2016-12-02 02:04 PM
We have Ontap 8.3.2.
I created a vserver to test the SVM DR in our cluster. It's successful. But when I try to snapmirror our production vserver to a SVM DR vserver, its failed to initialize the SVM DR snapmirror relatioinship.
The production vserver has a volume 1.9TB.
The error message was :
Last Transfer Error: Failed to apply the source Vserver configuration. Reason: Apply failed for Object: hosts_byname Method: baseline. Reason: The IPV4 address specified with "-address" is not supported because it is one of the following:
multicast, loopback or 0.0.0.0.
The SVM DR vserver contains only the volumes and nis services from source vserver but no dns service and no cifs protocol.
Solved! SEE THE SOLUTION
2016-12-05 08:23 AM
This the vserver svm1 host file table. Are those 127.x.x.x loopback entries?
cluster::> dns hosts show -vserver svm1
Vserver Address Hostname Aliases
---------- -------------- --------------- ----------------------
svm1 127.0.0.1 localhost localhost-stack
svm1 127.0.10.1 localhost-10 localhost-bsd
svm1 127.0.20.1 localhost-20 localhost-sk
svm1 x.x.x.218 svm1-e0a-280 -
svm1 x.x.x.108 nismaster -
svm1 x.x.x.112 svm1 svm1-e0a
svm1 x.x.x.x svm1-e0a-4038 -
2016-12-07 12:34 PM
Yes, those loopback IPs caused the SVM DR snapmirror failure.
So I have questions about the dns hosts file, if you think it is irrevalent to this SVM DR topic, then I will either post to another topic or open a case with NetApp support.
1. The host files entries sometime are necessery at the Unix but How critical is the host file entries for the C-DOT vserver?
2. All our vservers host file contain this loopback IPs. Maybe during the 7-mode transition we added them to the host file but I am not sure about this. Anyway, if I delete the 127.0.0.1 IP from the host file, how this will affect the NFS access?
3. is there any workaround without update the host file entries?