you can use a lookup per each then

You saved me, I think this will work for me well , till https://github.com/authzed/spicedb/issues/613

@Singha1 just be aware that if you need to check for a specific document, we recommend using a Check then

and, if you are looking directly in a

relation

, you can use a

read

which is faster

yes sure, that make total sense !!

@here We just released SpiceDB 1.8.0! It's mostly a performance-focused release, with lower latencies in a bunch of different scenarios. Full release notes here: https://github.com/authzed/spicedb/releases/tag/v1.8.0

Hi Team, Where is the permission data is stored in spiceDB

is it in memory or its persisted with some other datastore ?

It depends on you, what configuration you have selected, https://docs.authzed.com/spicedb/selecting-a-datastore

Do spiceDb runs in a cluster mode or its just one simple node ?

It runs on cluster mode, https://github.com/authzed/spicedb/blob/main/k8s/example.yaml , Increase replica as much as you want and connect it to persistent store like Postgresql.

do all the relation live in-memory?

or some of the relation which are not used frequently lives at persistent layer and are retrieve when needed?

This say's single node and not multiple nodes.

that's example if you want to increase the replica and use it.

if I am not wrong replica's are for high availability what about distributed load

what will happen if permission can-not fit in one single node ?

distributed load is handled via dispatcher. Look at its Architecture, https://authzed.com/blog/spicedb-architecture/#:~:text=SpiceDB%20has%20a%20datastore%20interface,which%20it%20can%20be%20retrieved.

still not clear. Can we have more than one spiceDb node running in a cluster

I am not talking about replicas

I am able it to run it with spec:dnsPolicy:"ClusterFirst"

The dispatching can be configured with the server 2 listen to request and then map specific request to a specific instance [ https://docs.authzed.com/spicedb/feature-overview ]

https://discord.com/channels/844600078504951838/844600078948630559/902618226528239637 also as another reference point

hey guys

we are facing the following errors on our production environment, things were working fine on all other environment apart from this one ↓, we've made sure that all the schema is loaded before we went on deployment, is there anything that we've missed?

canceled","pid":80278,"sql":"rollback","time":"2022-06-02T12:13:38Z","message":"Exec"}
{"level":"warn","grpc.component":"server","grpc.method":"CheckPermission","grpc.method_type":"unary","grpc.service":"authzed.api.v1.PermissionsService","peer.address":"192.168.58.163:48000","protocol":"grpc","requestID":"f58e35b24c99afa216885a6176a85f0d","grpc.start_time":"2022-06-02T12:13:38Z","grpc.code":"FailedPrecondition","grpc.error":"rpc error: code = FailedPrecondition desc = failed precondition: object definition `user` not found","grpc.time_ms":"12.243","time":"2022-06-02T12:13:38Z","message":"finished call"}
{"level":"warn","grpc.component":"server","grpc.method":"CheckPermission","grpc.method_type":"unary","grpc.service":"authzed.api.v1.PermissionsService","peer.address":"192.168.58.163:48000","protocol":"grpc","requestID":"92fb37116144bb7e86a081d2cbfeeb50","grpc.start_time":"2022-06-02T12:13:38Z","grpc.code":"FailedPrecondition","grpc.error":"rpc error: code = FailedPrecondition desc = failed precondition: object definition `user` not found","grpc.time_ms":"12.272","time":"2022-06-02T12:13:38Z","message":"finished call"}

This is our schema table

I think you are confusing the spicedb schema and the database schema

The database schema should be managed using the

spicedb migrate

command, and the spicedb schema should be managed with the

WriteSchema

api

Are you using memdb in your dev environments?