Subscribe
Accepted Solution

NDMPCOPY Process and performance

Hi

I am trying to migrate a volume with NDMPCOPY to other aggregate on same filer. I have initiated transfer as well. I am getting below messages on storage. I am confused to about these messages.

Can some one clarify the messages.

ndmpcopy: netapp1: Log: DUMP: XXXXX(date and time) :we have written 164816 KB.

ndmpcopy: netapp1: Log: Restore: XXXXX(date and time): We have read 162902 KB from the backup.

As per this meessage, this is written 160 MB to the destination..(if i am not wrong). But in the destiantion volume it still 20 MB.

Please clarify me on this message.

Thanks,

Re: NDMPCOPY Process and performance

could be a reporting issue...definitely wrote 160MB in the message... also local ndmpcopy uses the loopback so performance will be slower than if copying to another controller...but that doesn't explain the reporting of 160MB written and only 20MB used.

Re: NDMPCOPY Process and performance

Thanks for your response.

Could you please explain the passes of the NDMPCOPY.

PASS I

PASS II

PASS III

PASS IV

PASS V

We are presently in PASS III

Ndmpcopy: Netapp1: Log: DUMP: dumping (Pass III) [directories]

Is the speed increased in PASS IV (regular files)

Re: NDMPCOPY Process and performance

The passes are the same for the dump command.  Passes 1 and 2 create inode maps... pass 3 dumps directory structure, pass 4 dumps files and pass 5 is ACLs.  Speed increases typically during the dump phase and picks up speed as it goes...

Re: NDMPCOPY Process and performance

Thanks Scott