Volumes were found that must be changed to online state before attempting NDU

[tmferg]

Attempting a NDU from 8.3.1 to 9.1 and in the validation phase the subject error is reported. 

volume show -state !online -state !restricted reports 1 volume with a name of vol0 on my second node.

I found this KB: Volumes were found that must be changed to online state before attempting NDU - NetApp Knowledge Base

But it just states to contact support. Does anyone know how to correct this error? Our support ran out and I don't know how long it's going to take for it to be reinstated after the purchase has been approved. 

[SpindleNinja]

Volume do have to be online for upgrades, (or deleted).    

The question though is why it's vol0.    How many vol0s do you have? 

[tmferg]

So if I just do a vol show  node 1 shows a vol0 with an aggregate/size/available/used% populated, node 2 also shows a vol0 but no agregate/size/available/used data populated. 

I have tried to bring the vol0 for node 2 online through volume modify -vserver "node 2" -volume vol0 -state online and it reports aggregate not found.

[SpindleNinja]

Is the cluster healthy and not in takeover or anything? 

[tmferg]

Yes healthy and not in takeover. No issues reported in the upgrade validation besides this one offline volume.

[SpindleNinja]

if you drop into node shell on 02,  does vol0 look OK/healthy? 

[tmferg]

I'm not sure if I did it right or not, but doing system node run -node "node 2" then vol status, vol0 does not appear. My other volumes all look good in there. 

[SpindleNinja]

is there any volumes that show as root? under options? 

example - 

WOPR-02> vol status
         Volume State           Status                Options
           vol0 online          raid_dp, flex         root, nvfail=on, space_slo=none
                                64-bit

[tmferg]

AUTOROOT  is showing as root. On node 1 I have vol0 as root, nvfail=0n

[SpindleNinja]

Do you know if someone ever attempted to move to a new root aggr on this system?  

[tmferg]

No I don't know. Nobody currently working for the company was here when this system was last touched. 

[SpindleNinja]

Got ya. 

AUTOROOT is usually a remnant of node recovery or a root aggr move.  

Let me see what else I can dig up that might help.   But a support case would be best here.  

[tmferg]

Gotcha. I appreciate your help on this!

[SpindleNinja]

Following up on this, unfortunately, i wasn’t able to find anything that could be posted.

Few other questions thought - 

Is this part of a migration or something? i.e.  having to upgrade this system to migrate to a new ONTAP system?

Whats the output from the following - 

set d; debug vreport show

[tmferg]

Not migrating, just upgrading the firmware on our cluster to keep it current.

Output from the set d; debug vreport show:

aggregate Differences:

Name Reason Attributes
-------- ------- ---------------------------------------------------
ROOT_node_02 Present both in VLDB and WAFL with differences
Node Name: node-02
Aggregate UUID: e6e37fff-1793-49a7-984d-13674db15be6
Aggregate State: online
Aggregate Raid Status: raid_dp
Differing Attribute: Volume Count (Use commands 'volume add-other-volume' and 'volume remove-other-volume' to fix 7-Mode volumes on this aggregate)
WAFL Value: 1
VLDB Value: 0

[tmferg · Accepted solution]

Just wanted to let you know this was resolved by deleting the stale vol0 entry. 

[SpindleNinja]

Good to hear, the vreport is a handy command.   Sorry for the delay in replying, I was out of the country last week.