Joey
02/25/2022, 5:30 PMJoey
02/25/2022, 5:31 PMJoey
02/25/2022, 5:32 PM& feature_flag_new_admin_dashboard
, the graph walked to the flag, but did not find the user. The &
is not a boolean `&&`; its an intersection of the two sub pathswilliamdclt
02/25/2022, 5:36 PMwilliamdclt
02/25/2022, 5:36 PMJoey
02/25/2022, 5:37 PMJoey
02/25/2022, 5:37 PMwilliamdclt
02/25/2022, 5:43 PMJoey
02/25/2022, 5:44 PMwilliamdclt
02/25/2022, 5:48 PMsubject
and object
as input to my permission and would describe any path (or a boolean combination thereof) using these, which means that the paths don't have to end on the subject, they just have to exist (and it's up to me to use the subject
in the path description)williamdclt
02/25/2022, 5:50 PMJoey
02/25/2022, 5:51 PMJoey
02/25/2022, 5:51 PMJoey
02/25/2022, 5:51 PMwilliamdclt
02/25/2022, 5:51 PMwilliamdclt
02/25/2022, 5:51 PMJoey
02/25/2022, 5:52 PMwilliamdclt
02/26/2022, 12:28 PMwilliamdclt
02/26/2022, 12:43 PMError: rpc error: code = ResourceExhausted desc = grpc: received message larger than max (11078400 vs. 4194304)
(this is for part of my dataset, ~140000 relations).
Is there a way to import a large amount of data?
EDIT: I've just written a small bash script to chunk through the importJoey
02/26/2022, 5:03 PMJoey
02/26/2022, 5:06 PMdefinition platform {
relation user: user:*
}
then you just write the relationship platform:birdie#user@user:*
, and it'll be as if any user is therewilliamdclt
02/27/2022, 10:00 PMrelation flag_billing: user:*
👍
I'm having some questions about deploying SpiceDB to a production environment!
Primo, I'm having some trouble understanding how to configure a SpiceDB deployment, ie scale SpiceDB horizontally. I understand the concept of dispatching, but I'm struggling to understand the dispatch cluster configuration options. Is there a guide, or an example for a SpiceDB deployment? The docs talk about "when configured with [...] a source of peer discovery", but I'm not finding more practical information on how to do that
Secundo, does that mean that every SpiceDB instance in the deployment is able to serve traffic and to serve dispatch requests from other instances? So that SpiceDB is highly-available, traffic being load-balanced between all the instances
Tertio, can the datastore scale horizontally too? Eg can I have a Postgres read-replica and have SpiceDB make use of it for check and expand requests?Joey
02/28/2022, 5:08 AMJoey
02/28/2022, 5:11 AMwilliamdclt
02/28/2022, 10:50 AMJohannes Boyne
02/28/2022, 12:49 PMluka.h
02/28/2022, 2:12 PMecordell
02/28/2022, 2:14 PMluka.h
02/28/2022, 2:17 PMBryan
02/28/2022, 11:24 PMzed import
but even with the smallest file of 10k relationships, I got Error: bufio.Scanner: token too long
.