Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Flyte

My understanding of the cluster resources <https://docs.flyte.org/en/latest/deployment/cluster_config/general.html|docs> is that anything applied from `cra.yaml` will be applied in k8s via the `resourcequota` on the project-domain namespace. Is that not the case?

After updating `cluster-resources-attribute`, I'm seeing following mismatch:
```flytectl get cluster-resource-attribute -p flytesnacks -d development```
returns
```{"project":"flytesnacks","domain":"development","attributes":{"projectQuotaCpu":"10","projectQuotaMemory":"14G"}}```
 but
```kubectl -n flytesnacks-development get resourcequotas```
returns
```NAME            AGE    REQUEST   LIMIT
project-quota   142m             limits.cpu: 0/4, limits.memory: 0/3000Mi```

hey <@U04331ZRKPA> are you running the cluster resource controller sync command as a container? what are the template files passed to the cluster resource controller?

I'm not sure. I'm using the sandbox deployment, so I would think not. I do not know how to check flyteadmin config. I'm attempting to follow this <https://docs.flyte.org/en/latest/deployment/cluster_config/general.html|page>

I wrote the following `cra.yaml`:
```attributes:
    projectQuotaCpu: "10"
    projectQuotaMemory: "14G"
domain: development
project: flytesnacks```
And then applying with `flytectl update cluster-resource-attribute --attrFile cra.yaml`

ah one more question, are you running `flytectl sandbox start` or `flytectl demo start`?

great, i believe that should bring up the sync container, can i ask you to double check that the <https://github.com/flyteorg/flyte/blob/master/deployment/sandbox/flyte_helm_generated.yaml#L6828|deployment/pod> for it is happy?

The `syncresouces` pod is status ready. It's log shows an error, but I think it's from when the cluster started not from when I updated the cra.

```time="2022-10-06T17:42:59Z" level=info msg="Using config file: [/etc/flyte/config/cluster_resources.yaml /etc/flyte/config/db.yaml /etc/flyte/config/domain.yaml]"

2022/10/06 17:42:59 /go/src/github.com/flyteorg/flyteadmin/pkg/repositories/gormimpl/project_repo.go:80 ERROR: relation "projects" does not exist (SQLSTATE 42P01)
[0.455ms] [rows:0] SELECT * FROM "projects" WHERE state &lt;&gt; 1 ORDER BY created_at desc 
{"json":{},"level":"warning","msg":"Failed cluster resource creation loop with: cannot query with specified table attributes: relation \"projects\" does not exist","ts":"2022-10-06T17:42:59Z"}```

oh! this is an issue - are your flyteadmin and postgres pods running? it looks like your database is not migrated/misconfigured

no errors for flyteadmin.

postgres pod has these errors:
```2022-10-06 17:42:24.344 UTC [1] LOG:  database system is ready to accept connections
2022-10-06 17:42:52.020 UTC [74] FATAL:  database "datacatalog" does not exist
2022-10-06 17:42:59.099 UTC [78] ERROR:  relation "projects" does not exist at character 15
2022-10-06 17:42:59.099 UTC [78] STATEMENT:  SELECT * FROM "projects" WHERE state &lt;&gt; $1 ORDER BY created_at desc ```

do you mind restarting the admin pod? the init containers should run the migrations to create the databases and tables

oh and restart the datacatalog one as well

Will do. I have meetings for an hour, but will let you know what happens after.

so after restarting both (with `rollout restart`),  the `flytesnacks-development project-quota` is synced correctly.

If i update the cra again, then those changes don't make it to the `project-quota`. Restarting flytedmin again then causes them to sync.