Configuring Resource Limits

Variables
Variables have been created to be able to configure resource limits for Trilio pods.
metadataJobLimits - Specifies limits for all meta-data related and target mounting jobs like target-validator, meta-snapshot, pre-restore-validation, meta-restore etc
dataJobLimits : Specifies limits for Datamover jobs
deploymentLimits : Specifies limits for helm chart deployments. Not applicable for OCP
Configuring limits in Upstream Kubernetes
Directly specify the resource limits in the TVM CR spec:
1
apiVersion: triliovault.trilio.io/v1
2
kind: TrilioVaultManager
3
metadata:
4
  labels:
5
    triliovault: k8s
6
spec:
7
  trilioVaultAppVersion: latest
8
  helmVersion:
9
    version: v3
10
  applicationScope: Namespaced
11
  restoreNamespaces: ["kube-system", "default"]
12
  metadataJobLimits:
13
    cpu: 500m
14
    memory: 512Mi
15
  dataJobLimits:
16
    cpu: 1500m
17
    memory: 5120Mi
18
  deploymentLimits:
19
    cpu: 250m
20
    memory: 512Mi
Copied!
Configuring limits in OCP
During installation on OCP, a confimap k8s-triliovault-configcontaining the default limits is created. Edit the configMap to change default limits as per requirement.
Kubectl edit configmap k8s-triliovault-config -n <NAMESPACE>
ConfigMap example:
1
apiVersion: v1
2
kind: ConfigMap
3
metadata:
4
  name: k8s-triliovault-config
5
  labels:
6
    app.kubernetes.io/instance: k8s-triliovault-config
7
    app.kubernetes.io/managed-by: Helm
8
    app.kubernetes.io/name: k8s-triliovault
9
    app.kubernetes.io/part-of: k8s-triliovault
10
data:
11
  metadataJobResources: |-
12
    limits:
13
      cpu: 500m
14
      memory: 512Mi
15
    requests:
16
      cpu: 10m
17
      memory: 10Mi
18
  dataJobResources: |-
19
    limits:
20
      cpu: 1500m
21
      memory: 5120Mi
22
    requests:
23
      cpu: 100m
24
      memory: 800Mi
25
  tvk: |-
26
    name: tvk-instance
Copied!

Monitoring Velero with TrilioVault for Kubernetes

Next - Performance

S3 as Backup Target

Last modified 1mo ago

Copy link

Contents

Variables

Configuring limits in Upstream Kubernetes

Configuring limits in OCP