you'd think it would show a better error 🙂

If this is in the docs and I am blind please direct me. How do schema updates work? Eg if I need to add or remove a property in the future?

you call

WriteSchema

or use

zed schema write

to update

you can always add a

relation

or `permission`; removing one is allowed if it is unreferenced

got it, ty

Hi - I am trying to connect to my kubernetes authz instalment through java gRPC and getting this following error .

java.lang.RuntimeException: UNAVAILABLE: io exception
Channel Pipeline: [SslHandler#0, ProtocolNegotiators$ClientTlsHandler#0, WriteBufferingAndExceptionHandler#0, DefaultChannelPipeline$TailContext#0]

In my port forwarding window , I can see the following

E1224 19:37:25.756364 2333699 portforward.go:407] an error occurred forwarding 50051 -> 50051: error forwarding port 50051 to pod e9be608d50a07364633d2ae963ea53fe43b657fb8736f786d184429120250db4, uid : failed to execute portforward in network namespace "/var/run/netns/cni-ed183935-5abf-50ab-c25d-1a89990abab1": read tcp4 127.0.0.1:32768->127.0.0.1:50051: read: connection reset by peer
E1224 19:37:26.238240 2333699 portforward.go:233] lost connection to pod

Getting this following error when I try to evaluate permission , authz running in my kubernetes cluster , followed this steps https://authzed.com/docs/spicedb/operator to set it up. zed permission check blog/post:1 write blog/user:emilia --revision "GhUKEzE2NzE4OTA2NzIwMDM1Njg4NjY="

Error: rpc error: code = Unavailable desc = last connection error: connection error: desc = "transport: Error while dialing dial tcp 172.30.0.241:50053: connect: connection refused"

dispatching in your cluster is misconfigured

I keep on getting this error

Could not start the Development System. Please make sure you have WebAssembly enabled.

any Ideas how can I fix this on play.authzed.com

I have tried enabling WebAssembly it still throws error on chrome and edge both

👋🏻 where are you based? The

main.wasm

files are a bit heavy and I don't think we have CDN setup yet. Did you check with chrome inspector if the file is properly downloaded?

Is there any blog post or information on how to deploy SpiceDB in a production manner on premise without K8S? I've already deployed a postgresql cluster, and got the schema running, but I'm unsure of how many of which services etc.

the count is based on your traffic pattern and instance size, but should be at least 3 SpiceDB nodes for redundancy

datastore itself is based on however you run a production-ready version of Postgres/MySQL/CRDB/etc

The dispatcher is ran separate to the 3 SpiceDB nodes?

not sure I follow. in a non-K8s based environment, you'll have to configure the dispatch to point to the set of nodes via DNS

and hope updates are fast

its one of the major reasons we recommend k8s

I need to read about the dispatcher, I just started toying with this. Do you have something like https://github.com/temporalio/samples-server/blob/main/extensibility/authorizer/server/main.go#L44 Where I can run the SpiceDB server from go?

https://github.com/authzed/spicedb/blob/main/pkg/cmd/serve.go is the entrypoint and could be scripted if need be

https://authzed.com/blog/consistent-hash-load-balancing-grpc/ going to read this before asking more 🙂

Thanks For instant feed back

Very Refreshing !

anytime

and yeah, that's a great resource; I should have thought to link it

So it didn't mention much about how the non k8s way depends on DNS

But I did read " The upstream-addr should be the DNS address of the load balancer at which all SpiceDB nodes are accessible at the default dispatch port of :50053 "