No output after requesting service

[HomemadeAdvanced]

I added the app to the NC and then I added the ChatGPT API key to the NC. But when I want to test it, I write "/" in a talk group and I can select ChatGPT or Dalle. I write something in the text line and when I confirm to send it, nothing happens.

What am I doing wrong?

My DNS server resolves the ChatGPT API domain, but the ChatGPT API web page says that the key has "never" been used so far, so apparently no connection is happening.

[julien-nc]

Hi, When you say "nothing happens", is there really nothing happening?

• Is the picker closing itself after you click the "generate" button?
• Do you see any error in the browser console log? (F12 -> console tab)
• Do you see any line mentioning integration_openai in the server logs (nextcloud.log)?
• Would you be ready to share this API key with me and invalidate/revoke it after I had the chance to try it?

[HomemadeAdvanced]

• Thanks for your fast response
• No the picker is not closing itself after clicking the button. There is a short loading screen on the button but nothing happens after that
• There are two errors in the console log: xhr.js:220 POST https://examplecloud.com/apps/integration_openai/completions 400

OpenAI completions request error ChatGptCustomPickerElement.vue:211 i {message: 'Request failed with status code 400', name: 'AxiosError', code: 'ERR_BAD_REQUEST', config: {…}, request: XMLHttpRequest, …}

• NC error log: (...) "app":"integration_openai","method":"POST","url":"/apps/integration_openai/completions","message":"OpenAI API error : Client error: POST https://api.openai.com/v1/completions resulted in a 429 Too Many Requests response:\n{\n \"error\": {\n \"message\": \"You exceeded your current quota, please check your plan and billing details.\",\n (truncated...)\n" (...)
• I don't do something like this in general

I think it is because of the free plan I use. Because of what the NC log says I think I cannot use it with the API. The free plan browser version works for me. Kind of strange.

[julien-nc]

Thanks for the feedback. The next release will display more meaningful errors from the API responses.

What you get means you either ran out of credit or your free trial has expired. You can check that there: https://platform.openai.com/account/usage

The free plan browser version works for me.

What do you mean? Using ChatGPT from the OpenAI web interface does not require a valid API key. This integration uses their REST API which require to have credit and non-expired plan.

[julien-nc]

v1.0.4 is out.

Feel free to reopen if needed.

[Nortrix0]

I'm having the same issue, only for text generation. I'm not over the limits, but I am getting these warnings for every character I type or button I press

NcButton.js:2 You need to fill either the text or the ariaLabel props in the button component. {text: undefined, ariaLabel: null}ariaLabel: nulltext: undefined[[Prototype]]: Object o {_uid: 385, _isVue: true, __v_skip: true, _scope: e, $options: {…}, …}

[Nortrix0]

It does look like trying it in a new .md file does work

[julien-nc]

The warning triggered by the NcButton component is not related (and should not be triggered as the button text is defined...).

Switching to another file with Text should not have any effect either.

IMO this is more about OpenAI being sometimes overloaded and returning empty responses.

It would be interesting to know the content of the network response of the POST request made to apps/integration_openai/completions (triggered when you click the "Generate"button) when it fails.

[Nortrix0]

Yes, it should be unrelated, but when doing DALL-E, which always worked, it did not spew those errors

But this is the POST of the apps/integration_openai/completions

xhr.js:220

POST https://nextcloud/apps/integration_openai/completions 400

  | (anonymous) | @ | xhr.js:220 -- | -- | -- | --   | e.exports | @ | xhr.js:16   | e.exports | @ | dispatchRequest.js:56   | d.request | @ | Axios.js:109   | (anonymous) | @ | Axios.js:144   | (anonymous) | @ | bind.js:9   | onInputEnter | @ | ChatGptCustomPickerElement.vue:246   | En | @ | vue.runtime.esm.js:3017   | n | @ | vue.runtime.esm.js:1815   | click | @ | NcButton.js:2   | En | @ | vue.runtime.esm.js:3017   | n | @ | vue.runtime.esm.js:1815   | a._wrapper | @ | vue.runtime.esm.js:7473

[julien-nc]

Thanks for the feedback.

I need the response content. You can get it in the network tab of the browser dev tools. Select the mentioned request and then check the content of the "response" tab on the right side.

[Nortrix0]

Request URL: https://nextcloud/apps/integration_openai/completions Request Method: POST Status Code: 400 Remote Address: 45.27.82.162:443 Referrer Policy: no-referrer alt-svc: h3=":443"; ma=2592000 cache-control: no-cache, no-store, must-revalidate content-length: 207 content-security-policy: default-src 'none';base-uri 'none';manifest-src 'self';frame-ancestors 'none' content-type: application/json; charset=utf-8 date: Mon, 10 Apr 2023 17:38:26 GMT expires: Thu, 19 Nov 1981 08:52:00 GMT feature-policy: autoplay 'none';camera 'none';fullscreen 'none';geolocation 'none';microphone 'none';payment 'none' pragma: no-cache referrer-policy: no-referrer server: Caddy server: Apache/2.4.56 (Debian) strict-transport-security: max-age=31536000; x-content-type-options: nosniff x-frame-options: SAMEORIGIN x-permitted-cross-domain-policies: none x-powered-by: PHP/8.1.17 x-request-id: IrMfWooKjhdYOGugnaTD x-robots-tag: noindex, nofollow x-xss-protection: 1; mode=block :authority: nextcloud :method: POST :path: /apps/integration_openai/completions :scheme: https accept: application/json, text/plain, / accept-encoding: gzip, deflate, br accept-language: en-US,en;q=0.9 content-length: 115 content-type: application/json cookie: nc_sameSiteCookielax=true; nc_sameSiteCookiestrict=true; __Host-nc_sameSiteCookielax=true; __Host-nc_sameSiteCookiestrict=true; nc_username=nortrix0; oc_sessionPassphrase=Dc71oQQkwcrFA%2B7NYX8cK5XhuizYNvRsZILqn3MRobfv4VZSNlrSGf67Tm0rhMIqRh196O0qSM9boPwR7GgunAKdmiVYC2RsCTdA56Ey4XeKYx8Z9dOiHduJszWxhliG; oc8o7h6e132r=e0f086c87df894676cd4983fd2c91585; nc_token=ka%2BViLrTrimiA9Un0fT5UjkBq2CdNjzl; nc_session_id=e0f086c87df894676cd4983fd2c91585 dnt: 1 origin: https://nextcloud requesttoken: 9tnD5ouAMRzp5G8D5hNDSFagfLeKqbYGk1xCRm/xs3I=:saym0t6rdUawgF5UjVIQITzTEOXQ4fVAxTsKJDXe9Tc= sec-ch-ua: "Google Chrome";v="111", "Not(A:Brand";v="8", "Chromium";v="111" sec-ch-ua-mobile: ?0 sec-ch-ua-platform: "Windows" sec-fetch-dest: empty sec-fetch-mode: cors sec-fetch-site: same-origin user-agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/111.0.0.0 Safari/537.36

Payload was: {prompt: "Give me the full story of the events of assassins creed", n: 1, maxTokens: 1000,…} maxTokens : 1000 model : "gpt-3.5-turbo" n : 1 prompt : "Give me the full story of the events of assassins creed"

[julien-nc]

This is the content of the "headers" tab. I need the response one.

[Nortrix0]

Oh, sorry, having to speed read at work.

{"error":"cURL error 28: Operation timed out after 30001 milliseconds with 0 bytes received (see https:\/\/curl.haxx.se\/libcurl\/c\/libcurl-errors.html) for https:\/\/api.openai.com\/v1\/chat\/completions"}

[julien-nc]

Oh thanks a lot for taking time to help from work :grin: .

There we go. The request made from the Nextcloud server to the OpenAI API reaches a timeout. OpenAI is ofter overloaded.