1lsscsi(8) LSSCSI lsscsi(8)
2
6 lsscsi - list SCSI devices (or hosts), list NVMe devices
7
9 lsscsi [--brief] [--classic] [--controllers] [--device] [--generic]
10 [--help] [--hosts] [--kname] [--list] [--long] [--long-unit] [--lunhex]
11 [--no-nvme] [--pdt] [--protection] [--protmode] [--scsi_id] [--size]
12 [--sysfsroot=PATH] [--sz-lbs] [--transport] [--unit] [--verbose]
13 [--version] [--wwn] [H:C:T:L]
14
16 Uses information in sysfs (Linux kernel series 2.6 and later) to list
17 SCSI devices (or hosts) currently attached to the system. Many non-SCSI
18 storage devices (but not all) used the SCSI subsystem in Linux. In lss‐
19 csi version 0.30 support was added to list NVMe devices.
20 In single line per device (LU or NVMe namespace) mode, the default,
22 each line starts with a 4 element tuple surrounded by square brackets.
23 For SCSI devices the first element 'H' is the host number, the second
24 element 'C' is the controller number, the third element 'T' is the tar‐
25 get number and the final element is the Logical Unit Number (LUN). All
26 four are integers. For NVMe namespaces see two paragraphs down. When
27 the --hosts option is given for SCSI devices the tuple is reduced to
28 one element: the host number.
29 If a H:C:T:L tuple is given as an argument on the command line then it
31 acts as a filter and only devices that match it are listed. The colons
32 don't have to be present, and '-', '*', '?' or missing components at
33 the end are interpreted as wildcards. The default is '*:*:*:*' which
34 means to match devices (i.e. Logical Units). Any filter string using
35 '*' of '?' should be surrounded by single or double quotes to stop
36 shell expansions. If '-' is used as a wildcard then the whole filter
37 tuple should be prefixed by '-- ' to tell this utility there are no
38 more options on the command line to be interpreted. A leading '[' and
39 trailing ']' are permitted ( e.g. '[1:0:0]' matches all LUNs on 1:0:0).
40 May also be used to filter --hosts in which case only the H is active
41 and may be either a number or in the form "host<n>" where <n> is a host
42 number.
43 For NVMe devices and controllers almost all of the previous paragraph
45 applies. The main difference is that "N" appears in the 'H' (first)
46 position. The 'C' position for NVMe is the controller's Linux gener‐
47 ated "char" device minor number which is the first number that appears
48 in a typical NVMe controller name, for example: "/dev/nvme2". The 'T'
49 position for NVMe is the "CNTLID" value. The final 'L' position is the
50 NVMe namespace identifier which is typically a sequential value start‐
51 ing at 1. The leading explicit "N" for NVMe devices is converted inter‐
52 nally into a large value (32,767) that should not interfere with any
53 Linux generated SCSI host number; it also means that the numeric sort
54 used to show hosts (controllers) and devices (LUs or logical units)
55 will always place NVMe devices and controllers after those that use the
56 SCSI subsystem. To filter using a H:C:T:L argument for NVMe con‐
57 trollers, "hostN", "hostN:<num>", "N" or "N:<num>" may be used; when no
58 "<num>" is given, only NVMe controllers will be listed (i.e. it lists
59 no SCSI hosts (HBAs)).
60 By default in this utility device node names (e.g. "/dev/sda" or
62 "/dev/root_disk") are obtained by noting the major and minor numbers
63 for the listed device obtained from sysfs (e.g. the contents of
64 "/sys/block/sda/dev") and then looking for a match in the "/dev" direc‐
65 tory. This "match by major and minor" will allow devices that have been
66 given a different name by udev (for example) to be correctly reported
67 by this utility.
68 In some situations it may be useful to see the device node name that
70 Linux would produce by default, so the --kname option is provided. An
71 example of where this may be useful is kernel error logs which tend to
72 report disk error messages using the disk's default kernel name.
73
75 Arguments to long options are mandatory for short options as well. The
76 options are arranged in alphabetical order based on the long option
77 name. Hyphenated long options can also take underscore, and vice versa
78 (e.g. --scsi_id or --scsi-id are acceptable).
79 -b, --brief
81 reduces one line per device output to the tuple and the primary
82 device name. This may simplify scripts that process the output
83 of this utility. With the --generic option it will show on each
84 line the tuple (from which the bsg pass-through device name can
85 be deduced), the primary device name (which the block subsystem
86 uses) and the sg device name (also a pass-through).
87 When the --pdt option is used together with this option the SCSI
88 Peripheral Device Type (PDT) is displayed in hex (with a leading
89 "0x") between the tuple and the primary device name. For NVMe
90 namespaces "0x0" is displayed (for a disk or direct access
91 device).
92 -c, --classic
94 The output is similar to that obtained from 'cat
95 /proc/scsi/scsi'
96 -C, --controllers
98 Lists NVMe controllers and SCSI hosts. This is a synonym for the
99 --hosts option.
100 -d, --device
102 After outputting the (probable) SCSI device name the device node
103 major and minor numbers are shown in brackets (e.g.
104 "/dev/sda[8:0]").
105 -g, --generic
107 Output the SCSI generic device file name. Note that if the sg
108 driver is a module it may need to be loaded otherwise '-' may
109 appear.
110 NVMe does not have generic (char) devices in the same sense as
111 SCSI. Instead NVMe Admin, NVM (i.e. block type commands such as
112 Read and Write) and MI (Management Interface (e.g. to an enclo‐
113 sure)) commands are all sent to the containing controller whose
114 device name is shown when the --hosts option is used.
115 To unclutter the single line per device mode the --brief option
116 combined with this option should help.
117 -h, --help
119 Output the usage message and exit.
120 -H, --hosts
122 List the SCSI hosts and NVMe controllers currently attached to
123 the system. If this option is not given (and the --controllers
124 option is not given) then SCSI devices (logical units (LUs))
125 followed by NVMe devices (namespaces) are listed.
126 -k, --kname
128 Use Linux default algorithm for naming devices (e.g. block major
129 8, minor 0 is "/dev/sda") rather than the "match by major and
130 minor" in the "/dev" directory as discussed above.
131 -L, --list
133 Output additional information in <attribute_name>=<value> pairs,
134 one pair per line preceded by two spaces. This option has the
135 same effect as '-lll'.
136 -l, --long
138 Output additional information for each SCSI device (host). Can
139 be used multiple times for more output in which case the shorter
140 option form is more convenient (e.g. '-lll'). When used three
141 times (i.e. '-lll') outputs SCSI device (host) attributes one
142 per line; preceded by two spaces; in the form
143 "<attribute_name>=<value>".
144 -U, --long-unit
146 Output logical unit name in full, if available. It replaces the
147 normal vendor, product and revision strings given in the single
148 logical unit per line mode. If no logical unit name is found
149 "none" is printed. If the logical unit name is long (e.g. a
150 UUID) then following fields are pushed further to the right as
151 required. This option is functionally equivalent to the '-uuu'
152 option.
153 If the option is used twice (e.g. '-UU') then EUI, NAA, UUID and
154 T10 vendor ID formats are prefixed by "eui.", "naa.", "uuid."
155 and "t10." respectively. Note that SCSI name format used by
156 iSCSI should already be prefixed by 'iqn.'. Using the '--unit'
157 option 4 or more times (e.g. '-uuuu') will have the same action
158 as '-UU'.
159 --long_unit is also an acceptable form when invoking this
160 option.
161 -x, --lunhex
163 when this option is used once the LUN in the tuple (at the start
164 of each device line) is shown in "T10" format which is up to 16
165 hexadecimal digits. It is prefixed by "0x" to distinguish the
166 LUN from the decimal value shown in the absence of this option.
167 Also hierarchal LUNs are shown with a "_" character separating
168 the levels. For example the two level LUN: 0x0355006600000000
169 will appear as 0x0355_0066. If this option is given twice (e.g.
170 using the short form: '-xx') then the full 16 hexadecimal digits
171 are shown for each LUN, prefixed by "0x".
172 For NVMe, the namespace identifier (nsid) is shown in the "L"
173 position. The nsid is a 32 bit unsigned quantities with 0x0 and
174 0xffffffff reserved. Without this option, the nsid is shown in
175 decimal. When this option is used once the nsid is output in hex
176 with a lead 0x and with up to 3 leading zeros. When this option
177 is used twice the nsid is output in hex with up to 7 leading
178 zeros.
179 -N, --no-nvme
181 this option excludes NVMe devices and controllers for the out‐
182 put. This option may be needed to stop NVMe device output inter‐
183 fering with specific format output like that produced when the
184 --classic option is used.
185 To only show NVMe devices, use 'lsscsi N', to only show NVMe
186 controllers, use 'lsscsi -H N'.
187 -D, --pdt
189 this option displays the SCSI Peripheral Device Type (PDT) in
190 hex preceded by "0x". For NVME namespaces "0x0' is displayed
191 which corresponds to a disk ("Direct Access Device" or SSD). In
192 single line output this hex PDT replaces the device type abbre‐
193 viation (e.g. "0x0 " replaces "disk ") and appears after
194 the tuple.
195 -p, --protection
197 Output target (DIF) and initiator (DIX) protection types.
198 -P, --protmode
200 Output effective protection information mode for each disk
201 device.
202 -i, --scsi_id
204 outputs the udev derived matching id found in
205 /dev/disk/by-id/scsi* . This is only for disk (and disk like)
206 devices. If no match is found then "dm-uuid-mpath*" and "usb*"
207 are searched in the same directory. If there is still no match
208 then the /sys/class/block/<disk>/holders directory is searched.
209 The matching id is printed following the device name (e.g.
210 /dev/sdc) and if there is no match "-" is output.
211 -s, --size
213 Print disk capacity in human readable form. When given once,
214 normal base 10 SI units are used as a prefix for 'B' which is
215 bytes (aka octets). For example MB, GB and TB stand for 10^6,
216 10^9 and 10^12 bytes respectively. When given twice, IEC 80000-3
217 prefixes for 'B' are used; for example MiB, GiB and TiB stand
218 for 2^20, 2^30 and 2^40 bytes respectively. The output is
219 rounded to 3 or less significant figures in order to fit on a
220 single line. It will also output the size of RBC devices,
221 CD/DVD media and host managed ZBC disks. Host aware ZBC disks
222 have their "peripheral device type" set to 0 (the same as normal
223 disks) so their size is output.
224 If given three times (short form is the more convenient: '-sss')
225 then the disk capacity as a logical block count is given. This
226 is an exact figure in decimal reported by the storage device at
227 discovery. Discovery is typically just after boot time, or when
228 it was last attached if the storage device is removable.
229 To unclutter the single line per device mode the --brief option
230 combined with this option should help.
231 -y, --sysfsroot=PATH
233 assumes sysfs is mounted at PATH instead of the default '/sys' .
234 If this option is given PATH should be an absolute path (i.e.
235 start with '/').
236 -S, --sz-lbs
238 Print disk capacity as a number of logical blocks (which is the
239 same as '-sss'). When used twice a comma is added followed by
240 the logical block size in bytes. It should be a number like 512
241 or 4096.
242 If the logical block size cannot be found (e.g. because the ver‐
243 sion of Linux predates the /sys/block/<dev_name>/queue direc‐
244 tory) then the number of 512 byte blocks followed comma and then
245 '512' is output irrespective of what the true logical block size
246 of the device is. This special case action occurs whether this
247 option is given one or more times.
248 To unclutter the single line per device mode the --brief option
249 combined with this option should help.
250 -t, --transport
252 Output transport information. This will be target related infor‐
253 mation or, if --hosts is given, initiator related information.
254 When used without --list, a name or identifier (or both) are
255 output on a single line, usually prefixed by the type of trans‐
256 port. For devices this information replaces the normal vendor,
257 product and revision strings. When the --list option is also
258 given then additionally multiple lines of attribute_name=value
259 pairs are output, each indented by two spaces. See the section
260 on transports below.
261 -u, --unit
263 Output logical unit name, if available. If this option is given
264 once or twice, then the 30 character field where the vendor,
265 product and revision strings are usually placed is expanded to
266 32 characters and replaced by the logical unit name. If no logi‐
267 cal unit name is found "none" is printed. The first found of
268 the NAA, EUI-64 or SCSI name string is output unless a SCSI name
269 string is found and the associated target port indicates the
270 iSCSI protocol, in which case the SCSI name string is preferred.
271 Finally if there is no match on the above and a T10 Vendor ID
272 descriptor is found then it is used.
273 If the name cannot fit in the 32 character field then it is
274 truncated to the right and a trailing '_' character is used to
275 alert the reader to the truncation. The 32 character width is
276 chosen since that is large enough to hold 16 byte NAA or EUI-64
277 identifiers. However SCSI name strings as used by iSCSI can be
278 larger than that.
279 If this option is used twice then this field is also 32 charac‐
280 ter wide. If the logical unit name cannot fit then it will be
281 truncated to the left and a leading '_' character is used to
282 alert the reader to the truncation.
283 If this option is used three times the whole logical unit name
284 is output, followed by several spaces.
285 In order for this option to work, it needs a Linux kernel from
286 and including 3.15 . It accesses the sysfs vpd_pg83 file for the
287 device in question. Old SCSI and ATA (SATA) equipment may not
288 provide this information. If it is provided by ATA (SATA) then
289 it will be the WWN.
290 -v, --verbose
292 outputs directory names where information is found. Use multiple
293 times for more output.
294 -V, --version
296 outputs version information then exits. If used once outputs to
297 stderr; if used twice outputs to stdout and shortens the date to
298 yyyymmdd numeric format.
299 -w, --wwn
301 outputs the WWN for disks instead of manufacturer, model and
302 revision (or instead of transport information). The World Wide
303 Name (WWN) is typically 64 bits long (16 hex digits) but could
304 be up to 128 bits long. To indicate the WWN is hexadecimal, it
305 is prefixed by "0x". The ATA/SATA WWN is referred to as LU name
306 in SCSI jargon; hence this option is more or less superseded by
307 the --unit and --long-unit options.
308
310 This utility lists SCSI devices which are known as logical units (LU)
311 in the SCSI Architecture Model (ref: SAM-5 at http://www.t10.org) or
312 hosts when the --hosts option is given. A host is called an initiator
313 in SAM-5. A SCSI command travels out via an initiator, across some
314 transport to a target and then onwards to a logical unit. A target
315 device may contain several logical units. A target device has one or
316 more ports that can be viewed as transport end points. Each FC and SAS
317 disk is a single target that has two ports and contains one logical
318 unit. If both target ports on a FC or SAS disk are connected and visi‐
319 ble to a machine, then lsscsi will show two entries. Initiators (i.e.
320 hosts) also have one or more ports and some HBAs in Linux have a host
321 entry per initiator port while others have a host entry per initiator
322 device.
323 When the --transport option is given for devices (i.e. --hosts not
325 given) then most of the information produced by lsscsi is associated
326 with the target, or more precisely: the target port, through which SCSI
327 commands pass that access a logical unit.
328 Typically this utility provides one line of output per "device" or
330 host. Significantly more information can be obtained by adding the
331 --list option. When used together with the --transport option, after
332 the summary line, multiple lines of transport specific information in
333 the form "<attribute_name>=<value>" are output, each indented by two
334 spaces. Using a filter argument will reduce the volume of output if a
335 lot of devices or hosts are present.
336 The transports that are currently recognized are: IEEE 1394, ATA, FC,
338 iSCSI, SAS, SATA, SPI, SRP and USB.
339 For IEEE 1394 (a.k.a. Firewire and "SBP" when storage is involved), the
341 EUI-64 based target port name is output when --transport is given, in
342 the absence of the --hosts option. When the --hosts option is given
343 then the EUI-64 initiator port name is output. Output on the summary
344 line specific to the IEEE 1394 transport is prefixed by "sbp:".
345 To detect ATA and SATA devices a crude check is performed on the driver
347 name (after the checks for other transports are exhausted). Based on
348 the driver name either the ATA or SATA transport type is chosen. Output
349 on the summary line is either "ata:" or "sata:". A search is made for
350 an associated vpd_pg83 file in sysfs, if found it may contain the
351 device's WWN which is output if present. The WWN will not appear in
352 Linux kernels before 3.15 and with old PATA and SATA devices. Most
353 device and hosts flagged as "ata:" will use the parallel ATA transport
354 (PATA). For SATA devices that are attached via a SAS expander, see the
355 SAS paragraph below.
356 For Fibre Channel (FC) the port name and port identifier are output
358 when --transport is given. In the absence of the --hosts option these
359 ids will be for the target port associated with the device (logical
360 unit) being listed. When the --hosts option is given then the ids are
361 for the initiator port used by the host. Output on the summary line
362 specific to the FC transport is prefixed by "fc:". If FCoE (over Eth‐
363 ernet) is detected the prefix is changed to "fcoe:".
364 For iSCSI the target port name is output when --transport is given, in
366 the absence of the --hosts option. This is made up of the iSCSI name
367 and the target portal group tag. Since the iSCSI name starts with "iqn"
368 no further prefix is used. When the --hosts option is given then only
369 "iscsi:" is output on the summary line.
370 For Serial Attached SCSI the SAS address of the target port (or initia‐
372 tor port if --hosts option is also given) is output. This will be a
373 naa-5 address. For SAS HBAs and SAS targets (such as SAS disks and tape
374 drives) the SAS address will be world wide unique. For SATA disks
375 attached to a SAS expander, the expander provides the SAS address by
376 adding a non zero value to its (i.e. the expander's) SAS address (e.g.
377 expander_sas_address + phy_id + 1). SATA disks directly attached to SAS
378 HBAs seem to have an indeterminate SAS address. Output on the summary
379 line specific to the SAS transport is prefixed by "sas:".
380 For SATA devices, see the paragraph above on ATA devices. As noted in
382 the previous paragraph, SATA devices attached to SAS expanders will
383 display a manufactured SAS transport address (manufactured by the
384 expander) rather than the SATA device's WWN.
385 For the SCSI Parallel Interface (SPI) the target port identifier (usu‐
387 ally a number between 0 and 15 inclusive) is output when --transport is
388 given, in the absence of the --hosts option. When the --hosts option is
389 given then only "spi:" is output on the summary line.
390 For the PCIe transport (a.k.a. PCI Express) there at two possible stor‐
392 age types: NVMe and SOP/PQI (SCSI over PCIe). There are very few exam‐
393 ples of the latter currently so this utility concentrates on NVMe. NVMe
394 uses its own command set and not SCSI but has many things in common.
395 Rather than re-invent everything currently in use that SCSI has accumu‐
396 lated over nearly 40 years, NVMe is beginning to use some parts of
397 SCSI. A recent example is the SES-3 standard for enclosure management
398 which has been adopted by NVMe. In SCSI a SES device is a logical unit
399 with a peripheral device type (PDT) of 0xd (for enclosure) so it will
400 appear when the lsscsi utility is invoked without any options. In NVMe
401 is seems that an enclosure with appear as attached to the management
402 interface (MI) of a NVMe controller. This means it should appear when
403 "lsscsi --hosts" is invoked. It is unclear whether such a NVMe con‐
404 troller can have any storage namespaces associated with it. The sg_ses
405 utility (in the sg3_utils package) can then be given that NVMe con‐
406 troller's device name (e.g. /dev/nmve1).
407 When the --transport option is given, after "pcie" the NVMe con‐
408 troller's subsystem vendor id and device id are output, separated by a
409 colon (e.g. "pcie 0x8086:0x390a").
410 For the SCSI RDMA Protocol (SRP) the IB (InfiniBand) port's GUID is
412 given. As an example, it has a form like this: 0002:c903:00fa:abcd .
413 When a USB transport is detected, the summary line will contain "usb:"
415 followed by a USB device name. The USB device name has the form
416 "<b>-<p1>[.<p2>[.<p3>]]:<c>.<i>" where <b> is the USB bus number, <p1>
417 is the port on the host. <p2> is a port on a host connected hub, if
418 present. If needed <p3> is a USB hub port closer to the USB storage
419 device. <c> refers to the configuration number while <i> is the inter‐
420 face number. There is a separate SCSI host for each USB (SCSI) target.
421 A USB SCSI target may contain multiple logical units. Thus the same
422 "usb: <device_name>" string appears for a USB SCSI host and all logical
423 units that belong to the USB SCSI target associated with that USB SCSI
424 host.
425
427 For historical reasons and as used by several other Unix based Operat‐
428 ing Systems, Linux uses a tuple of integers to describe (a path to) a
429 SCSI device (also know as a Logical Unit (LU)). The last element of
430 that tuple is the so-called Logical Unit Number (LUN). And originally
431 in SCSI a LUN was an integer, at first 3 bits long, then 8 then 16
432 bits. SCSI LUNs today (SAM-5 section 4.7) are 64 bits but SCSI stan‐
433 dards now consider a LUN to be an array of 8 bytes.
434 Up until 2013, Linux mapped SCSI LUNs to a 32 bit integer by taking the
436 first 4 bytes of the SCSI LUN and ignoring the last 4 bytes. Linux
437 treated the first two bytes of the SCSI LUN as a unit (a word) and it
438 became the least significant 16 bits in the Linux LUN integer. The next
439 two bytes of the SCSI LUN became the upper 16 bits in the Linux LUN
440 integer. The rationale for this was to keep commonly used LUNs small
441 Linux LUN integers. The most common LUN (by far) in SCSI LUN (hex)
442 notation is 00 00 00 00 00 00 00 00 and this becomes the Linux LUN
443 integer 0. The next most common LUN is 00 01 00 00 00 00 00 00 and this
444 becomes the Linux LUN integer 1.
445 In 2013 it is proposed to increase Linux LUNs to a 64 bit integer by
447 extending the mapping outlined above. In this case all information that
448 is possible to represent in a SCSI LUN is mapped a Linux LUN (64 bit)
449 integer. And the mapping can be reversed without losing information.
450 This version of the utility supports both 32 and 64 bit Linux LUN inte‐
452 gers. By default the LUN shown at the end of the tuple commencing each
453 line is a Linux LUN as a decimal integer. When the --lunhex option is
454 given then the LUN is in SCSI LUN format with the 8 bytes run together,
455 with the output in hexadecimal and prefixed by '0x'. The LUN is decoded
456 according to SAM-5's description and trailing zeros (i.e. digits to the
457 right) are not shown. So LUN 0 (i.e. 00 00 00 00 00 00 00 00) is shown
458 as 0x0000 and LUN 65 (i.e. 00 41 00 00 00 00 00 00) is shown as 0x0041.
459 If the --lunhex option is given twice then the full 64 bits (i.e. 16
460 hexadecimal digits) are shown.
461 If the --lunhex option is not given on the command line then the envi‐
463 ronment variable LSSCSI_LUNHEX_OPT is checked. If LSSCSI_LUNHEX_OPT is
464 present then its associated value becomes the number of times the
465 --lunhex is set internally. So, for example, 'LSSCSI_LUNHEX_OPT=2 lss‐
466 csi' and 'lsscsi -xx' are equivalent.
467
469 Information about this utility including examples can also be found at:
470 http://sg.danny.cz/scsi/lsscsi.html .
471
473 Information for this command is derived from the sysfs file system,
474 which is assumed to be mounted at /sys unless specified otherwise by
475 the user. SCSI (pseudo) devices that have been detected by the SCSI
476 mid level will be listed even if the required upper level drivers (i.e.
477 sd, sr, st, osst or ch) have not been loaded. If the appropriate upper
478 level driver has not been loaded then the device file name will appear
479 as '-' rather than something like '/dev/st0'. Note that some devices
480 (e.g. scanners and medium changers) do not have a primary upper level
481 driver and can only be accessed via a SCSI generic (sg) device name.
482 Generic SCSI devices can also be accessed via the bsg driver in Linux.
484 By default, the bsg driver's device node names are of the form
485 '/dev/bsg/H:C:T:L'. So, for example, the SCSI device shown by this
486 utility on a line starting with the tuple '6:0:1:2' could be accessed
487 via the bsg driver with the '/dev/bsg/6:0:1:2' device node name.
488 lsscsi version 0.21 or later is required to correctly display SCSI
490 devices in Linux kernel 2.6.26 (and possibly later) when the CON‐
491 FIG_SYSFS_DEPRECATED_V2 kernel option is not defined.
492
494 Written by Doug Gilbert
495
497 Report bugs to <dgilbert at interlog dot com>.
498
500 Copyright © 2003-2020 Douglas Gilbert
501 This software is distributed under the GPL version 2. There is NO war‐
502 ranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PUR‐
503 POSE.
504
506 lspci lsusb lsblk
507lsscsi-0.31 February 2020 lsscsi(8)