Host Initiators not logged in to NetApp cluster using some of the paths
Applies to
- ONTAP 9
- FCP
- SAN
- VMware ESXi
Issue
- Host initiators are zoned, but not logged into NetApp resulting in path mismatch on host end.
- The lifs and physical ports are up and online on NetApp end, the same can be verified using fcp adapter show and net int show command.
- SFP tx and rx power are in optimal range on storage end.
- Some of the ESXi host initiators are not logged in to all the target LIF.
fcp initiator.xml log
shows the host not logged in using the lifs of all the nodes of the cluster. - There are other unaffected hosts logged in to the same NetApp ports using the same lifs.
- Every time there is a host reboot, we are seeing some or the other paths as missing. This tells me that post reboot the host is unable to initiate the session back to storage.
vserver fcp ping-igroup
show shows zone-info-not-available for some of the paths.
fcp ping-igroup show -vserver abc -igroup xyz -check-fabric true
AFF250::> vserver fcp ping-igroup show
Igroup Logical Node Ping Extended
Vserver Name WWPN Interface Name Status Status
--------- ----------- -------------- ---------- --------- -------- -----------
AFFVS01
ESX01 10:00:00:10:9b:xx:xx:dc AFFA250-01_fc_lif_1 AFFA250-01 reachable wwpn-logged-in
ESX01 10:00:00:10:9b:xx:xx:dc AFFA250-01_fc_lif_2 AFFA250-01 reachable wwpn-logged-in
ESX01 10:00:00:10:9b:xx:xx:dc AFFA250-01_fc_lif_3 AFFA250-01 not-zoned -
ESX01 10:00:00:10:9b:xx:xx:dc AFFA250-01_fc_lif_4 AFFA250-01 not-zoned -
ESX01 10:00:00:10:9b:xx:xx:dc AFFA250-02_fc_lif_1 AFFA250-02 zone-info-not-available -
ESX01 10:00:00:10:9b:xx:xx:dc AFFA250-02_fc_lif_2 AFFA250-02 reachable wwpn-logged-in
- On switch end , both host and storage ports are logged in as FC4 type under
nsshow
log output.
N 685546; 3;10:00:00:10:9b:cc:xy:xx;20:00:00:10:9b:cc:yy:xx; 0x00000003
SCR: Fabric-Detected Nx-Port-Detected
FC4s: FCP FC-NVMe
PortSymb: [34] "Emulex PPN-10:00:00:10:9b:cc:ab:xx"
NodeSymb: [94] "Emulex LPm32002-D FV14.0.539.26 DV14.0.622.0 HN:abcc.hbcxxm.com OS:VMware ESXi 7.0.3"
Fabric Port Name: 20:55:88:94:71:a2:88:xx
Permanent Port Name: 20:11:d8:1f:cc:69:75:xx
Device type: NPIV Initiator
Port Index: 85
Redirect: No
Partial: No
LSAN: No
Slow Drain Device: No
Device link speed: 32G
Connected through AG: Yes
Real device behind AG: Yes
FCoE: No
Connected through FC-LAG: No
FC4 Features [FCP]: Initiator
FC4 Features [FC-NVMe]: Initiator
N 68fb41; 3;20:07:00:a0:98:f3:yy:xy;20:08:00:a0:98:f3:xx:xx; 0x00000003
SCR: Fabric-Detected Nx-Port-Detected
FC4s: FCP Application-Services
PortSymb: [63] "NetApp FC Target Port (LPe16000) VVMxxx_DATA_SVM:lif"
NodeSymb: [34] "NetApp Vserver SVM"
Fabric Port Name: 20:fb:88:94:71:a2:88:ab
Permanent Port Name: 50:0a:09:81:80:61:fa:aa
Device type: NPIV Target
Port Index: 251
Redirect: No
Partial: No
LSAN: No
Slow Drain Device: No
Device link speed: 16G
Connected through AG: No
Real device behind AG: No
FCoE: No
Connected through FC-LAG: No
FC4 Features [FCP]: Target
- Ensure the correct ONTAP cli context to prevent false possitive output fcp ping-initiator.
- Below steps need to be implemeted to siolate the issue further:
- Rescan storage adapters (
# esxcli storage core adapter rescan -A vmhbaX )
- Toggle the switch port connected to host initiator and verify, This can be done using
portdisable
command on Brocade switch andshutdown
command on Cisco switch. - Reset the Host HBA on VMware (
# esxcli storage san fc reset -A vmhbaX )
- Reset the LIFs on NetApp end using below command:
NOTE : If the storage LIFs are zoned with any other hosts, ensure those hosts have proper redundant paths and multipathing configured. |
::> net int modify -vserver <vserver_name> -lif <lif_name> -status-admin down
::> net int modify -vserver <vserver_name> -lif <lif_name> -status-admin up