Improve health check to return more granular details #85
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The Dashboard AI team have requested that the plugin health check returns more details about whether the OpenAI functionality is configured, and whether that configuration is valid. This PR adds it, and similar functionality for the vector service. It's backwards incompatible with past health checks unfortunately, but the llmclient package does include a change which should be compatible with both the current and previous version of the plugin. We're definitely due a refactor of the way we handle OpenAI config and requests, but I think that will be part of #80.
|
TODO:
|
3 tasks
The app is recreated by the plugin management system if settings change, so the cached value will be discarded when the user updates the settings to fix the health check error. Still unsure if this is an ideal way of doing things, but it speeds up health checks significantly while saving on OpenAI costs.
sd2k
force-pushed
the
upgraded-health-check
branch
from
44a4448
to
3967408
Compare
| Error: "", | ||
| Models: map[string]openAIModelHealth{ | ||
| "gpt-3.5-turbo": {OK: true, Error: ""}, | ||
| "gpt-4": {OK: false, Error: `unexpected status code: 404: {"error": "model does not exist"}`}, |
There was a problem hiding this comment.
If we return the errors like foo: bar: {error: baz} it is hard to parse in the frontend. Could we just return an object? is this any type of convention?
I'd love to return to the frontend as much information as possible from OpenAI since this is part of the configuration phase.
This is the best we can do when the request fails when using the query:
Also disable the button if we're waiting for the health check.
sd2k
force-pushed
the
upgraded-health-check
branch
from
6922f10
to
2d0d880
Compare
ivanortegaalba
approved these changes
Oct 4, 2023
Merged
SandersAaronD
pushed a commit
that referenced
this pull request
Oct 27, 2023
Fix expected field names in type in health check details
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The Dashboard AI team have requested that the plugin health check returns more details about whether the OpenAI functionality is configured, and whether that configuration is valid. This PR adds it, and similar functionality for the vector service.
It's backwards incompatible with past health checks unfortunately, but the llmclient package does include a change which should be compatible with both the current and previous version of the plugin.
We're definitely due a refactor of the way we handle OpenAI config and requests, but I think that will be part of #80.