# OpenAI network infrastructure and or bad network connection?

**URL:** <https://community.openai.com/t/openai-network-infrastructure-and-or-bad-network-connection/1396089>\
**Category:** Bugs\
**Created:** [September 9, 2026, 2:44pm UTC](https://community.openai.com/t/openai-network-infrastructure-and-or-bad-network-connection/1396089 "2026-09-09T14:44:16Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![none1](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/none1/32/751834_2.png) [@none1](https://community.openai.com/u/none1)\
**Post date:** [September 9, 2026, 2:44pm UTC](https://community.openai.com/t/openai-network-infrastructure-and-or-bad-network-connection/1396089/1 "2026-09-09T14:44:16Z")

</div>

I noticed that the network connection is really unstable and often drops; the way codex handles this is to silently re-send the whole previous conversation up to the point where the turns dropped. if it’s a wss connection after a few dropped wss connection retries; it falls back to http

This might seem reasonable to open ai if they have unlimited internal compute but is this really reasonable behavior? while it keeps the conversation flowing; there’s a few problems even on cache hits

it wastes time

you’re paying for work that wasn’t delivered

i see my usage get burned then i get interrupted connection drops; not only that its wasting time!!!

any api infrastructure engineers have anything to say about dealing with this?

---

<div class="post-metadata">

**Author:** ![EricGT](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/ericgt/32/20571_2.png) [@EricGT](https://community.openai.com/u/EricGT)\
**Post date:** [September 10, 2026, 1:19pm UTC](https://community.openai.com/t/openai-network-infrastructure-and-or-bad-network-connection/1396089/2 "2026-09-10T13:19:44Z")

</div>

I have been following this topic since it was posted, but until now there has been little I could do beyond observe.

Today, that changed.

I asked ChatGPT to search for the closest related issue, and it identified:

> <https://github.com/openai/codex/issues/43045>
>
> \### What version of Codex CLI is running?
> 
> codex-cli 0.153.4
> 
> \### What subscript…ion do you have?
> 
> ChatGPT Plus with purchased Codex credits
> 
> \### Which model were you using?
> 
> gpt-6-astra
> 
> \### What platform is your computer?
> 
> Microsoft Windows NT 10.0.26200.0 x64
> 
> \### What terminal emulator and version are you using (if applicable)?
> 
> VS Code
> 
> \### Codex doctor report
> 
> \`\`\`json
> {
> "schemaVersion": 1,
> "generatedAt": "1788623592s since unix epoch",
> "overallStatus": "warning",
> "codexVersion": "0.153.4",
> "checks": {
> "app\_server.status": {
> "id": "app\_server.status",
> "category": "app-server",
> "status": "ok",
> "summary": "background server is not running",
> "details": {
> "control socket": "C:\\\\Users\\\\okamo\\\\.codex\\\\app-server-control\\\\app-server-control.sock",
> "daemon state dir": "C:\\\\Users\\\\okamo\\\\.codex\\\\app-server-daemon",
> "mode": "ephemeral",
> "pid file": "C:\\\\Users\\\\okamo\\\\.codex\\\\app-server-daemon\\\\app-server.pid (missing)",
> "settings": "C:\\\\Users\\\\okamo\\\\.codex\\\\app-server-daemon\\\\settings.json (missing)",
> "status": "not running",
> "update-loop pid file": "C:\\\\Users\\\\okamo\\\\.codex\\\\app-server-daemon\\\\app-server-updater.pid (missing)"
> },
> "remediation": null,
> "durationMs": 0
> },
> "auth.credentials": {
> "id": "auth.credentials",
> "category": "auth",
> "status": "ok",
> "summary": "auth is configured",
> "details": {
> "auth env vars present": "OPENAI\_API\_KEY",
> "auth file": "C:\\\\Users\\\\okamo\\\\.codex\\\\auth.json",
> "auth storage mode": "File",
> "stored API key": "false",
> "stored ChatGPT tokens": "true",
> "stored agent identity": "false",
> "stored auth mode": "chatgpt"
> },
> "remediation": null,
> "durationMs": 0
> },
> "config.load": {
> "id": "config.load",
> "category": "config",
> "status": "ok",
> "summary": "config loaded",
> "details": {
> "CODEX\_HOME": "C:\\\\Users\\\\okamo\\\\.codex",
> "config.toml": "C:\\\\Users\\\\okamo\\\\.codex\\\\config.toml",
> "config.toml parse": "ok",
> "cwd": "C:\\\\repository\\\\auto-codex",
> "enabled feature flags": "\<redacted\>",
> "feature flag overrides": "none",
> "feature flags enabled": "48",
> "log dir": "C:\\\\Users\\\\okamo\\\\.codex\\\\log",
> "mcp servers": "1",
> "model": "gpt-6-astra",
> "model provider": "openai",
> "sqlite home": "C:\\\\Users\\\\okamo\\\\.codex"
> },
> "remediation": null,
> "durationMs": 0
> },
> "git.environment": {
> "id": "git.environment",
> "category": "git",
> "status": "ok",
> "summary": "git version 2.40.0.windows.1",
> "details": {
> ".git entry": "directory",
> "PATH git #1": "C:\\\\Program Files\\\\Git\\\\cmd\\\\git.exe",
> "PATH git entries": "1",
> "git branch": "\_c/20260905-1",
> "git build options": "git version 2.40.0.windows.1; cpu: x86\_64; built from commit: 1d90ca2906dd4b7ddaf0669a13c173ec579d794a; sizeof-long: 4; sizeof-size\_t: 8; shell-path: /bin/sh; feature: fsmonitor--daemon",
> "git exec path": "C:/Program Files/Git/mingw64/libexec/git-core",
> "git version": "git version 2.40.0.windows.1",
> "repo detected": "true",
> "repo root": "C:\\\\repository\\\\auto-codex",
> "selected git": "C:\\\\Program Files\\\\Git\\\\cmd\\\\git.exe"
> },
> "remediation": null,
> "durationMs": 51
> },
> "git.worktree.dev\_drive": {
> "id": "git.worktree.dev\_drive",
> "category": "git",
> "status": "warning",
> "summary": "this worktree is not on a Windows Dev Drive; moving it to a trusted Dev Drive can significantly improve repository and filesystem performance",
> "details": {},
> "remediation": "create a trusted Windows Dev Drive for source repositories: https://learn.microsoft.com/en-us/\<redacted\>",
> "durationMs": 0
> },
> "installation": {
> "id": "installation",
> "category": "install",
> "status": "ok",
> "summary": "installation looks consistent",
> "details": {
> "PATH codex #1": "C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\codex",
> "PATH codex #2": "C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\codex.cmd",
> "PATH codex entries": "2",
> "current executable": "C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc\\\\bin\\\\codex.exe",
> "install context": "npm (package C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc, bin C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc\\\\bin, resources C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc\\\\codex-resources, path C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc\\\\codex-path)",
> "managed by Vite+": "false",
> "managed by bun": "false",
> "managed by npm": "true",
> "managed by pnpm": "false",
> "managed package root": "C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex",
> "npm update target": "C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex"
> },
> "remediation": null,
> "durationMs": 131
> },
> "mcp.config": {
> "id": "mcp.config",
> "category": "mcp",
> "status": "ok",
> "summary": "MCP configuration is locally consistent",
> "details": {
> "configured servers": "1",
> "disabled servers": "0",
> "streamable\_http servers": "1"
> },
> "remediation": null,
> "durationMs": 93
> },
> "network.env": {
> "id": "network.env",
> "category": "network",
> "status": "ok",
> "summary": "network-related environment looks readable",
> "details": {
> "managed proxy": "not configured",
> "proxy env vars": "none",
> "respect system proxy": "disabled"
> },
> "remediation": null,
> "durationMs": 0
> },
> "network.provider\_reachability": {
> "id": "network.provider\_reachability",
> "category": "reachability",
> "status": "ok",
> "summary": "active provider endpoints are reachable over HTTP",
> "details": {
> "ChatGPT inference URL": "https://chatgpt.com/backend-api/\<redacted\> reachable (HTTP 405)",
> "reachability mode": "ChatGPT auth"
> },
> "remediation": null,
> "durationMs": 91
> },
> "network.websocket\_reachability": {
> "id": "network.websocket\_reachability",
> "category": "websocket",
> "status": "ok",
> "summary": "Responses WebSocket handshake succeeded",
> "details": {
> "DNS": "2 IPv4, 2 IPv6, first IPv6",
> "auth mode": "chatgpt",
> "connect timeout": "15000 ms",
> "endpoint": "wss://chatgpt.com/backend-api/\<redacted\>",
> "handshake result": "HTTP 101 Switching Protocols",
> "model provider": "openai",
> "provider name": "OpenAI",
> "proxy env vars": "none",
> "reasoning header": "false",
> "server model present": "false",
> "supports websockets": "true",
> "wire API": "responses"
> },
> "remediation": null,
> "durationMs": 835
> },
> "runtime.provenance": {
> "id": "runtime.provenance",
> "category": "runtime",
> "status": "ok",
> "summary": "running npm on windows-x86\_64",
> "details": {
> "commit": "unknown",
> "current executable": "C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc\\\\bin\\\\codex.exe",
> "install method": "npm (package C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc, bin C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc\\\\bin, resources C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc\\\\codex-resources, path C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc\\\\codex-path)",
> "platform": "windows-x86\_64",
> "version": "0.153.4"
> },
> "remediation": null,
> "durationMs": 0
> },
> "runtime.search": {
> "id": "runtime.search",
> "category": "search",
> "status": "ok",
> "summary": "search is OK (bundled)",
> "details": {
> "search command": "C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex\\\\node\_modules\\\\@openai\\\\codex-win32-x64\\\\vendor\\\\x86\_64-pc-windows-msvc\\\\codex-path\\\\rg.exe",
> "search command readiness": "file exists",
> "search provider": "bundled"
> },
> "remediation": null,
> "durationMs": 0
> },
> "sandbox.helpers": {
> "id": "sandbox.helpers",
> "category": "sandbox",
> "status": "ok",
> "summary": "sandbox configuration is readable",
> "details": {
> "approval policy": "Never",
> "codex-linux-sandbox helper": "none",
> "denied-read restrictions": "false",
> "execve wrapper helper": "none",
> "filesystem sandbox": "unrestricted",
> "network sandbox": "enabled",
> "sandbox backend": "\<redacted\>"
> },
> "remediation": null,
> "durationMs": 0
> },
> "security.endpoint": {
> "id": "security.endpoint",
> "category": "security",
> "status": "warning",
> "summary": "endpoint protection detected; Codex exclusions are unverified",
> "details": {
> "Codex exclusions": "not verified",
> "endpoint products": "Microsoft Defender",
> "exclusion targets": "signed Codex app; codex.exe; codex-windows-sandbox-setup.exe; codex-command-runner.exe; codex-code-mode-host.exe" 
> },
> "issues": \[
> {
> "severity": "warning",
> "cause": "Microsoft Defender can interfere with Codex. Verify Codex exclusions.",
> "measured": "not verified",
> "expected": "Codex application and helper exclusions",
> "remedy": "Microsoft Defender: Add a certificate or executable-path exclusion for Codex and its helpers. If Attack Surface Reduction blocks Codex, add a rule exclusion. If Controlled Folder Access blocks Codex, allow the app.",
> "fields": \[
> "Codex exclusions"
> \]
> }
> \],
> "remediation": "ask your security administrator to verify Codex exclusions and required helper allowances",
> "durationMs": 11
> },
> "state.paths": {
> "id": "state.paths",
> "category": "state",
> "status": "ok",
> "summary": "state paths and databases are inspectable",
> "details": {
> "CODEX\_HOME": "C:\\\\Users\\\\okamo\\\\.codex (dir)",
> "active rollout files": "3546 files, 2411753571 total bytes, 680133 average bytes",
> "archived rollout files": "0 files, 0 total bytes, 0 average bytes",
> "goals DB": "C:\\\\Users\\\\okamo\\\\.codex\\\\goals\_1.sqlite (file)",
> "goals DB integrity": "ok",
> "log DB": "C:\\\\Users\\\\okamo\\\\.codex\\\\logs\_2.sqlite (file)",
> "log DB integrity": "ok",
> "log dir": "C:\\\\Users\\\\okamo\\\\.codex\\\\log (dir)",
> "memories DB": "C:\\\\Users\\\\okamo\\\\.codex\\\\memories\_1.sqlite (file)",
> "memories DB integrity": "ok",
> "queue DB": "C:\\\\Users\\\\okamo\\\\.codex\\\\queue\_1.sqlite (file)",
> "queue DB integrity": "ok",
> "sqlite home": "C:\\\\Users\\\\okamo\\\\.codex (dir)",
> "state DB": "C:\\\\Users\\\\okamo\\\\.codex\\\\state\_5.sqlite (file)",
> "state DB integrity": "ok",
> "thread history DB": "C:\\\\Users\\\\okamo\\\\.codex\\\\thread\_history\_1.sqlite (file)",
> "thread history DB integrity": "ok"
> },
> "remediation": null,
> "durationMs": 181
> },
> "state.rollout\_db\_parity": {
> "id": "state.rollout\_db\_parity",
> "category": "threads",
> "status": "ok",
> "summary": "rollout files and state DB thread inventory agree",
> "details": {
> "default model provider": "openai",
> "rollout DB active files": "3546",
> "rollout DB active rows": "3546",
> "rollout DB archive mismatches": "0",
> "rollout DB archived files": "0",
> "rollout DB archived rows": "0",
> "rollout DB duplicate DB paths": "0",
> "rollout DB duplicate rollout thread ids": "0",
> "rollout DB malformed file names": "0",
> "rollout DB missing active rows": "0",
> "rollout DB missing archived rows": "0",
> "rollout DB model providers": "openai=3511, amazon-bedrock=35",
> "rollout DB rows": "3546",
> "rollout DB scan cap reached": "false",
> "rollout DB scan errors": "0",
> "rollout DB sources": "exec=3443, cli=93, subagent:thread\_spawn=9, vscode=1",
> "rollout DB stale rows": "0"
> },
> "remediation": null,
> "durationMs": 20384
> },
> "system.disk": {
> "id": "system.disk",
> "category": "disk",
> "status": "ok",
> "summary": "sufficient free disk space (679.4 GiB)",
> "details": {
> "CODEX\_HOME available": "679.4 GiB",
> "failure threshold": "1.0 GiB",
> "warning threshold": "5.0 GiB",
> "worktree available": "679.4 GiB"
> },
> "remediation": null,
> "durationMs": 0
> },
> "system.environment": {
> "id": "system.environment",
> "category": "system",
> "status": "ok",
> "summary": "OS language ja-JP",
> "details": {
> "EDITOR": "not set",
> "LANG": "ja\_JP.UTF-8",
> "VISUAL": "not set",
> "os": "Windows 10.0.26200 (Windows 11 Core) \[64-bit\]",
> "os language": "ja-JP",
> "os type": "Windows",
> "os version": "10.0.26200"
> },
> "remediation": null,
> "durationMs": 0
> },
> "terminal.env": {
> "id": "terminal.env",
> "category": "terminal",
> "status": "warning",
> "summary": "height 22 rows - content may scroll off (recommended \>=24)",
> "details": {
> "COLORTERM": "truecolor",
> "TERM\_PROGRAM": "vscode",
> "VSCODE\_INJECTION": "present",
> "color output": "enabled",
> "console input code page": "932",
> "console output code page": "932",
> "effective locale": "ja\_JP.UTF-8",
> "stderr console mode": "0x00000007 (VT processing: true)",
> "stderr is terminal": "true",
> "stdin is terminal": "true",
> "stdout console mode": "0x00000007 (VT processing: true)",
> "stdout is terminal": "true",
> "terminal": "VS Code",
> "terminal size": "144x22",
> "terminal version": "1.109.0"
> },
> "issues": \[
> {
> "severity": "warning",
> "cause": "height 22 rows - content may scroll off (recommended \>=24)",
> "measured": "144 x 22",
> "expected": "\>= 24 rows",
> "remedy": "resize the window to at least 24 rows",
> "fields": \[
> "terminal size"
> \]
> }
> \],
> "remediation": null,
> "durationMs": 0
> },
> "terminal.title": {
> "id": "terminal.title",
> "category": "title",
> "status": "ok",
> "summary": "terminal title default",
> "details": {
> "terminal title activity": "true",
> "terminal title items": "activity, project-name",
> "terminal title project source": "git repo root",
> "terminal title project value": "auto-codex",
> "terminal title source": "default"
> },
> "remediation": null,
> "durationMs": 0
> },
> "updates.status": {
> "id": "updates.status",
> "category": "updates",
> "status": "ok",
> "summary": "update configuration is locally consistent",
> "details": {
> "cached latest version": "0.153.4",
> "check for update on startup": "true",
> "last checked at": "2026-09-05T02:14:21.973748600Z",
> "latest version": "0.153.4",
> "latest version status": "current version is not older",
> "npm update target": "C:\\\\Users\\\\okamo\\\\AppData\\\\Roaming\\\\npm\\\\node\_modules\\\\@openai\\\\codex",
> "update action": "npm install -g @openai/codex",
> "version cache": "C:\\\\Users\\\\okamo\\\\.codex\\\\version.json"
> },
> "remediation": null,
> "durationMs": 299
> }
> }
> }
> \`\`\`
> 
> \### What issue are you seeing?
> 
> 
> Since switching to GPT-6 Astra, Codex CLI frequently enters a WebSocket reconnection loop during \`codex exec\`.
> 
> The following messages are repeatedly displayed:
> 
> \`\`\`text
> ERROR: Reconnecting... 2/5
> ERROR: Reconnecting... 2/5
> ERROR: Reconnecting... 3/5
> ERROR: Reconnecting... 4/5
> ERROR: Reconnecting... 5/5
> warning: Falling back from WebSockets to HTTPS transport. stream disconnected before completion: idle timeout waiting for websocket
> \`\`\`
> 
> Once this happens, Codex can remain in the reconnect loop for approximately 10 minutes before finally falling back from WebSockets to HTTPS.
> 
> The serious problem is that \*\*purchased Codex credits continue to be consumed while Codex is stuck in the reconnect loop\*\*.
> 
> In my observations, approximately \*\*3 credits are consumed every 10 seconds during the reconnect loop\*\*, even though no useful output is being produced and no files are being modified.
> 
> This means a single ~10-minute reconnect incident can potentially consume a very large number of purchased credits without making any visible progress.
> 
> After the retries finally reach \`5/5\` and Codex prints:
> 
> \`\`\`text id="m8fqdd"
> warning: Falling back from WebSockets to HTTPS transport.
> \`\`\`
> 
> the behavior changes immediately: Codex starts producing output and modifying/generating files normally over the HTTPS transport.
> 
> The observed sequence is:
> 
> \`\`\`text id="r6dbu2"
> GPT-6 Astra starts a task
> ↓
> WebSocket becomes idle/disconnected
> ↓
> Reconnecting... 1/5 ... 5/5
> ↓
> Purchased credits continuously decrease
> ↓
> No useful progress for ~10 minutes
> ↓
> Codex falls back to HTTPS
> ↓
> The task immediately starts progressing normally
> \`\`\`
> 
> \#### How I am monitoring the credit consumption
> 
> This issue is easiest to observe \*\*after the included 5-hour or weekly usage limit has already been exhausted\*\*, so that subsequent usage is charged against purchased Codex credits.
> 
> I am specifically monitoring the \*\*purchased credit balance\*\*, not estimating consumption from the 5-hour/weekly percentage shown by Codex CLI.
> 
> There is an important difference between the two monitoring methods:
> 
> \##### Codex CLI \`/status\`
> 
> Running \`/status\` during the reconnect loop does not show the purchased credit balance decreasing.
> 
> From the CLI's perspective, nothing visibly changes while the reconnect loop is occurring.
> 
> Therefore, \`/status\` does \*\*not\*\* expose the credit consumption described in this report.
> 
> \##### ChatGPT Codex usage page
> 
> Instead, I monitor the purchased credit balance using the official Codex usage page:
> 
> https://chatgpt.com/codex/settings/usage
> 
> While \`ERROR: Reconnecting...\` is repeatedly displayed in the terminal, refreshing the usage page shows the purchased credit balance continuously decreasing.
> 
> I observe approximately:
> 
> \`\`\`text
> ~3 purchased credits consumed every ~10 seconds
> \`\`\`
> 
> during the reconnect loop.
> 
> This continues even though:
> 
> \* Codex is only displaying \`ERROR: Reconnecting...\`
> \* no useful model output is appearing
> \* no files are being modified/generated
> \* \`/status\` does not indicate any corresponding change
> \* the task does not appear to make useful progress
> 
> The correlation is especially noticeable because \*\*the rapid credit consumption stops once the WebSocket retry loop ends and Codex falls back to HTTPS\*\*, at which point normal task progress resumes.
> 
> Therefore, the reproduction/measurement procedure is:
> 
> 1. Exhaust the included 5-hour or weekly Codex usage allowance so that purchased credits are being used.
> 2. Ensure the account has a purchased Codex credit balance.
> 3. Open the official usage page and note the current purchased credit balance:
> https://chatgpt.com/codex/settings/usage
> 4. Start a task using GPT-6 Astra (medium) with \`codex exec\`.
> 5. Wait until Codex enters the WebSocket reconnect loop.
> 6. While \`ERROR: Reconnecting... x/5\` is displayed, periodically refresh the official usage page.
> 7. Observe the purchased credit balance decreasing approximately every 10 seconds.
> 8. At the same time, observe that Codex is producing no useful output or file modifications.
> 9. Optionally run \`/status\`; it does not show this purchased-credit depletion.
> 10. Wait until Codex reaches \`5/5\` and falls back to HTTPS.
> 11. Observe that the task then immediately begins making normal progress.
> 
> This makes the unexpected consumption directly observable from OpenAI's own usage page rather than being inferred from local token counts or the CLI \`/status\` display.
> 
> 
> \### What steps can reproduce the bug?
> 
> 1. Use an account where the included 5-hour or weekly Codex usage limit has been exhausted and purchased credits are now being consumed.
> 2. Note the purchased credit balance at https://chatgpt.com/codex/settings/usage.
> 3. Use GPT-6 Astra with medium reasoning.
> 4. Run a relatively long task using \`codex exec\`.
> 5. Wait until the WebSocket stream hits an idle timeout.
> 6. Observe repeated \`ERROR: Reconnecting... x/5\` messages.
> 7. Refresh the Codex usage page periodically while the reconnect loop is running.
> 8. Observe that purchased credits continue decreasing, approximately 3 credits every 10 seconds in my case, even though the task is making no visible progress.
> 9. After \`5/5\`, Codex falls back to HTTPS.
> 10. Observe that normal task execution then immediately resumes.
> 
> \### What is the expected behavior?
> 
> A transport-level WebSocket failure should not repeatedly consume significant purchased Codex credits.
> 
> In particular, retrying a failed transport should not cause repeated billable model inference for the same unfinished turn.
> 
> Ideally:
> 
> 1. A failed WebSocket connection should fall back to HTTPS/SSE quickly.
> 2. Transport retries should not create new billable inference requests unnecessarily.
> 3. If repeating the model request is unavoidable, Codex should not automatically perform multiple expensive retries without warning the user.
> 4. Failed/retried requests should not repeatedly charge purchased credits when no useful response is delivered.
> 5. There should be an option to disable WebSockets entirely and use HTTPS/SSE from the beginning.
> 
> For example:
> 
> \`\`\`text
> codex exec --transport https
> \`\`\`
> 
> or an equivalent configuration option.
> 
> \### Additional information
> 
> The most important part of this issue is \*\*not simply that WebSockets are unstable\*\*.
> 
> The critical issue is:
> 
> \*\*Purchased Codex credits are continuously depleted while Codex is stuck in an automatic WebSocket reconnect loop, despite no useful work being produced.\*\*
> 
> This is directly observable by refreshing the official Codex usage page while the reconnect loop is occurring.
> 
> With GPT-6 Astra, the rate of consumption can be substantial. At approximately 3 credits every 10 seconds, a reconnect loop lasting around 10 minutes could consume roughly 180 purchased credits while apparently making no useful progress.
> 
> Once Codex gives up on WebSockets and switches to HTTPS, the same task begins working normally.
> 
> Related discussion about forcing HTTPS transport:
> 
> https://github.com/openai/codex/issues/27381#issuecomment-5514150210

If that issue describes the same problem you are experiencing, consider adding a 👍 reaction to the GitHub issue itself, rather than only reacting to this forum topic. GitHub reactions can help the maintainers gauge how many users are affected.

If it does not match your problem closely enough, search the [existing Codex issues](https://github.com/openai/codex/issues) for a better match. If you cannot find one, consider opening a new issue and include as much relevant information as possible, such as your Codex version, operating system, environment, reproduction steps, expected behavior, actual behavior, and any relevant logs or error messages.

---

<div class="post-metadata">

**Author:** ![frogjiangs](https://avatars.discourse-cdn.com/v4/letter/f/50afbb/32.png) [@frogjiangs](https://community.openai.com/u/frogjiangs)\
**Post date:** [October 25, 2026, 2:16pm UTC](https://community.openai.com/t/openai-network-infrastructure-and-or-bad-network-connection/1396089/3 "2026-10-25T14:16:56Z")

</div>

Adding a newer controlled observation, without assuming all reconnect errors have the same cause. On Windows desktop package26.1021.3705.0 / bundled core0.164.0-alpha.11.1, an independent localhost-only mock with dummy credentials reproduces persistent per-chat HTTP fallback:

| Same app-server process | WS requests, including warmup | HTTP requests |
| --- | --- | --- |
| ChatA, mock WS fails then HTTP succeeds | 3 | 1 |
| ChatA next turn, mock WS now healthy | 0 | 1 |
| Fresh ChatB, same process and healthy mock | 2 | 0 |

All three turns completed/error=null. No real model inference, persistent settings change, or original-chat edit was involved. This verifies fallback lifetime, not the initial cause of a real TLS/server error.

Separately, actual US wired residential use showed TLS decrypt errors, invalid TLS records, WS closure before response.completed, and HTTP200 followed by response-body decode failures across multiple chats. One affected chat had only about550KB raw request bytes; fresh same-history controls worked. Reloading that inactive original chat preserved history hashes/model/provider and was followed by an explicitly associated new model WS connection and a completed real desktop minimal reply. Long-run stability remains unproven.

Detailed incremental evidence is already posted in the closest existing OpenAI Codex GitHub issue, number 53139, comment 6398542559, rather than opening duplicates. No external link is included here because new community accounts cannot post links.

Private OpenAI Support has the correlatable request IDs and has confirmed escalation to a support specialist. A later-turn bounded WS re-probe or safe history-preserving reconnect path would help. Initial network/TLS failures still require separate diagnosis; high access bandwidth does not settle client state, TLS, routing or service behavior. There is no claim of a universal fix or measured global failure-rate increase.

---

<div class="post-metadata">

**Author:** ![OpenAI\_Support](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/openai_support/32/561845_2.png) [@OpenAI\_Support](https://community.openai.com/u/OpenAI_Support)\
**Post date:** [November 2, 2026, 8:22pm UTC](https://community.openai.com/t/openai-network-infrastructure-and-or-bad-network-connection/1396089/4 "2026-11-02T20:22:25Z")

</div>

Hey @none1, is this still happening? If so, could you share which Codex client and version you’re using, along with any exact error message?

@frogjiangs, thanks for the detailed testing and for sharing what you found! Since you mentioned your private support case has already been escalated, please continue working with the Support team there and add any newer findings to that same case. Please keep sensitive logs and account details in that private request too.

Craig

---

<div class="post-metadata">

**Author:** ![chirag.parcha](https://avatars.discourse-cdn.com/v4/letter/c/a88e57/32.png) [@chirag.parcha](https://community.openai.com/u/chirag.parcha)\
**Post date:** [November 10, 2026, 4:54pm UTC](https://community.openai.com/t/openai-network-infrastructure-and-or-bad-network-connection/1396089/5 "2026-11-10T16:54:25Z")

</div>

Closed on November 10, 2026, 4:54pm UTC


