timescaledb service end point about helm-charts HOT 9 CLOSED

Could you elaborate a bit? This may be an issue with Digital Ocean, these Helm Charts, or something very different.

To reproduce I did this:

helm3 upgrade --install bugtest ./charts/timescaledb-single/

And this is what I run into with DO:

kubectl describe service/bugtest

  Warning  SyncLoadBalancerFailed  31s                service-controller  Error syncing load balancer: failed to ensure load balancer: failed to create load-balancer: POST https://api.digitalocean.com/v2/load_balancers: 429 (request "20b5a9be-d807-42b2-a772-72e926d3d61f") You have reached your load balancer limit, maximum allowed 10.

So, clearly I'm running into a Digital Ocean limit here.

For troubleshooting, could you provide the output of:

kubectl get service/$RELEASE

from helm-charts.

thanks for replying, i am attaching screenshot of nodebalancer as well as sharing output of service, timescale db is running, i have checked via port forward of pod, but not able to connect to DB remotely

from helm-charts.

Looks good, you may be running into a firewall or something similar; that's hard to establish from here.

From inside the kubernetes cluster (from another timescale pod for example) what does the following do:

pg_isready -h 157.230.192.232 -p 5432
pg_isready -h timescale-16 -p 5432

Also, to address your actual question, however it would only be a debugging/troubleshooting step IMHO:

kubectl port-forward $(kubectl get pod -l cluster-name=timescale-16,role=master -o name) 15432:5432

In a separate console, you should be able to connect:

psql -h localhost -p 15432 [...]

from helm-charts.

I have checked the port forwarding method and it's working, for other method I will share update

from helm-charts.

pg_isready -h timescale-16 -p 5432

is fine, but not with IP

from helm-charts.

Ok, that means the Service is working, but you cannot connect to the public IP.
That seems like a firewall to me, I'd advise you to take that up with Digital Ocean, either their support documentation or their support channels.

from helm-charts.

ok thanks for your time, nodebalancer always say down, I logged in server, port is listening but only ipv6

from helm-charts.

i can access other services but connection timeout on the timescale service, and even can not connect pod direct address from other non timescale pod

from helm-charts.

build new cluster and from this github repo, and it worked now, thanks for your time

from helm-charts.