ONTAP Discussions
ONTAP Discussions
Hello everybody,
I have the following problem every time I restart NetApp:
Because of the false "is Home" the vm stored inside are not communicating and I don't have access, so I always use the migrate command to fix it or I enter the web interface to change it to True.
And there are times when the current port is changed in the cluster or in the NFS.
So I would like to know if there is a way to fix those paranormal changes (since it happens very rarely). Is there a boot configuration file or something like that so that it always loads the right configuration?
Thank you very much and best regards.
Hi,
Probably, the auto-revert property is set to 'false', I am guessing, bydefault it is false for 'data LIFs'.
First verify:
::> network interface show -vserver vserver_name -fields auto-revert
If it's set to false, make it 'true'.
::> network interface modify -vserver vserver_name -lif lif_name -auto-revert true
With this setting, next time when the Data LIFs failed-over, they will automatically revert to home-port once the Home-port is operational/status-admin up.
Thanks!
I check it:
FNC FNC-01_mgmt1 true
FNC FNC-02_mgmt1 true
FNC cluster_mgmt false
NFS01 NFS01_lif false
NFS01 NFS01_nfs_lif1 false
NFS02 NFS02_lif false
NFS02 NFS02_nfs_lif1 false
Would you recommend me to put everything to True?
(With that every reboot all things will go to their place, isn´t it?)
Thanks for your help!
Yes, please.
With that every reboot all things will go to their place, isn´t it? yes, that's correct.
You're welcome.
Perfect!
Tomorrow I´ll let you know the results 🙂
Thanks again!!!
I have this issue with the -auto-revert true in NFS01 and NFS02 lifs:
There are no entries matching your query
I don´t know why :s because the cluster_mgmt lif was changed correctly
What is the current output of : network interface show command ?
This current output has the migrate applied 2 weeks ago (that´s why the configuration are in his place with the trues and correctly current ports).
All the ports looks like are at 'Home', so that's good.
Also, I noticed, the Ports in your first screenshot was showing as 'a0s', probably that's why it said 'There are no entries...'.
e0*, is what I would expect.
The first screen was from the day that I rebooted the NetApp (then I could solved using the migrate and the web interface client).
But as you told me, for prevent that issues every reboot I tried to apply the -lif auto-revert but it doesn´t works in NFS01 and NFS02.
What is the output of this command;
network port show
network port show -port a0s -instance
Kindly give the output of :
::> network port show
::> network interface show -vserver NFS01 -fields home-node,home-port,curr-node,curr-port
::> network interface show -vserver NFS02 -fields home-node,home-port,curr-node,curr-port
Sorry (the e0s is e0a, because the first screen was taken from other NetApp with the same configuration)
Take the outputs:
Looks like you have under-lines issues (vifmgr/Node) and therefore when the LIF fails-over it is not reachable. Ideally, even if the LIFs fail-over, they should be reachable. However, if you want LIFs to go back to its home-port then auto-revert is the property that will turn on.
I see the issue in the output you shared:
Warning: Unable to list entries for vifmgr on node "FNC-01": RPC: Port Mapper failure. Therefore, it's only listing ports for Node FNC-02. Perhaps, that's why all your LIFs are attached to this Node.
Try to check:
::> set adv
Warning: These advanced commands are potentially dangerous; use them only when directed to do so by NetApp personnel.
Do you want to continue? {y|n}: y
::*> cluster ring show -unitname vifmgr
I would request you to log a ticket with NetApp Phone support.
Thanks!
Ok,
I´ll investigate and check
Thanks for your help!!
Thanks and good luck. Feel free to call NetApp phone support to sort this out.
My only thought is that it could be a broadcast domain/LIF failover group configuration.
I'd check those commands and see if you can make some headway. Without knowing your config it's hard to dig further. A case definitely isn't a bad option.