NVMe-oF host configuration for RHEL 9.4 with ONTAP

11/13/2024 Contributors

NVMe over Fabrics (NVMe-oF), including NVMe over Fibre Channel (NVMe/FC) and other transports, is supported with Red Hat Enterprise Linux (RHEL) 9.4 with Asymmetric Namespace Access (ANA). In NVMe-oF environments, ANA is the equivalent of ALUA multipathing in iSCSI and FC environments and is implemented with in-kernel NVMe multipath.

The following support is available for NVMe-oF host configuration for RHEL 9.4 with ONTAP:

Support for NVMe over TCP (NVMe/TCP) in addition to NVMe/FC. The NetApp plug-in in the native nvme-cli package displays ONTAP details for both NVMe/FC and NVMe/TCP namespaces.
Use of NVMe and SCSI co-existent traffic on the same host on a given host bus adapter (HBA) without the explicit dm-multipath settings to prevent claiming NVMe namespaces.

For additional details on supported configurations, see the NetApp Interoperability Matrix Tool.

Features

RHEL 9.4 has in-kernel NVMe multipath enabled for NVMe namespaces by default; therefore, there is no need for explicit settings.
SAN booting using the NVMe/FC protocol is supported.

Known limitations

There are no known limitations.

Validate software versions

You can use the following procedure to validate the minimum supported RHEL 9.4 software versions.

Steps

Install RHEL 9.4 on the server. After the installation is complete, verify that you are running the specified RHEL 9.4 kernel:
```
# uname -r
```
Example output:
```
5.14.0-423.el9.x86_64
```

Install the nvme-cli package:

# rpm -qa|grep nvme-cli

Example output:

nvme-cli-2.6-4.el9.x86_64

Install the libnvme package:

#rpm -qa|grep libnvme

Example output

libnvme-1.6-1.el9.x86_64

On the RHEL 9.4 host, check the hostnqn string at /etc/nvme/hostnqn:

# cat /etc/nvme/hostnqn

Example output

nqn.2014-08.org.nvmexpress:uuid: uuid:4c4c4544-0036-5610-804a-c7c04f365a32

Verify that the hostnqn string matches the hostnqn string for the corresponding subsystem on the ONTAP array:

::> vserver nvme subsystem host show -vserver vs_coexistence_LPE36002

Example output:

Vserver     Subsystem          Host NQN
----------- --------------- ----------------------------------------------------------
vs_coexistence_LPE36002   nvme    nqn.2014-08.org.nvmexpress:uuid: 4c4c4544-0036-5610-804a-

If the hostnqn strings do not match, use the vserver modify command to update the hostnqn string on your corresponding ONTAP array subsystem to match the hostnqn string from /etc/nvme/hostnqn on the host.

Configure NVMe/FC

You can configure NVMe/FC for Broadcom/Emulex or Marvell/Qlogic adapters.

Broadcom/Emulex

Steps

Verify that you are using the supported adapter model:

# cat /sys/class/scsi_host/host*/modelname

Example output:

LPe36002-M64
LPe36002-M64

# cat /sys/class/scsi_host/host*/modeldesc

Example output:

Emulex LightPulse LPe36002-M64 2-Port 64Gb Fibre Channel Adapter
Emulex LightPulse LPe36002-M64 2-Port 64Gb Fibre Channel Adapter

Verify that you are using the recommended Broadcom lpfc firmware and inbox driver:
```
# cat /sys/class/scsi_host/host*/fwrev
14.2.673.40, sli-4:6:d
14.2.673.40, sli-4:6:d


# cat /sys/module/lpfc/version
0:14.2.0.16
```
For the most current list of supported adapter driver and firmware versions, see the NetApp Interoperability Matrix Tool.

Verify that lpfc_enable_fc4_type is set to 3:

# cat /sys/module/lpfc/parameters/lpfc_enable_fc4_type
3

Verify that the initiator ports are up and running and that you can see the target LIFs:

# cat /sys/class/fc_host/host*/port_name
0x100000109b3c081f
0x100000109b3c0820

# cat /sys/class/fc_host/host*/port_state
Online
Online

# cat /sys/class/scsi_host/host*/nvme_info
NVME Initiator Enabled
XRI Dist lpfc0 Total 6144 IO 5894 ELS 250
NVME LPORT lpfc0 WWPN x100000109b3c081f WWNN x200000109b3c081f DID x062300 ONLINE
NVME RPORT       WWPN x2143d039ea165877 WWNN x2142d039ea165877 DID x061b15 TARGET DISCSRVC ONLINE
NVME RPORT       WWPN x2145d039ea165877 WWNN x2142d039ea165877 DID x061115 TARGET DISCSRVC ONLINE
NVME Statistics
LS: Xmt 000000040b Cmpl 000000040b Abort 00000000
LS XMIT: Err 00000000  CMPL: xb 00000000 Err 00000000
Total FCP Cmpl 000000001f5c4538 Issue 000000001f58da22 OutIO fffffffffffc94ea
abort 00000630 noxri 00000000 nondlp 00001071 qdepth 00000000 wqerr 00000000 err 00000000
FCP CMPL: xb 00000630 Err 0001bd4a
NVME Initiator Enabled
XRI Dist lpfc1 Total 6144 IO 5894 ELS 250
NVME LPORT lpfc1 WWPN x100000109b3c0820 WWNN x200000109b3c0820 DID x062c00 ONLINE
NVME RPORT       WWPN x2144d039ea165877 WWNN x2142d039ea165877 DID x060215 TARGET DISCSRVC ONLINE
NVME RPORT       WWPN x2146d039ea165877 WWNN x2142d039ea165877 DID x061815 TARGET DISCSRVC ONLINE
NVME Statistics
LS: Xmt 000000040b Cmpl 000000040b Abort 00000000
LS XMIT: Err 00000000  CMPL: xb 00000000 Err 00000000
Total FCP Cmpl 000000001f5c3618 Issue 000000001f5967a4 OutIO fffffffffffd318c
abort 00000629 noxri 00000000 nondlp 0000044e qdepth 00000000 wqerr 00000000 err 00000000
FCP CMPL: xb 00000629 Err 0001bd3d

Marvell/QLogic FC Adapter for NVMe/FC

The native inbox qla2xxx driver included in the RHEL 9.4 GA kernel has the latest fixes. These fixes are essential for ONTAP support.

Steps

Verify that you are running the supported adapter driver and firmware versions:

# cat /sys/class/fc_host/host*/symbolic_name

Example output

QLE2872 FW:v9.12.01 DVR:v10.02.09.100-k
QLE2872 FW:v9.12.01 DVR:v10.02.09.100-k

Verify that ql2xnvmeenable is set. This enables the Marvell adapter to function as an NVMe/FC initiator:
```
# cat /sys/module/qla2xxx/parameters/ql2xnvmeenable
1
```

Enable 1MB I/O (Optional)

ONTAP reports an MDTS (Max Data Transfer Size) of 8 in the Identify Controller data. This means the maximum I/O request size can be up to 1MB. To issue I/O requests of size 1 MB for a Broadcom NVMe/FC host, you must increase the lpfc value of the lpfc_sg_seg_cnt parameter to 256 from the default value of 64.

The following steps don't apply to Qlogic NVMe/FC hosts.

Steps

Set the lpfc_sg_seg_cnt parameter to 256:

cat /etc/modprobe.d/lpfc.conf

Example output

options lpfc lpfc_sg_seg_cnt=256

Run the dracut -f command, and reboot the host:
Verify that lpfc_sg_seg_cnt is 256:
```
cat /sys/module/lpfc/parameters/lpfc_sg_seg_cnt
```
The expected value is 256.

Configure NVMe/TCP

NVMe/TCP does not have an auto-connect functionality. Instead, you can discover the NVMe/TCP subsystems and namespaces by performing the NVMe/TCP connect or connect-all operations manually.

Steps

Verify that the initiator port can fetch the discovery log page data across the supported NVMe/TCP LIFs:

nvme discover -t tcp -w host-traddr -a traddr

Example output:

# nvme discover -t tcp -w 192.168.167.1 -a 192.168.167.16

Discovery Log Number of Records 8, Generation counter 10
=====Discovery Log Entry 0======
trtype:  tcp
adrfam:  ipv4
subtype: current discovery subsystem
treq:    not specified
portid:  11
trsvcid: 8009
subnqn:  nqn.1992-08.com.netapp:sn.983de7f4b39411ee871ed039ea954d18:
discovery
traddr:  192.168.167.8
eflags:  explicit discovery connections, duplicate discovery information
sectype: none
=====Discovery Log Entry 1======
trtype:  tcp
adrfam:  ipv4
subtype: current discovery subsystem
treq:    not specified
portid:  9
trsvcid: 8009
subnqn:  nqn.1992-08.com.netapp:sn.983de7f4b39411ee871ed039ea954d18:
discovery
traddr:  192.168.166.8
eflags:  explicit discovery connections, duplicate discovery information
sectype: none
=====Discovery Log Entry 2======
trtype:  tcp
adrfam:  ipv4
subtype: current discovery subsystem
treq:    not specified
portid:  12
trsvcid: 8009
subnqn:  nqn.1992-08.com.netapp:sn.983de7f4b39411ee871ed039ea954d18:
discovery
traddr:  192.168.167.7
eflags:  explicit discovery connections, duplicate discovery information
sectype: none
=====Discovery Log Entry 3======
trtype:  tcp
adrfam:  ipv4
subtype: current discovery subsystem
treq:    not specified
portid:  10
trsvcid: 8009
subnqn:  nqn.1992-08.com.netapp:sn.983de7f4b39411ee871ed039ea954d18:
discovery
traddr:  192.168.166.7
eflags:  explicit discovery connections, duplicate discovery information
sectype: none
=====Discovery Log Entry 4======
trtype:  tcp
adrfam:  ipv4
subtype: nvme subsystem
treq:    not specified
portid:  11
trsvcid: 4420
subnqn:  nqn.1992-08.com.netapp:sn.983de7f4b39411ee871ed039ea954d18:subsystem.nvme_tcp_1
traddr:  192.168.167.8
eflags:  none
sectype: none
=====Discovery Log Entry 5======
trtype:  tcp
adrfam:  ipv4
subtype: nvme subsystem
treq:    not specified
portid:  9
trsvcid: 4420
subnqn:  nqn.1992-08.com.netapp:sn.983de7f4b39411ee871ed039ea954d18:subsystem.nvme_tcp_1
traddr:  192.168.166.8
eflags:  none
sectype: none
=====Discovery Log Entry 6======
trtype:  tcp
adrfam:  ipv4
subtype: nvme subsystem
treq:    not specified
portid:  12
trsvcid: 4420
subnqn:  nqn.1992-08.com.netapp:sn.983de7f4b39411ee871ed039ea954d18:subsystem.nvme_tcp_1
traddr:  192.168.167.7
eflags:  none
sectype: none
=====Discovery Log Entry 7======
trtype:  tcp
adrfam:  ipv4
subtype: nvme subsystem
treq:    not specified
portid:  10
trsvcid: 4420
subnqn:  nqn.1992-08.com.netapp:sn.983de7f4b39411ee871ed039ea954d18:subsystem.nvme_tcp_1
traddr:  192.168.166.7
eflags:  none
sectype: none

Verify that the other NVMe/TCP initiator-target LIF combinations are able to successfully fetch discovery log page data:

nvme discover -t tcp -w host-traddr -a traddr

Example output:

#nvme discover -t tcp -w 192.168.166.6 -a 192.168.166.7
#nvme discover -t tcp -w 192.168.166.6 -a 192.168.166.8
#nvme discover -t tcp -w 192.168.167.6 -a 192.168.167.7
#nvme discover -t tcp -w 192.168.167.6 -a 192.168.167.8

Run the nvme connect-all command across all the supported NVMe/TCP initiator-target LIFs across the nodes:

nvme connect-all -t tcp -w host-traddr -a traddr

Example output:

#	nvme	connect-all	-t	tcp	-w	192.168.166.6	-a	192.168.166.7
#	nvme	connect-all	-t	tcp	-w	192.168.166.6	-a	192.168.166.8
#	nvme	connect-all	-t	tcp	-w	192.168.167.6	-a	192.168.167.7
#	nvme	connect-all	-t	tcp	-w	192.168.167.6	-a	192.168.167.8

Beginning with RHEL 9.4, the default setting for the NVMe/TCP ctrl_loss_tmo timeout is turned off. This means there is no limit on the number of retries (indefinite retry). Consequently, you don't need to manually configure a specific ctrl_loss_tmo timeout duration when using the nvme connect or nvme connect-all commands (option -l ). With this default behavior, the NVMe/TCP controllers don't experience timeouts in the event of a path failure and remain connected indefinitely.

Validate NVMe-oF

You can use the following procedure to validate NVME-oF.

Steps

Verify that the in-kernel NVMe multipath is enabled:
```
# cat /sys/module/nvme_core/parameters/multipath
Y
```
Verify that the appropriate NVMe-oF settings (such as, model set to NetApp ONTAP Controller and load balancing iopolicy set to round-robin) for the respective ONTAP namespaces correctly reflect on the host:
```
# cat /sys/class/nvme-subsystem/nvme-subsys*/model
NetApp ONTAP Controller
NetApp ONTAP Controller
```
```
# cat /sys/class/nvme-subsystem/nvme-subsys*/iopolicy
round-robin
round-robin
```

Verify that the namespaces are created and correctly discovered on the host:

# nvme list

Example output:

Node         SN                   Model
---------------------------------------------------------
/dev/nvme4n1 81Ix2BVuekWcAAAAAAAB	NetApp ONTAP Controller


Namespace Usage    Format             FW             Rev
-----------------------------------------------------------
1                 21.47 GB / 21.47 GB	4 KiB + 0 B   FFFFFFFF

Verify that the controller state of each path is live and has the correct ANA status:

NVMe/FC

# nvme list-subsys /dev/nvme5n21

Example output:

nvme-subsys4 - NQN=nqn.1992-08.com.netapp:sn.efd7989cb10111ee871ed039ea954d18:subsystem.nvme
            hostnqn=nqn.2014-08.org.nvmexpress:uuid:d3b581b4-c975-11e6-8425-0894ef31a074
 iopolicy=round-robin
 \
  +- nvme2 fc traddr=nn-0x2013d039ea951c45:pn-0x2018d039ea951c45,host_traddr=nn-0x200000109bdacc76:pn-0x100000109bdacc76 live non-optimized
  +- nvme3 fc traddr=nn-0x2013d039ea951c45:pn-0x2017d039ea951c45,host_traddr=nn-0x200000109bdacc75:pn-0x100000109bdacc75 live non-optimized
  +- nvme5 fc traddr=nn-0x2013d039ea951c45:pn-0x2016d039ea951c45,host_traddr=nn-   0x200000109bdacc76:pn-0x100000109bdacc76 live optimized
  +- nvme6 fc traddr=nn-0x2013d039ea951c45:pn-0x2014d039ea951c45,host_traddr=nn-  0x200000109bdacc75:pn-0x100000109bdacc75 live optimized

NVMe/TCP

# nvme list-subsys /dev/nvme1n1

Example output:

nvme-subsys1 -NQN=nqn.1992-08.com.netapp:
sn.983de7f4b39411ee871ed039ea954d18:subsystem.nvme_tcp_1         hostnqn=nqn.2014-08.org.nvmexpress:uuid:
4c4c4544-0035-5910-804b-c2c04f444d33
iopolicy=round-robin
\
+- nvme5 tcp traddr=192.168.166.7,trsvcid=4420,host_traddr=192.168.166.6,src_addr=192.168.166.6 live
+- nvme4 tcp traddr=192.168.166.8,trsvcid=4420,host_traddr=192.168.166.6,src_addr=192.168.166.6 live
+- nvme2 tcp traddr=192.168.167.7,trsvcid=4420,host_traddr=192.168.167.6,src_addr=192.168.167.6 live
+- nvme1 tcp traddr=192.168.167.8,trsvcid=4420,host_traddr=192.168.167.6,src_addr=192.168.167.6 live

Verify that the NetApp plug-in displays the correct values for each ONTAP namespace device:

Column

# nvme netapp ontapdevices -o column

Example output:

Device        Vserver   Namespace Path
----------------------- ------------------------------
/dev/nvme0n1 vs_tcp           /vol/vol1/ns1



NSID       UUID                                   Size
------------------------------------------------------------
1          6fcb8ea0-dc1e-4933-b798-8a62a626cb7f	21.47GB

JSON

# nvme netapp ontapdevices -o json

Example output

{

"ONTAPdevices" : [
{

"Device" : "/dev/nvme1n1", "Vserver" : "linux_tcnvme_iscsi", "Namespace_Path" : "/vol/tcpnvme_1_0_0/tcpnvme_ns", "NSID" : 1,
"UUID" : "1a42c652-1450-4a29-886a-b4ccc23e637d", "Size" : "21.47GB",
"LBA_Data_Size" : 4096,
"Namespace_Size" : 5242880
},

]
}

Known issues

There are no known issues for the NVMe-oF host configuration for RHEL 9.4 with ONTAP release.

NVMe-oF host configuration for RHEL 9.4 with ONTAP

Creating your file...

Features

Known limitations

Validate software versions

Configure NVMe/FC

Enable 1MB I/O (Optional)

Configure NVMe/TCP

Validate NVMe-oF

Known issues