I believe the wafl_check has to be run from maintenance mode. wafl_iron can be run while the filer is online and the aggregate will be offline for a certain amount of time during the wafl_iron. Depending on the size of the aggregate and the vols in the aggregate it can take a while to run.
Yes, if you destroy the aggregate and create a new one the issue will be gone. We have done this several times to get an aggregate back since they were large enough that the wafl_iron or check would run for many hours.
Kelley R. Green IT Specialist Global Technology Services - Storage Access Line 801-415-0449 Cell 801-916-1273 e-mail: krgreen@us.ibm.com
"Nils Vogels" bacardicoke@gmail.com Sent by: owner-toasters@mathworks.com 12/04/2007 07:17 AM
To Bryano Bryan.Price@firstassistinsurance.co.uk cc toasters@mathworks.com Subject Re: Aggregate shows need check
Hi Bryano,
On Dec 4, 2007 1:40 PM, Bryano Bryan.Price@firstassistinsurance.co.uk wrote:
According to the manuals this means that i need to run a wafl consistency check ran on it (Wafl_check), can I do this while the filer is online and serving data or do i have to boot the filer in to maintenance mode?.
Yes, you can run a wafl_check on an aggregate while the filer is online, as long as the aggregate DOESN'T contain the root volume.
The aggregate will be taken offline while running the check, so all the data that you are serving from the aggregate will have availability issues :)
If i was to delete the aggregate and start again would I take it that would negate the need to run the wafl_check? reason I ask is that the volumes only contain test data and this is more feasable than bringing the whole filer down for several hours (its an R200)
In theory, yes, it should remove the status, since you will be creating a completely set of raidgroups. I haven't ever tried that personally, though.
HTH & HAND!