The cloud-native orchestrator for the whole development lifecycle, with integrated lineage and observability.

dagster

Hi all,
Is there an easy way to trigger a dagster job from an AWS Lambda function that I already have or will I need to make a sensor instead?

if you have a reachable dagit server probably <https://docs.dagster.io/concepts/dagit/graphql-client#using-the-graphql-client>

<@UH3RM70A2> I cant seem to import the dagster_graphql library in my lambda, not sure why

<@UH3RM70A2> I have gotten past that issue ^ now my lambda keeps timing out on any of the graphql functions I may try to run (get status, start run, etc.).
I am guessing the issue is how I have my `client` set up.
`client = DagsterGraphQLClient("&lt;endpoint name&gt;", port_number=3000)`
Do i need the `/graphql` at the end of my endpoint name?

have you gotten a local version working or do you not have network access to the dagit instance from your laptop?

I have network access to the dagit instance at the endpoint name

&gt; Do i need the /graphql at the end of my endpoint name?
I dont believe so based on the documentation and examples i see in the code base

No longer timing out but definitely getting that it cannot connect
```[ERROR] DagsterGraphQLClientError: Error when connecting to url http://&lt;endpoint name&gt;/graphql. Did you specify hostname: &lt;endpoint name&gt; correctly?```

and the url it returns is one you can visit in your browser successfully?

I have given the lambda admin access to eliminate it being a permission issue

you may need to set the `transport` argument on `DagsterGraphQLClient` to pass any headers or however identity is established. `DagsterGraphQLClient` is just a thin wrapper atop <https://github.com/graphql-python/gql>

is there more context in the error output?

I was able to fix it based on the tip you gave me - It was an issue with access within a VPC. Thanks for the help!