Temporal with Yugabyte db

Community Support
1

Hi,

Has anyone tried Temporal with Yugabyte db ?
Yugabyte is PostgreSQL compatible database and since Temporal supports PostgreSQL, I believe it should not be an issue.
It would be helpful if anyone has any insights regarding that. Any suggestions on how can I start with the same.

2

@tihomir @maxim Request you to please guide on the topic

3

Hi, check out this blog that I wrote on the subject:

2 Likes
4

Looks very useful. Thanks a lot for the information !!

5

@eric1pratt I am trying to follow the steps as mentioned in above blog post. Facing issue while doing helm install :

Error: INSTALLATION FAILED: unable to build kubernetes objects from release manifest: [unable to recognize “”: no matches for kind “ClusterRole” in version “rbac.authorization.k8s.io/v1beta1”, unable to recognize “”: no matches for kind “ClusterRoleBinding” in version “rbac.authorization.k8s.io/v1beta1”]

Please guide if I am missing anything.

6

@eric1pratt 1 thing to be rectified in above blog post. In section : Create and update the schema version of the temporal_visibility database, command to update schema for visibility database needs to be modified as below :
./temporal-sql-tool
–tls
–tls-disable-host-verification
–tls-cert-file $TLS_CERT_FILE
–ep $SQL_HOST
-p $SQL_PORT
-u $SQL_USER
-pw $SQL_PASSWORD
–pl $SQL_PLUGIN
–db $SQL_TEMPORAL_VISIBILITY_DB
update-schema -schema-dir ./schema/postgresql/v96/visibility/versioned

7

Thanks for the feedback. I’ll get it updated. Did you get past the previous error as well?

8

@eric1pratt Also, the copy commands need to be rectified as below :
kubectl cp basic-test.json temporal/${ADMIN_TOOLS_CONTAINER}:/tmp/
kubectl cp basic-const12k.json temporal/${ADMIN_TOOLS_CONTAINER}:/tmp/
kubectl cp basic-payload.json temporal/${ADMIN_TOOLS_CONTAINER}:/tmp/
kubectl cp basic-spike.json temporal/${ADMIN_TOOLS_CONTAINER}:/tmp/

Yet, I could not get past the previous error, it is occurring when prometheus is enabled.

9

@poojabhutada I believe the Prometheus issue is actually with Prometheus. I think I ran into the same thing. Let me dig around a bit. Apologies for the delay, I was out on vacation over the holidays.

10

@poojabhutada Looks like we need an upgrade to the api. ClusterRole is in rback.authorization.k8s.io/v1. Appears that the error is stemming from using beta1 version:

Error: INSTALLATION FAILED: unable to build kubernetes objects from release manifest: [unable to recognize “”: no matches for kind “ClusterRole” in version “rbac.authorization.k8s.io/v1beta1”

@tihomir any thoughts on the above? It seems that the helm chart is still using 11.04 which is using rbac with v1beta1:

The current prometheus version is 15. Can we just bump the version in the helm chart to work around it?

11

@eric1pratt can you please give it a try and report back, same issue was reported regarding grafana as well, see Temporal deployment with Helm charts, and associated fix Fix deprecated role APIs by zanhsieh · Pull Request #89 · grafana/helm-charts · GitHub