest", "--executablePath=/usr/bin/chromium-browser", "--headless=true", "--isolated=true", "--chrome-arg=--no-sandbox", "--chrome-arg=--disable-setuid-sandbox", "--chrome-arg=--disable-dev-shm-usage", "--chrome-arg=--disable-gpu", "--logFile=/tmp/chrome-devtools-mcp.log"]
env = { PYTHONIOENCODING = "utf-8", REQUEST_TIMEOUT = "30", DEBUG = "*" }
startup_timeout_sec = 20
tool_timeout_sec = 120

[features]
unified_exec = true
shell_snapshot = true

etraut-openai contributor · 6 months ago

You're using two experimental features: unified_exec and shell_snapshot. If you disable these, does the problem go away?

LeXwDeX · 6 months ago

You're using two experimental features: unified_exec and shell_snapshot. If you disable these, does the problem go away?

I’ve been using it continuously for three hours, and so far I haven’t seen that error anymore.

jif-oai contributor · 6 months ago

Could you /feedback on the session and send us the session ID please?

LeXwDeX · 6 months ago

Could you /feedback on the session and send us the session ID please?

I’m sorry—I can’t find the specific conversation from when the bug occurred. But I can enable those two experimental parameters and keep using it; if I run into the bug again, I’ll report it to you.

etraut-openai contributor · 6 months ago

I think we understand this issue and have a proposed fix under review. Tracking in #9360, so I'm going to close this as duplicate.