DM5000H Ontap 9.7 Alert Mesage

[ahmar]

Hi Team,

Kindly advice and suggest.As we are getting alerts as per attached doc.

RUH-PS-######::> network interface show -failover
Logical Home Failover Failover
Vserver Interface Node:Port Policy Group
-------- --------------- --------------------- --------------- ---------------
Cluster
RUH-PS-DM5###H-01_clus1 RUH-PS-DM5000H-01:e0a
local-only Cluster
Failover Targets: RUH-PS-DM5000H-01:e0a,
RUH-PS-DM5###H-01:e0b
RUH-PS-DM5000H-01_clus2 RUH-PS-DM5000H-01:e0b
local-only Cluster
Failover Targets: RUH-PS-DM5000H-01:e0b,
RUH-PS-DM5000H-01:e0a
RUH-PS-DM5000H-02_clus1 RUH-PS-DM5000H-02:e0a
local-only Cluster
Failover Targets: RUH-PS-DM5000H-02:e0a,
RUH-PS-DM5000H-02:e0b
RUH-PS-DM5000H-02_clus2 RUH-PS-DM5000H-02:e0b
local-only Cluster
Failover Targets: RUH-PS-DM5000H-02:e0b,
RUH-PS-DM5000H-02:e0a
RUH-PS-DM5000H
RUH-PS-DM5000H-01_mgmt1 RUH-PS-DM5000H-01:e0M
local-only Management
Failover Targets: RUH-PS-DM5000H-01:e0M
RUH-PS-DM5000H-02_mgmt1 RUH-PS-DM5000H-02:e0M
local-only Management
Failover Targets: RUH-PS-DM5000H-02:e0M
cluster_mgmt RUH-PS-DM5000H-01:e0M broadcast-domain-wide
Management
Failover Targets: RUH-PS-DM5000H-01:e0M,
RUH-PS-DM5000H-02:e0M
SVM_iSCSI
RUH-PS-DM5000H-01_iscsi_lif_1 RUH-PS-DM5000H-01:e0c
disabled
Failover Targets: RUH-PS-DM5000H-01:e0c
RUH-PS-DM5000H-01_iscsi_lif_2 RUH-PS-DM5000H-01:e0d
disabled
Failover Targets: RUH-PS-DM5000H-01:e0d
RUH-PS-DM5000H-02_iscsi_lif_1 RUH-PS-DM5000H-02:e0c
disabled -
Failover Targets: RUH-PS-DM5000H-02:e0c
RUH-PS-DM5000H-02_iscsi_lif_2 RUH-PS-DM5000H-02:e0d
disabled -
Failover Targets: RUH-PS-DM5000H-02:e0d
11 entries were displayed.

[dbenadib]

Hi,

Regarding the messages you get it says that mgmt lifs (node mgmt ips) does not have any failover target. This means that if e0M is failing the mgmt ip won't be abble to move to any other port.

No big deal .. at least for iSCSI connectivity .

If you want to avoid this error msg U could then add mgmt vlan to any other port of the node and add it to mgmt broadcast domain then those lifs will have a failover target 😉

BR

[ahmar]

Thanks for your input, As per the  best practice, what your suggest?

Also if we make it true for the below, will it help anyway to get rid of the alert?

  Logical Status Network Current Current Is
Vserver Interface Admin/Oper Address/         Mask Node               Port     Home

cluster_mgmt up/up 10.2.211.170/24 RUH-PS-DM5000H-02     e0M      false

[dbenadib · Accepted solution]

The cluster mgmt can faiover to different node into a cluster. The node mgmt cannot that is the reason why you get those alert.

On a best practice point of view, I would add another 1GbE port to the mgmt broadcast domain.

To validate lif failover you can run the cmd: net int show -failover

it will show you on which port a lif could then jump in case of failure.

[TMACMD]

Also, not a great idea to use an access port (e0M) and a VLAN-tag (a0a-123) in the same broadcast-domain.

Active IQ Config Advisor will complain about that configuration.

It is best (as indicated) to use only access ports when including e0M (typically, the Default broadcast-domain)

[ahmar]

Thanks for your suggestion.

So what you suggest us to do, just ignore the message?

And live with it as it will do no harm.

regards

[TMACMD]

If you have any free ports that are not in use you can configure the switch and add them. Most of the smaller platforms only have the single e0M port and that is that.

 You see the e0M will fail over to the partner port. The message is about local failover ports. If you can’t do it, you can’t do it. 

We see this a lot and can ignore if the rest is correct. Remember, there is no data going over e0M. Additionally, your SVMs should have a mgmt port

[ahmar]

Good observation, since this is our only concern and rest all is good. So will ignore it, but is there a way to disable this alert?

Thanks

[TMACMD]

What platform is this?

 Output of “net port show -role !Cluster”

[ahmar]

Pls see if the attached file help to get the info

[TMACMD]

I did ask for the output of another command (which would have made the output much shorter).

I see what I need here. You could order 2 x X6568-R6 (332-00374, FCLF8522P2BTL-NQ) which are supported 1Gb transceivers. (one for each controller)

Now, this is a rebranded Lenovo unit, unless Lenovo has another support matrix for transceivers, that should work.

You could plug those in, get 1Gb (RJ45) connections and add to the broadcast domain if you really want to get rid of those messages.

Or get a 2 x  X6589-R6 (332-00279R6, AFBR-703SMZ-NA2, AFBR-709SMZ-NA2, FTLX8571D3BCL-(NQ), FTLX8574D3BCL-NQ) which are supported 10Gb optics. Plug up the optic as an access port and be done.

The third option I just remembered that I occasionally do: Set the NATIVE VLAN of the ifgroup to the same as e0M (if your network supports it). The Native VLAN is not tagged and if you then add a0a (or whatever your ifgrp is), that should stop the message.

You could also get 

[ahmar]

Thanks for your deep investigation and input, really appreciate it.

I will see if we can proceed with the third option. Otherwise will stay as it is.

Thanks again