FS for Windows fails - NFS mounts on Linux fails at attach on EC2

Question

IHAC who is using FS and mounting it as an NFS drive on Linux EC2 (ECS), and using this setup for an AI/ML workload.
They have reported an issue where they have observed that the mount-point stops reflecting on the EC2 instances during random reboots or when scaled.

I found a PFR with a similar type of issue - [Link](https://aws-crm.lightning.force.com/lightning/r/Product_Feature_Request__c/a2v4z0000035Vl7AAE/view)

Do we know a workaround until this gets resolved by the product team?

Thanks

Accepted Answer

Per the [documentation](https://docs.aws.amazon.com/fsx/latest/WindowsGuide/high-availability-multiAZ.html#MulitAZ-Failover), Linux clients do not support automatic DNS-based failover - which occurs during file system maintenance, infrastructure component replacement, and customer-initiated throughput scaling.

As a workaround, you can connect directly to the secondary IP of the file system while it is in a failed-over state. (However, only one of the IPs will be active depending on whether the file system is on the primary or standby server. You will need to monitor both IPs to see which is returning traffic at any given time.)

FS for Windows fails - NFS mounts on Linux fails at attach on EC2

Contenido relevante