TzuHsiang (Andy) Su

com.google.apphosting.api.ApiProxy$CallNotFoundException: Can't make API call mail.Send in a thread that is neither the original request thread nor a thread created by ThreadManager

1

I am trying to make a REST api call, so when users make a POST request, it will send emails However, I encounter the problem when trying to send the...

Change the hardcoded value on country "Taiwan"

2

Hi, I'm not trying to be nitpicky here, but when I looked at the code at line 1166 on paper-chip/demo/index.html, it marks ``` { "label": "Taiwan, Province of China", "value":...

Trino gateway health state design

22

Problem: Currently the health state of trino cluster is stored in-memory in trino-gateway. This is a problem because there can be multiple instances of trino-gateway, and their trino cluster states...

## Description As different routing rules are implemented, `routingManager.updateBackEndStats` method could be overridden. An example is https://github.com/trinodb/trino-gateway/blob/c052bb237450d7fc2f2f7af1b06f7fc8f4cdee48/gateway-ha/src/main/java/io/trino/gateway/ha/router/QueryCountBasedRouter.java#L238-L241 Therefore, it will be more useful to log unhealthy clusters at `ActiveClusterMonitor` level...

If http endpoint returns null or empty response, clusterstat is missing static attributes such as `proxyTo`. These attributes belong to backend and not on cluster stats. This is due to...

Feature request: Introduce different states to prevent clusters from being turned inactive programmatically

10

**Problem:** Currently, if the backends are not responding to healthcheck, trino-gateway will inactivate the backends. Since healthcheck only checks against active backends, inactive backends will never be turned to active...

## Description If no adhoc clusters are found, we should throw a meaningful exception with some explanations. ## Additional context and related issues ## Release notes ( ) This is...

Add PENDING type to healthchecks

1

## Description Resolves https://github.com/trinodb/trino-gateway/issues/222 part 1 ## Additional context and related issues ## Release notes (X) This is not user-visible or is docs only, and no release notes are required....

INFO_API doesn't expose queuedQueryCount and runningQueryCount

2

By design, `queuedQueryCount` and `runningQueryCount` are updated from health checks. However, `/v1/info` doesn't expose `queuedQueryCount` and `runningQueryCount`, which are the two important values for `QueryCountBasedRouter` to work properly. When using...

It looks like the purpose of BackendStateManager class is to provide states of clusters to frontend. We can use RouterManager to achieve the same result.