2012-11-14 09:48 AM
We have a faulty disk. The disk is faulty (not failed), so NetApp sent a replacement disk. In the meantime, I was told by NetApp to perform a "disk fail" on the faulty disk.
I ran the "disk fail <diskid>" without the -i switch. This results in ONTAP performing a "prefail copy" to a spare disk. We have the replacement disk on-site and would like to replace it immediately but the "prefail copy" is taking a lot of time because it is a SATA disk and the disk is faulty. The "prefail copy" has been running for 3 hours and is only 20% complete.
Is there any way to cancel the "disk fail" process once initiated? I'd like to cancel the copy to the spare disk and re-run the disk fail command using the -i switch.
Here is what I am talking about. It is essentially a disk swap before marking the disk as failed.
|data||1d.06.22 1d||6 22 SA:A 0 BSAS 7200 847555/1735794176 847884/1736466816 (prefail, copy in progress)|
|-> copy 1d.06.23 1d||6 23 SA:A 0 BSAS 7200 847555/1735794176 847884/1736466816 (copy 22% completed)|
2012-11-14 09:45 PM
You should be able to just pull failed disk out. Of course, I would contact NetApp support before to ensure there are no known bugs that result in controller panic or something.
Otherwise disk copy usually takes less time and impose less load on a system than full rebuild. And you will need to make rebuild anyway.