HPE EVA Storage

Bad performance for EVA-CA full copy replication

Occasional Contributor

Bad performance for EVA-CA full copy replication

Hi there,

due to less space in the logging disk group I have to delete my old DR-Groups and built it new (one after the other) with logging in another DiskGroup with more disks in it.
Afterwards I had to start a Full-Copy. At this point I got the following error/warning messages:

Problem Found: CUST: CA warning: EVAxx: Excessive Vdisk response time at the DR Destination has been detected

Problem Found: CUST: CA informational: EVAxx: Copy resources on the inter site link have been reduced to the minimum allowed value.

I had two EVA8100 that replicate via CA in a 4 GB FC environment.

My Perfmon monitor shows me a max. throughput of 40-50 MB/s before the above messages occur, afterwards I got only 20-25 MB/s,
so I need for a 100 GB LUN-copy approximately 50 minutes.

Did anyone know if it is normal CA-performance characteristic? In my opinion that is bad performance and I don't know why.

Any hints or explanations are welcome.
Víctor Cespón
Honored Contributor

Re: Bad performance for EVA-CA full copy replication

Full copy is done in synchronous mode. Your destination array seems to be saturated and not being able to cope with it's own traffic and the CA traffic at the same time.

These kind of situation require a data capture with EVAperf and an analysys.