1NVME-CONNECT-ALL(1) NVMe Manual NVME-CONNECT-ALL(1)
2
6 nvme-connect-all - Discover and Connect to Fabrics controllers.
7
9 nvme connect-all
10 [--transport=<trtype> | -t <trtype>]
11 [--traddr=<traddr> | -a <traddr>]
12 [--trsvcid=<trsvcid> | -s <trsvcid>]
13 [--host-traddr=<traddr> | -w <traddr>]
14 [--hostnqn=<hostnqn> | -q <hostnqn>]
15 [--hostid=<hostid> | -I <hostid>]
16 [--raw=<filename> | -r <filename>]
17 [--keep-alive-tmo=<#> | -k <#>]
18 [--reconnect-delay=<#> | -c <#>]
19 [--ctrl-loss-tmo=<#> | -l <#>]
20 [--hdr-digest | -g]
21 [--data-digest | -G]
22 [--nr-io-queues=<#> | -i <#>]
23 [--nr-write-queues=<#> | -W <#>]
24 [--nr-poll-queues=<#> | -P <#>]
25 [--queue-size=<#> | -Q <#>]
26
28 Send one or more Discovery requests to a NVMe over Fabrics Discovery
29 Controller, and create controllers for the returned discovery records.
30 If no parameters are given, then nvme connect-all will attempt to find
32 a /etc/nvme/discovery.conf file to use to supply a list of connect-all
33 commands to run. If no /etc/nvme/discovery.conf file exists, the
34 command will quit with an error.
35 Otherwise a specific Discovery Controller should be specified using the
37 --transport, --traddr and if necessary the --trsvcid and a Diѕcovery
38 request will be sent to the specified Discovery Controller.
39 See the documentation for the nvme-discover(1) command for further
41 background.
42
44 -t <trtype>, --transport=<trtype>
45 This field specifies the network fabric being used for a
46 NVMe-over-Fabrics network. Current string values include:
47 ┌──────┬────────────────────────────┐
49 │Value │ Definition │
50 ├──────┼────────────────────────────┤
51 │rdma │ The network fabric is an │
52 │ │ rdma network (RoCE, iWARP, │
53 │ │ Infiniband, basic rdma, │
54 │ │ etc) │
55 ├──────┼────────────────────────────┤
56 │fc │ WIP The network fabric is │
57 │ │ a Fibre Channel network. │
58 ├──────┼────────────────────────────┤
59 │loop │ Connect to a NVMe over │
60 │ │ Fabrics target on the │
61 │ │ local host │
62 └──────┴────────────────────────────┘
63 -a <traddr>, --traddr=<traddr>
65 This field specifies the network address of the Discovery
66 Controller. For transports using IP addressing (e.g. rdma) this
67 should be an IP-based address (ex. IPv4).
68 -s <trsvcid>, --trsvcid=<trsvcid>
70 This field specifies the transport service id. For transports using
71 IP addressing (e.g. rdma) this field is the port number. By
72 default, the IP port number for the RDMA transport is 4420.
73 -w <traddr>, --host-traddr=<traddr>
75 This field specifies the network address used on the host to
76 connect to the Discovery Controller.
77 -q <hostnqn>, --hostnqn=<hostnqn>
79 Overrides the default Host NQN that identifies the NVMe Host. If
80 this option is not specified, the default is read from
81 /etc/nvme/hostnqn first. If that does not exist, the autogenerated
82 NQN value from the NVMe Host kernel module is used next. The Host
83 NQN uniquely identifies the NVMe Host, and may be used by the the
84 Discovery Controller to control what NVMe Target resources are
85 allocated to the NVMe Host for a connection.
86 -I <hostid>, --hostid=<hostid>
88 UUID(Universally Unique Identifier) to be discovered which should
89 be formatted.
90 -r <filename>, --raw=<filename>
92 This field will take the output of the nvme connect-all command and
93 dump it to a raw binary file. By default nvme connect-all will dump
94 the output to stdout.
95 -k <#>, --keep-alive-tmo=<#>
97 Overrides the default keep alive timeout (in seconds). This option
98 will be ignored for discovery, but will be passed on to the
99 subsequent connect call.
100 -c <#>, --reconnect-delay=<#>
102 Overrides the default delay (in seconds) before reconnect is
103 attempted after a connect loss.
104 -l <#>, --ctrl-loss-tmo=<#>
106 Overrides the default controller loss timeout period (in seconds).
107 -g, --hdr-digest
109 Generates/verifies header digest (TCP).
110 -G, --data-digest
112 Generates/verifies data digest (TCP).
113 -i <#>, --nr-io-queues=<#>
115 Overrides the default number of I/O queues create by the driver.
116 This option will be ignored for discovery, but will be passed on to
117 the subsequent connect call.
118 -W <#>, --nr-write-queues=<#>
120 Adds additional queues that will be used for write I/O.
121 -P <#>, --nr-poll-queues=<#>
123 Adds additional queues that will be used for polling latency
124 sensitive I/O.
125 -Q <#>, --queue-size=<#>
127 Overrides the default number of elements in the I/O queues created
128 by the driver. This option will be ignored for discovery, but will
129 be passed on to the subsequent connect call.
130
132 · Connect to all records returned by the Discover Controller with IP4
133 address 192.168.1.3 for all resources allocated for NVMe Host name
134 host1-rogue-nqn on the RDMA network. Port 4420 is used by default:
135 # nvme connect-all --transport=rdma --traddr=192.168.1.3 \
137 --hostnqn=host1-rogue-nqn
138 · Issue a nvme connect-all command using a /etc/nvme/discovery.conf
140 file:
141 # Machine default 'nvme discover' commands. Query the
143 # Discovery Controller's two ports (some resources may only
144 # be accessible on a single port). Note an official
145 # nqn (Host) name defined in the NVMe specification is being used
146 # in this example.
147 -t rdma -a 192.168.69.33 -s 4420 -q nqn.2014-08.com.example:nvme:nvm-subsystem-sn-d78432
148 -t rdma -a 192.168.1.4 -s 4420 -q nqn.2014-08.com.example:nvme:nvm-subsystem-sn-d78432
149 At the prompt type "nvme connect-all".
151
153 nvme-discover(1) nvme-connect(1)
154
156 Part of the nvme-user suite
157NVMe 01/07/2020 NVME-CONNECT-ALL(1)