ONTAP 9.6 FC issue with vmware HBA don't see NetApp

[StephaneGilliers]

Hello,

It's the first time I post on this community,

I have one problem with FC and would like just have some precisions with Vmware don't see lun.

The connexion to vmware don't working

igroup show
          Vserver Name: svm_fc
           Igroup Name: ESX01
              Protocol: fcp
               OS Type: vmware
Portset Binding Igroup: fc_pset_1
                  ALUA: true
            Initiators: 21:00:00:24 (not logged in)

And if I make a fcp ping initiator on ESX's WWPN I have

    Remote WWPN: 21:00:00:24:ff
       LIF Name: NODE-01_fc_lif_1
           Node: NODE-01
    Ping Status: zone-info-not-available

The topology show see the fabric 

vserver fcp topology show -vserver svm_fc
svm_fc
        NODE-01_fc_lif_1
                       1      SWITCH_BROCADE300_1 10:00:00:05:33:  24

Do you know what exactly "zone-info-not-available" are ?

Do you know where I can found more logs ?

Firmware Version of switch are check in IMT.

Thank you very much,

[SpindleNinja]

Are there other luns that you can see?  

Do you have zoning in place between the initiators (host(s)) and the Targets (NetApp) ?    It doesn't look like that host is logging in to the storage.  

[StephaneGilliers]

Hello,

Thank you for your quick reply.

No AFF are the first connected to existent Fabric with EMC FC Storage.

- Zoning (it's not me which manage brocade) at make by specify WWPN of LIF Netapp (Start by 20: and not 50:) and WWPN of Vmware.

- We check also NPIV on Brocade.

- We make a portset in SVM mapped to igroup

- We make an igroup in SVP mapped to LUN with WWPN of Vmware HBA

- We check also the LUN ID for not have an existing LUN ID in ESX

If I do a "network fcp topology show" I can see the switch with all ports and WWPN connected

And if I do "network fcp zone show" (I don't have a valid system for check what he must return) :

::> network fcp zone show
Warning: Unable to list entries for kernel on node "NODE-01": Fabric query operation timed out or was rejected.
         Unable to list entries for kernel on node "NODE-02": Fabric query operation timed out or was rejected.
Error: show failed: Fabric query operation timed out or was rejected

I have no more idea what I can do for check if this is a netapp problem.

Thank's,

[SpindleNinja]

if you run "fcp initiator show"  does it show anything?  if nothing, then the issue is most likely in the zoning.   

Do you see the WWPNs flogi-ing on the switchs?   

can you also post the output of the following? 

net port show 

net int show -vsever <fcp vserver name> 

portset show 

igroup show 

lun mapping show 

[StephaneGilliers]

I try to give you all information (from my note) since i have no access to storage until tomorrow.

I hope you see someting !!

::> vserver fcp initiator show -vserver svm_fc
There are no entries matching your query.

::> ucadmin show
                       Current  Current    Pending  Pending    Admin
Node          Adapter  Mode     Type       Mode     Type       Status
------------  -------  -------  ---------  -------  ---------  -----------
NODE-01    0c       cna      target     -        -          online
NODE-01    0d       cna      target     -        -          online
NODE-02    0c       cna      target     -        -          online
NODE-02    0d       cna      target     -        -          online

::> network interface show -fields wwpn,curr-port,lif (LIF are fine data and fc protocol)
svm_fc        NODE-01_fc_lif_1                         0c        20:05:d0:39:ea:09
svm_fc        NODE-01_fc_lif_2                         0d        20:06:d0:39:ea:09
svm_fc        NODE-02_fc_lif_1                         0c        20:07:d0:39:ea:09
svm_fc        NODE-02_fc_lif_2                         0d        20:08:d0:39:ea:09

::vserver fcp interface> show
           Logical    Status                      Current       Current Is
Vserver    Interface  Admin/Oper  WWPN            Node          Port    Home
---------- ---------- ----------- --------------- ------------- ------- ------
svm_fc 
           NODE-01_fc_lif_1 
                         up/up    20:05:d0:39:ea:09
                                                  NODE-01   0c      true
svm_fc 
           NODE-02_fc_lif_1 
                         up/up    20:07:d0:39:ea:09
                                                  NODE-02   0c      true

::> portset show -instance
    Vserver Name: svm_fc
    Portset Name: fc_pset_1
 LIF Or TPG Name: NODE-01_fc_lif_1, NODE-02_fc_lif_1
        Protocol: fcp
 Number Of Ports: 2
Bound To Igroups: IGROUP_ESX

::> igroup show -instance
          Vserver Name: svm_fc
           Igroup Name: IGROUP_ESX
              Protocol: fcp
               OS Type: vmware
Portset Binding Igroup: fc_pset_1
           Igroup UUID: c817c090-06dd-11ea-ab7c-xxxxxxxxxx
                  ALUA: true
            Initiators: 21:00:00:24:ff (ESX01) (not logged in)

lun show -m
Vserver    Path                                      Igroup   LUN ID  Protocol
---------- ----------------------------------------  -------  ------  --------
svm_fc 
           /vol/lun_test_01_vol/lun_test_01          IGROUP_ESX  50  fcp

-> OS Type : vmware

::> vol show -instance        
Vserver   Volume       Aggregate    State      Type       Size  Available Used%
--------- ------------ ------------ ---------- ---- ---------- ---------- -----
svm_fc 
          lun_test_01_vol 
                       NODE_01_SSD_1 
                                    online     RW      330.7GB    29.46GB   91%

Thank you,

[SpindleNinja]

Huh.  Your ports need to be in FC mode, not cna.       Are there FC SFPs in there? 

give this a read through how to convert:   

https://docs.netapp.com/ontap-9/index.jsp?topic=%2Fcom.netapp.doc.dot-cm-sanag%2FGUID-E2CD31BE-28EC-468E-AF49-83A641575799.html

[StephaneGilliers]

Ah sory I take the wrong copy/paste.

Yes the convert operation are good make and nose rebooted (even if I don’t see fabric characteristics and topology).

For sure I launch the command tomorrow.

Do you know how works the fcp ping command ? I don’t see result on docs for the explanation of results. 

Thanks.

[SpindleNinja]

fcp ping-initiator and -igroup   will ping between target fc lifs and the initiators.   

https://docs.netapp.com/ontap-9/index.jsp?topic=%2Fcom.netapp.doc.dot-cm-cmpr-960%2Fvserver__fcp__ping-initiator__show.html&cp=4_1_28_23_10_0

https://docs.netapp.com/ontap-9/index.jsp?topic=%2Fcom.netapp.doc.dot-cm-cmpr-960%2Fvserver__fcp__ping-igroup__show.html&cp=4_1_28_23_9_0

[StephaneGilliers]

Ok if I good understand :

- ping igroup : I must specify in WWPN the LIF

- ping initiator : I must specify in WWPN vmware

do you think it’s « technically » working if I specify the WWPN of Fabric Switch ?

if I have « zone-info-not-available » it seems netapp correctly go until the switch and after zoning it’s not good ? That’s seems good configured in AFF side ?

thank you ,

[StephaneGilliers]

Hello,

I check on AFF ucadmin are fine :

::> ucadmin show
                       Current  Current    Pending  Pending    Admin
Node          Adapter  Mode     Type       Mode     Type       Status
------------  -------  -------  ---------  -------  ---------  -----------
NODE-01   0c       fc       target     -        -          online
NODE-02   0c       fc       target     -        -          online

[Ontapforrum]

Hi,

I see that it's a brocade switch. What is the Brocade FabOS version ?

Could you try bouncing the LIF once:

::> net int modify -vserver <vserver_name> -lif <lif_name> -status-admin down
::> net int modify -vserver <vserver_name> -lif <lif_name> -status-admin up

After bouncing, check if that makes Initiators login ?

If not, then send the output of this command from the Brocade switch:

switch1:admin> nsshow

[StephaneGilliers]

Hi,

Yesterday I down/up the LIF again. I have restarted ESX today but not better.

In switch the version are check in IMT for brocade and ONTAP 9.6

Version:
Kernel:     2.6.14.2
Fabric OS:  v7.4.2c
Made on:    Thu Feb 1 00:07:10 2018
Flash:    Mon Nov 18 15:14:58 2019
BootProm:   1.0.11

I found in supportshow the LIF WWPN (start 20:05) and WWNN (start 20:09) and in permanent port name the Hardware WWPN of NODE1-0c port.

N    021001;      3;20:05:d0:39:ea:09:10:df;20:09:d0:39:ea:09:10:df; 0x00000003
    FC4s: FCP 
    PortSymb: [61] "NetApp FC Target Port (8324) svm_fc:NODE-01_fc_lif_1"
    NodeSymb: [26] "NetApp Vserver svm_fc"
    Fabric Port Name: 20:10:c4:f5:7c:6c:78:18 
    Permanent Port Name: 50:0a:09:83:80:83:1c:4c
    Port Index: 16
    Share Area: No
    Device Shared in Other AD: No
    Redirect: No 
    Partial: No
    LSAN: No

I see that also with LIF WWPN

portId:    021000
portIfId:    43020007
portWwn:   20:10:c4:f5:7c:6c:78:18
portWwn of device(s) connected:
20:05:d0:39:ea:09:10:df
50:0a:09:83:80:83:1c:4c
Distance:  normal
portSpeed: N8Gbps

I see portloginshow with WWPN LIF and NODE1-0c WWPN :

portloginshow 16
Type  PID     World Wide Name        credit df_sz cos
=====================================================
  fd  021001 20:05:d0:39:ea:09:10:df     6  2048   8  scr=0x3
  fe  021000 50:0a:09:83:80:83:1c:4c     6  2048   8  scr=0x0
  ff  021001  20:05:d0:39:ea:09:10:df     0     0   8  d_id=FFFFFC  
  ff  021000  50:0a:09:83:80:83:1c:4c     0     0   8  d_id=FFFFFA  
  ff  021000  50:0a:09:83:80:83:1c:4c     0     0   8  d_id=FFFFFC  

I see also ESX (WWPN I enter in igroup : 21:00:00:24:ff:13:03:90)

 N    020d00;      3;21:00:00:24:ff:13:03:90;20:00:00:24:ff:13:03:90; 0x00000003
    FC4s: FCP 
    NodeSymb: [32] "QLE2560 FW:v8.05.00 DVR:v3.0.1.0"
    Fabric Port Name: 20:0d:c4:f5:7c:6c:78:18 
    Permanent Port Name: 21:00:00:24:ff:13:03:90
    Port Index: 13
    Share Area: No
    Device Shared in Other AD: No
    Redirect: No 
    Partial: No
    LSAN: No

I'll ask (I'm afraid than I have only the answer monday morning) to make a fcp ping in brocade to the netapp and also to check the zoning with WWPN of LIF.

Thank you very much,

[SpindleNinja]

Is there a zone in place between the host and the netapp port? 

[StephaneGilliers]

Hello,

I see that on switch (is information you need) ?

alias:	PRDGFPSTSPA01_e0c	
		20:05:d0:39:ea:09:10:df
alias:	PRDGFPSTSPB01_e0c	
		20:07:d0:39:ea:09:10:df
alias:	ESX1_DELL_PORT0	
		21:00:00:24:ff:13:03:90

zone:	ESX1_DELL_PORT0_PRDGFPSTSPA01_e0c	
		ESX1_DELL_PORT0; PRDGFPSTSPA01_e0c
zone:	ESX1_DELL_PORT0_PRDGFPSTSPB01_e0c	
		ESX1_DELL_PORT0; PRDGFPSTSPB01_e0c

Thank you,

[StephaneGilliers]

Just for information I have the hand on switch brocade and try to do a fcp ping on WWPN netapp (hardware and LIF) :

SWITCH_PROD_1:admin> fcping --number 5 20:05:d0:39:ea:09:10:df
Destination:	20:05:d0:39:ea:09:10:df
Pinging 20:05:d0:39:ea:09:10:df [0x021001] with 12 bytes of data:
received reply from 20:05:d0:39:ea:09:10:df: 12 bytes time:583 usec
received reply from 20:05:d0:39:ea:09:10:df: 12 bytes time:571 usec
received reply from 20:05:d0:39:ea:09:10:df: 12 bytes time:560 usec
received reply from 20:05:d0:39:ea:09:10:df: 12 bytes time:560 usec
received reply from 20:05:d0:39:ea:09:10:df: 12 bytes time:593 usec
5 frames sent, 5 frames received, 0 frames rejected, 0 frames timeout
Round-trip min/avg/max = 560/573/593 usec
SWITCH_PROD_1:admin> 
SWITCH_PROD_1:admin> fcping --number 5 50:0a:09:83:80:83:1c:4c
Destination:	50:0a:09:83:80:83:1c:4c
Pinging 50:0a:09:83:80:83:1c:4c [0x021000] with 12 bytes of data:
received reply from 50:0a:09:83:80:83:1c:4c: 12 bytes time:763 usec
received reply from 50:0a:09:83:80:83:1c:4c: 12 bytes time:661 usec
received reply from 50:0a:09:83:80:83:1c:4c: 12 bytes time:562 usec
received reply from 50:0a:09:83:80:83:1c:4c: 12 bytes time:590 usec
received reply from 50:0a:09:83:80:83:1c:4c: 12 bytes time:558 usec
5 frames sent, 5 frames received, 0 frames rejected, 0 frames timeout
Round-trip min/avg/max = 558/626/763 usec

It's seems to be a good point.

[Ontapforrum]

To be honest, all seems ok to me. I don't have FC environment, so can't really verify things from my end.

Do you mind sharing this screenshot ?

1) Open System Manager
2) Under storage on the left side menu, go to SVM
3) Select the SVM whcih is serving the FC LUN
4) Click SVM settings
5) In the Protocols pane, click FC

Do you see FC service is running ?

Plz send the screenshot showing Network Interface, WWPN, Current Port & Status ?

Also, ::>event log show  [Do you see anything here that needs attention or informational related to FC]

[StephaneGilliers]

Hello,

Just for your information someone know well FC/Vmware had look an eye.

He check all and for him the problem are in Vmware.

Switch zoning and Netapp FC are fine (portcamshow for see SID and DID, I'm not expert on FC but I remark that).

Some answer I shared with you :

With FCP ping command you cannot ping the WWPN of brocade switch just the initiator (but the switch can make a fcp ping to NetApp.

A case will be open on Vmware, of course I take you a feedback when I have the answer.

I hope I have the answer tomorrow,

Thank you very much for your reactivity which permit to confort me in not an ONTAP problem.

[Ontapforrum]

Thanks for the update and for sharing information with us.  Will be glad to hear from your interaction with VMware.

[StephaneGilliers · Accepted solution]

Hello,

It's ok the LUN are mounting.

In fact drivers of Vmware's HBA need to be upgraded  IMT was checked just for fabric.

Thank you very much for your implication.