Unable to install dremio helm chart on kubernetes AWS EKS

1

Hello,

I am trying to install dremio on kubernetes AWS eks cluster with below configuration.
Instancetype: r5.4xlarge which has cpu 16 and memory:128Gib
NodeAutoScalingGroupMinSize: 1
NodeAutoScalingGroupMaxSize: 20
NodeAutoScalingGroupDesiredCapacity: 3

Also,
In values.yaml below is conf:
coordinator:
memory: 100000
cpu: 2
count: 0
web:
port: 9047
client:
port: 31010
volumeSize: 10Gi
executor:
memory: 100000
cpu: 2
count: 3
volumeSize: 10Gi
zookeeper:
memory: 1024
cpu: 0.5
count: 3
volumeSize: 10Gi

But dremio is failing to get installed with insufficient memory error.

From resource perspective, this configuration seems to be sufficient.Could you please give some understanding on how this is achievable.

Thanks in Advance!

2

Hi @meg,

What is the specific error message you are getting? Can you attach the output from your install?

3

@meg, can you also do a kubectl describe nodes and share the output?

4

Hi,

Good news! I was able to deploy dremio helm chart.
But still the pods are in pending status.

  1. kubectl get nodes
    o/p All nodes are in ready state.
    2.kubectl describe pod dremio-master-0
    o/p : No failed events But Tolerance is as below :
    node.kubernetes.io/not-ready:NoExecute for 300s
    node.kubernetes.io/unreachable:NoExecute for 300s

  2. $ kubectl get svc
    NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
    dremio-client LoadBalancer ClusterIPiscoming pending 31010:32121/TCP,9047:31747/TCP 56m

4.$ kubectl get pod
NAME READY STATUS RESTARTS AGE
dremio-executor-0 0/1 Init:0/2 0 56m
dremio-master-0 0/1 Init:1/4 0 56m

5.$kubectl get services dremio-client
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
dremio-client LoadBalancer CLUSTER-IPiscoming pending 31010:32121/TCP,9047:31747/TCP 3m21s

ISSUE
1.External_IP is in pending state.
2.POD state for executor and master isnot in running state.
3.Tolerance shows unreachable or not ready even when nodes are ready.
Could you please help with this?

Thanks in advance.