[
https://issues.apache.org/jira/browse/YUNIKORN-86?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17088693#comment-17088693
]
Adam Antal commented on YUNIKORN-86:
------------------------------------
Tested the helm charts for the latest patch on local docker-desktop:
{noformat}
$ helm install --name yunikorn ./yunikorn/
[...]
$ kubectl describe pod yunikorn-scheduler-76b97c9d8b-xkp2s
Name: yunikorn-scheduler-76b97c9d8b-xkp2s
Namespace: default
Priority: 0
Node: docker-desktop/192.168.65.3
Start Time: Tue, 21 Apr 2020 15:24:54 +0200
Labels: app=yunikorn
component=yunikorn-scheduler
pod-template-hash=76b97c9d8b
release=yunikorn
Annotations: <none>
Status: Running
IP: 10.1.0.179
Controlled By: ReplicaSet/yunikorn-scheduler-76b97c9d8b
Containers:
yunikorn-scheduler-k8s:
Container ID:
docker://fa657544037701c57e2831163ef9b450dad5bf6c4a57fceac057d9f4dd0c656a
Image: yunikorn/yunikorn-scheduler-k8s:latest
Image ID:
docker-pullable://yunikorn/yunikorn-scheduler-k8s@sha256:7c6fa1d2d05e9af947eadcded57ad61e69ed858bc417bda1835966d86f5a04e3
Port: 9080/TCP
Host Port: 0/TCP
State: Running
Started: Tue, 21 Apr 2020 15:24:57 +0200
Ready: True
Restart Count: 0
Limits:
cpu: 4
memory: 2Gi
Requests:
cpu: 200m
memory: 1Gi
Environment:
NAMESPACE: default (v1:metadata.namespace)
Mounts:
/etc/yunikorn/ from config-volume (rw)
/var/run/secrets/kubernetes.io/serviceaccount from
yunikorn-admin-token-cjcfn (ro)
yunikorn-scheduler-web:
Container ID:
docker://44d444d352e7093116ce55ab98822aed077ceadb707c59dcfa8909a1adcf709e
Image: yunikorn/yunikorn-web:latest
Image ID:
docker-pullable://yunikorn/yunikorn-web@sha256:4ac7ebbd00321a3f3d6fa90c9ea419052acd127930be9ef840c76dc32253fbd8
Port: 9889/TCP
Host Port: 0/TCP
State: Running
Started: Tue, 21 Apr 2020 15:24:59 +0200
Ready: True
Restart Count: 0
Limits:
cpu: 200m
memory: 64Mi
Requests:
cpu: 100m
memory: 32Mi
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from
yunikorn-admin-token-cjcfn (ro)
Conditions:
Type Status
Initialized True
Ready True
ContainersReady True
PodScheduled True
Volumes:
config-volume:
Type: ConfigMap (a volume populated by a ConfigMap)
Name: yunikorn-configs
Optional: false
yunikorn-admin-token-cjcfn:
Type: Secret (a volume populated by a Secret)
SecretName: yunikorn-admin-token-cjcfn
Optional: false
QoS Class: Burstable
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 16s default-scheduler Successfully assigned
default/yunikorn-scheduler-76b97c9d8b-xkp2s to docker-desktop
[...]
{noformat}
So the limits are set accordingly.
Updating the helm chart also works. After setting 2Gi for request and 3Gi as
limit in values.yaml:
{noformat}
$ helm upgrade yunikorn ./yunikorn/
Release "yunikorn" has been upgraded.
[...]
$ kubectl describe pod yunikorn-scheduler-56bddd5895-txcsx
[...]
Containers:
yunikorn-scheduler-k8s:
Container ID:
docker://ac93a8545bc9747e6973d0fd0d5b25f833686f2b5842f464a6da4d8d5382d32b
Image: yunikorn/yunikorn-scheduler-k8s:latest
Image ID:
docker-pullable://yunikorn/yunikorn-scheduler-k8s@sha256:7c6fa1d2d05e9af947eadcded57ad61e69ed858bc417bda1835966d86f5a04e3
Port: 9080/TCP
Host Port: 0/TCP
State: Running
Started: Tue, 21 Apr 2020 15:30:32 +0200
Last State: Terminated
Reason: Error
Exit Code: 2
Started: Tue, 21 Apr 2020 15:30:24 +0200
Finished: Tue, 21 Apr 2020 15:30:29 +0200
Ready: True
Restart Count: 1
Limits:
cpu: 2
memory: 3Gi
Requests:
cpu: 200m
memory: 2Gi
Environment:
NAMESPACE: default (v1:metadata.namespace)
Mounts:
/etc/yunikorn/ from config-volume (rw)
/var/run/secrets/kubernetes.io/serviceaccount from
yunikorn-admin-token-cjcfn (ro)
yunikorn-scheduler-web:
Container ID:
docker://f4f935607ed37b12a667bc4dbea93db6359eaa9f3dedcbc95edfef158f4558eb
Image: yunikorn/yunikorn-web:latest
Image ID:
docker-pullable://yunikorn/yunikorn-web@sha256:4ac7ebbd00321a3f3d6fa90c9ea419052acd127930be9ef840c76dc32253fbd8
Port: 9889/TCP
Host Port: 0/TCP
State: Running
Started: Tue, 21 Apr 2020 15:30:26 +0200
Ready: True
Restart Count: 0
Limits:
cpu: 200m
memory: 64Mi
Requests:
cpu: 100m
memory: 32Mi
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from
yunikorn-admin-token-cjcfn (ro)
[...]
{noformat}
So it looks good to me.
> Set proper resource request and limit for YuniKorn pods
> -------------------------------------------------------
>
> Key: YUNIKORN-86
> URL: https://issues.apache.org/jira/browse/YUNIKORN-86
> Project: Apache YuniKorn
> Issue Type: Task
> Components: core - scheduler
> Reporter: Adam Antal
> Assignee: Adam Antal
> Priority: Critical
> Labels: pull-request-available
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Right now we don't specify the resource request and limit for neither the
> scheduler container nor the web container. We should define it.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
