Scale up trident controller pods | NetApp | Page 1

noble canopy Apr 10, 2025, 5:06 PM

#

Hi Team,
I am using trident for the NAS volume provisioning for the Kubernetes workload, when we are doing cluster maintenance we always into to issue such pods failed to mount the volume. until the trident controllers comes online the application pods cannot start. is there any way can we increase the replica count for the trident controller? i tried editing the deployment but it is getting overwrite by the operator

bronze temple Apr 11, 2025, 12:58 PM

#

No, the trident controller is not designed to be duplicated. If you force it the two trident-controllers will eventually cross talk and step on each other and cause major issues. The best solution is to set up k8s in such a way that the trident-controller pod can move to another node when the one it is currently on is down for maintenance.

noble canopy Apr 15, 2025, 7:26 AM

#

@bronze temple thanks for your reply, is there any way we can reduce the pod startup time (image pull delay) ? any idea how other customers overcome with this issue ?

tribal granite Apr 15, 2025, 7:38 AM

#

I guess you could proactively pull the new images on all worker nodes (ie. docker pull docker.io/netapp/trident-operator:<tag> etc)?

#Scale up trident controller pods