/
githubmirror
/
sg3_utils
Обзор
Документация
Войти
/
githubmirror
/
sg3_utils
Код
Запросы
0
Пакеты
0
Релизы
0
Аналитика
Безопасность
main
sg3_utils.man8.html
1 182 строки
55 KB
Douglas Gilbert
post release 1.49, start at version/release 1.49.5
25 июл 2026, 18:00
25 июл 2026, 18:00
b7952b7
Код
Авторство
О чём код?
Content-type: text/html; charset=UTF-8 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <HTML><HEAD><TITLE>Man page of SG3_UTILS</TITLE> </HEAD><BODY> <H1>SG3_UTILS</H1> Section: SG3_UTILS (8)<BR>Updated: 2026-07-18<BR><A HREF="#index">Index</A> <A HREF="../index.html">Return to Main Contents</A><HR> <A NAME="lbAB"> </A> <H2>NAME</H2> sg3_utils - a package of utilities for sending SCSI commands <A NAME="lbAC"> </A> <H2>SYNOPSIS</H2> <B>sg_*</B> [<I>--dry-run</I>] [<I>--enumerate</I>] [<I>--help</I>] [<I>--hex</I>] [<I>--in=FN</I>] [<I>--inhex=FN</I>] [<I>--json[=JO]</I>] [<I>--js-file=JFN</I>] [<I>--maxlen=LEN</I>] [<I>--raw</I>] [<I>--timeout=SECS</I>] [<I>--verbose</I>] [<I>--version</I>] [<I>--</I>] [<I>OTHER_OPTIONS</I>] [<I>DEVICE</I>] <A NAME="lbAD"> </A> <H2>DESCRIPTION</H2> sg3_utils is a package of utilities that send SCSI commands to the given <I>DEVICE</I> via a SCSI pass through interface provided by the host operating system. <P> The names of all utilities start with "sg" and most start with "sg_" often followed by the name, or a shortening of the name, of the SCSI command that they send. For example the "sg_verify" utility sends the SCSI VERIFY command. A mapping between SCSI commands and the sg3_utils utilities that issue them is shown in the COVERAGE file. The sg_raw utility can be used to send an arbitrary SCSI command (supplied on the command line) to the given <I>DEVICE</I>. <P> sg_decode_sense can be used to decode SCSI sense data given on the command line or in a file. sg_raw -vvv will output the T10 name of a given SCSI CDB which is most often 16 bytes or less in length. <P> SCSI draft standards can be found at <A HREF="https://www.t10.org">https://www.t10.org</A> . The standards themselves can be purchased from ANSI and other standards organizations. A good overview of various SCSI standards can be seen in <A HREF="https://www.t10.org/scsi-3.htm">https://www.t10.org/scsi-3.htm</A> with the SCSI command sets in the upper part of the diagram. The highest level (i.e. most abstract) document is the SCSI Architecture Model (SAM) with SAM-5 being the most recent standard (ANSI INCITS 515-2016) with the most recent draft being SAM-6 revision 4 . SCSI commands in common with all device types can be found in SCSI Primary Commands (SPC) of which SPC-5 is the most recent standard (ANSI INCITS 502-2020). The most recent SPC draft is SPC-6 revision 6. Block device specific commands (e.g. as used by disks) are in SBC, those for tape drives in SSC, those for SCSI enclosures in SES and those for CD/DVD/BD drives in MMC. <P> It is becoming more common to control ATA disks with the SCSI command set. This involves the translation of SCSI commands to their corresponding ATA equivalents (and that is an imperfect mapping in some cases). The relevant standard is called SCSI to ATA Translation (SAT, SAT-2 and SAT-3) are now standards at INCITS(ANSI) and ISO while SAT-4 is at the draft stage. The logic to perform the command translation is often called a SAT Layer or SATL and may be within an operating system, in host bus adapter firmware or in an external device (e.g. associated with a SAS expander). See <A HREF="https://www.t10.org">https://www.t10.org</A> for more information. <P> There is some support for SCSI tape devices but not for their basic operation. The reader is referred to the "mt" utility. <P> There are two generations of command line option usage. The newer utilities (written since July 2004) use the getopt_long() function to parse command line options. With that function, each option has two representations: a short form (e.g. '-v') and a longer form (e.g. '--verbose'). If an argument is required then it follows a space (optionally) in the short form and a "=" in the longer form (e.g. in the sg_verify utility '-l 2a6h' and '--lba=2a6h' are equivalent). Note that with getopt_long(), short form options can be elided, for example: '-all' is equivalent to '-a -l -l'. The <I>DEVICE</I> argument may appear after, between or prior to any options. <P> The older utilities, including as sg_inq, sg_logs, sg_modes, sg_opcode, sg_rbuff, sg_readcap, sg_senddiag, sg_start and sg_turs had individual command line processing code typically based on a single "-" followed by one or more characters. If an argument is needed then it follows a "=" ( e.g. '-p=1f' in sg_modes with its older interface). Various options can be elided as long as it is not ambiguous (e.g. '-vv' to increase the verbosity). <P> Over time the command line interface of these older utilities became messy and overloaded with options. So in sg3_utils version 1.23 the command line interface of these older utilities was altered to have both a cleaner getopt_long() interface and their older interface for backward compatibility. By default these older utilities use their getopt_long() based interface. The getopt_long() is a GNU extension (i.e. not yet POSIX certified) but more recent command line utilities tend to use it. That can be overridden by defining the SG3_UTILS_OLD_OPTS environment variable or using '-O' or '--old' as the first command line option. The man pages of the older utilities documents the details. <P> Several sg3_utils utilities are based on the Unix dd command (e.g. sg_dd) and permit copying data at the level of SCSI READ and WRITE commands. sg_dd is tightly bound to Linux and hence is not ported to other OSes. A more generic utility (than sg_dd) called ddpt in a package of the same name has been ported to other OSes. <A NAME="lbAE"> </A> <H2>COMMAND DESCRIPTOR BLOCK</H2> This term is often used when describing SCSI commands, usually abbreviated to CDB. SCSI commands such as READ(16) refer to the 16 byte CDB used to fetch data from a storage device. Typical CDB lengths are 6, 10, 12, 16 and 32 bytes long. Any integers in a CDB that are longer than 8 bits (i.e. 1 byte) use a big endian repesentation. <A NAME="lbAF"> </A> <H2>ENVIRONMENT VARIABLES</H2> The SG3_UTILS_OLD_OPTS environment variable is explained in the previous section. It is only for backward compatibility of the command line options for older utilities. <P> The SG3_UTILS_INVOCATION environment variable has been implemented on some utilities that decode a lot of data. When SG3_UTILS_INVOCATION is set those utilities will print out a line of 60 "v" characters. After that the utility name and version (which includes a date) followed by a line by line breakdown of the command line arguments that the utility was invoked with. The command line arguments include the utility name and options in the order they were given. The output is sent to stdout. The option is designed to help automatic testing of utilities in the package. <P> The SG3_UTILS_DSENSE environment variable may be set to a number. It is only used by the embedded SNTL within the library used by the utilities in this library. SNTL is a SCSI to NVMe Translation Layer. This environment variable defaults to 0 which will lead to any utility that issues a SCSI command that is translated to a NVMe command (by the embedded SNTL) that fails at the NVMe device, to return SCSI sense in 'fixed' format. If this variable is non-zero then then the returned SCSI sense will be in 'descriptor' format. <P> Several utilities have their own environment variable setting (e.g. sg_persist has SG_PERSIST_IN_RDONLY). See individual utility man pages for more information. <P> There is a Linux specific environment variable called SG3_UTILS_LINUX_NANO that if defined and the sg driver in the system is 4.0.30 or later, will show command durations in nanoseconds rather than the default milliseconds. Command durations are typically only shown if --verbose is used 3 or more times. Due to an interface problem (a 32 bit integer that should be 64 bits with the benefit of hindsight) the maximum duration that can be represented in nanoseconds is about 4.2 seconds. If longer durations may occur then don't define this environment variable (or undefine it). <A NAME="lbAG"> </A> <H2>LINUX DEVICE NAMING</H2> Most disk block devices have names like /dev/sda, /dev/sdb, /dev/sdc, etc. SCSI disks in Linux have always had names like that but in recent Linux kernels it has become more common for many other disks (including SATA disks and USB storage devices) to be named like that. Partitions within a disk are specified by a number appended to the device name, starting at 1 (e.g. /dev/sda1 ). <P> Tape drives are named /dev/st<num> or /dev/nst<num> where <num> starts at zero. Additionally one letter from this list: "lma" may be appended to the name. CD, DVD and BD readers (and writers) are named /dev/sr<num> where <num> start at zero. There are less used SCSI device type names, the dmesg and the lsscsi commands may help to find if any are attached to a running system. <P> There is also a SCSI device driver which offers alternate generic access to SCSI devices. It uses names of the form /dev/sg<num> where <num> starts at zero. The "lsscsi -g" command may be useful in finding these and which generic name corresponds to a device type name (e.g. /dev/sg2 may correspond to /dev/sda). In the lk 2.6 series a block SCSI generic driver was introduced and its names are of the form /dev/bsg/<h:c:t:l> where h, c, t and l are numbers. Again see the lsscsi command to find the correspondence between that SCSI tuple (i.e. <h:c:t:l>) and alternate device names. <P> Prior to the Linux kernel 2.6 series these utilities could only use generic device names (e.g. /dev/sg1 ). In almost all cases in the Linux kernel 2.6 series, any device name can be used by these utilities. <P> Very little has changed in Linux device naming in the Linux kernel 3 and 4 series. <A NAME="lbAH"> </A> <H2>WINDOWS DEVICE NAMING</H2> Storage and related devices can have several device names in Windows. Probably the most common in the volume name (e.g. "D:"). There are also a "class" device names such as "PhysicalDrive<n>", "CDROM<n>" and "TAPE<n>". <n> is an integer starting at 0 allocated in ascending order as devices are discovered (and sometimes rediscovered). <P> Some storage devices have a SCSI lower level device name which starts with a SCSI (pseudo) adapter name of the form "SCSI<n>:". To this is added sub-addressing in the form of a "bus" number, a "target" identifier and a LUN (Logical Unit Number). The "bus" number is also known as a "PathId". These are assembled to form a device name of the form: "SCSI<n>:<bus>,<target>,<lun>". The trailing ",<lun>" may be omitted in which case a LUN of zero is assumed. This lower level device name cannot often be used directly since Windows blocks attempts to use it if a class driver has "claimed" the device. There are SCSI device types (e.g. Automation/Drive interface type) for which there is no class driver. At least two transports ("bus types" in Windows jargon): USB and IEEE 1394 do not have a "scsi" device names of this form. <P> In keeping with DOS file system conventions, the various device names can be given in upper, lower or mixed case. Since "PhysicalDrive<n>" is tedious to write, a shortened form of "PD<n>" is permitted by all utilities in this package. <P> A single device (e.g. a disk) can have many device names. For example: "PD0" can also be "C:", "D:" and "SCSI0:0,1,0". The two volume names reflect that the disk has two partitions on it. Disk partitions that are not recognized by Windows are not usually given a volume name. However Vista does show a volume name for a disk which has no partitions recognized by it and when selected invites the user to format it (which may be rather unfriendly to other OSes). <P> These utilities assume a given device name is in the Win32 device namespace. To make that explicit "\\.\" can be prepended to the device names mentioned in this section. Beware that backslash is an escape character in Unix like shells and the C programming language. In a shell like Msys (from MinGW) each backslash may need to be typed twice. <P> The sg_scan utility within this package lists out Windows device names in a form that is suitable for other utilities in this package to use. <A NAME="lbAI"> </A> <H2>FREEBSD DEVICE NAMING</H2> SCSI disks have block names of the form /dev/da<num> where <num> is an integer starting at zero. The "da" is replaced by "sa" for SCSI tape drives and "cd" for SCSI CD/DVD/BD drives. Each SCSI device has a corresponding pass-through device name of the form /dev/pass<num> where <num> is an integer starting at zero. The "camcontrol devlist" command may be useful for finding out which SCSI device names are available and the correspondence between class and pass-through names. <P> FreeBSD allows device names to be given without the leading "/dev/" (e.g. da0 instead of /dev/da0). That worked in this package up until version 1.43 when the unadorned device name (e.g. "da0") gave an error. The original action (i.e. allowing unadorned device names) has been restored in version 1.46 . Also note that symlinks (to device names) are followed before prepending "/dev/" if the resultant name doesn't start with a "/". <P> FreeBSD's NVMe naming has been evolving. The controller naming is the same as Linux: "/dev/nvme<n>" but the namespaces have an extra "s" (e.g. "/dev/nvme0ns1"). The latter is not a block (GEOM) device (strictly speaking FreeBSD does not have block devices). Initially FreeBSD had "/dev/nvd<m>" GEOM devices that were not based on the CAM subsystem. Then in FreeBSD release 12 a new nda driver was added that is CAM (and GEOM) based for NVMe namespaces; it has names like "/dev/nda0". The preferred device nodes for this package are "/dev/nvme0" for NVMe controllers and "/dev/nda0" for NVMe namespaces. <A NAME="lbAJ"> </A> <H2>SOLARIS DEVICE NAMING</H2> SCSI device names below the /dev directory have a form like: c5t4d3s2 where the number following "c" is the controller (HBA) number, the number following "t" is the target number (from the SCSI parallel interface days) and the number following "d" is the LUN. Following the "s" is the slice number which is related to a partition and by convention "s2" is the whole disk. <P> OpenSolaris also has a c5t4d3p2 form where the number following the "p" is the partition number apart from "p0" which is the whole disk. So a whole disk may be referred to as either c5t4d3, c5t4d3s2 or c5t4d3p0 . <P> And these device names are duplicated in the /dev/dsk and /dev/rdsk directories. The former is the block device name and the latter is for "raw" (or char device) access which is what sg3_utils needs. So in OpenSolaris something of the form 'sg_inq /dev/rdsk/c5t4d3p0' should work. If it doesn't work then add a '-vvv' option for more debug information. Trying this form 'sg_inq /dev/dsk/c5t4d3p0' (note "rdsk" changed to "dsk") will result in an "inappropriate ioctl for device" error. <P> The device names within the /dev directory are typically symbolic links to much longer topological names in the /device directory. In Solaris cd/dvd/bd drives seem to be treated as disks and so are found in the /dev/rdsk directory. Tape drives appear in the /dev/rmt directory. <P> There is also a sgen (SCSI generic) driver which by default does not attach to any device. See the /kernel/drv/sgen.conf file to control what is attached. Any attached device will have a device name of the form /dev/scsi/c5t4d3 . <P> Listing available SCSI devices in Solaris seems to be a challenge. "Use the 'format' command" advice works but seems a very dangerous way to list devices. [It does prompt again before doing any damage.] 'devfsadm -Cv' cleans out the clutter in the /dev/rdsk directory, only leaving what is "live". The "cfgadm -v" command looks promising. <A NAME="lbAK"> </A> <H2>NVME SUPPORT</H2> NVMe (or NVM Express) is a relatively new storage transport and command set. The level of abstraction of the NVMe command set is somewhat lower the SCSI command sets, closer to the level of abstraction of ATA (and SATA) command sets. NVMe claims to be designed with flash and modern "solid state" storage in mind, something unheard of when SCSI was originally developed in the 1980s. <P> The SCSI command sets' advantage is the length of time they have been in place and the existing tools (like these) to support it. Plus SCSI command sets level of abstraction is both and advantage and disadvantage. Recently the NVME-MI (Management Interface) designers decide to use the SCSI Enclosure Services (SES-3) standard "as is" with the addition of two tunnelling NVME-MI commands: SES Send and SES Receive. This means after the OS interface differences are taken into account, the sg_ses, sg_ses_microcode and sg_senddiag utilities can be used on a NVMe device that supports a newer version of NVME-MI. <P> The NVME-MI SES Send and SES Receive commands correspond to the SCSI SEND DIAGNOSTIC and RECEIVE DIAGNOSTIC RESULTS commands respectively. There are however a few other commands that need to be translated, the most important of which is the SCSI INQUIRY command to the NVMe Identify controller/namespace. Starting in version 1.43 these utilities contain a small SNTL (SCSI to NVMe Translation Layer) to take care of these details. <P> As a side effect of this "juggling" if the sg_inq utility is used (without the --page= option) on a NVMe <I>DEVICE</I> then the actual NVMe Identifier (controller and possibly namespace) responses are decoded and output. However if 'sg_inq --page=sinq <device>' is given for the same <I>DEVICE</I> then parts of the NVMe Identify controller and namespace response are translated to a SCSI standard INQUIRY response which is then decoded and output. <P> Apart from the special case with the sg_inq, all other utilities in the package assume they are talking to a SCSI device and decode any response accordingly. One easy way for users to see the underlying device is a NVMe device is the standard INQUIRY response Vendor Identification field of "NVMe " (an 8 character long string with 4 spaces to the right). <P> The following SCSI commands are currently supported by the SNTL library: INQUIRY, MODE SELECT(10), MODE SENSE(10), READ(10,16), READ CAPACITY(10,16), RECEIVE DIAGNOSTIC RESULTS, REQUEST SENSE, REPORT LUNS, REPORT SUPPORTED OPERATION CODES, REPORT SUPPORTED TASK MANAGEMENT FUNCTIONS, SEND DIAGNOSTICS, START STOP UNIT, SYNCHRONIZE CACHE(10,16), TEST UNIT READY, VERIFY(10,16), WRITE(10,16) and WRITE SAME(10,16). <A NAME="lbAL"> </A> <H2>EXIT STATUS</H2> To aid scripts that call these utilities, the exit status is set to indicate success (0) or failure (1 or more). Note that some of the lower values correspond to the SCSI sense key values. <P> The exit status values listed below can be given to the sg_decode_sense utility (which is found in this package) as follows: <P> <BR> sg_decode_sense --err=<exit_status> <P> and a short explanatory string will be output to stdout. <P> The exit status values are: <DL COMPACT> <DT><B>0</B> <DD> success. Also used for some utilities that wish to return a boolean value for the "true" case (and that no error has occurred). The false case is conveyed by exit status 36. <DT><B>1</B> <DD> syntax error. Either illegal command line options, options with bad arguments or a combination of options that is not permitted. <DT><B>2</B> <DD> the <I>DEVICE</I> reports that it is not ready for the operation requested. The <I>DEVICE</I> may be in the process of becoming ready (e.g. spinning up but not at speed) so the utility may work after a wait. In Linux the <I>DEVICE</I> may be temporarily blocked while error recovery is taking place. <BR> See exit status values 12, 13 and 16 below which refine this exit value. <DT><B>3</B> <DD> the <I>DEVICE</I> reports a medium or hardware error (or a blank check). For example an attempt to read a corrupted block on a disk will yield this value. <DT><B>5</B> <DD> the <I>DEVICE</I> reports an "illegal request" with an additional sense code other than "invalid command operation code". This is often a supported command with a field set requesting an unsupported capability. For commands that require a "service action" field this value can indicate that the command with that service action value is not supported. <DT><B>6</B> <DD> the <I>DEVICE</I> reports a "unit attention" condition. This usually indicates that something unrelated to the requested command has occurred (e.g. a device reset) potentially before the current SCSI command was sent. The requested command has not been executed by the device. Note that unit attention conditions are usually only reported once by a device. <DT><B>7</B> <DD> the <I>DEVICE</I> reports a "data protect" sense key. This implies some mechanism has blocked writes (or possibly all access to the media). <DT><B>9</B> <DD> the <I>DEVICE</I> reports an illegal request with an additional sense code of "invalid command operation code" which means that it doesn't support the requested command. <DT><B>10</B> <DD> the <I>DEVICE</I> reports a "copy aborted". This implies another command or device problem has stopped a copy operation. The EXTENDED COPY family of commands (including WRITE USING TOKEN) may return this sense key. <DT><B>11</B> <DD> the <I>DEVICE</I> reports an aborted command. In some cases aborted commands can be retried immediately (e.g. if the transport aborted the command due to congestion). <DT><B>12</B> <DD> the <I>DEVICE</I> reports a sense key of not ready together with an additional sense code of "target port in standby state". <DT><B>13</B> <DD> the <I>DEVICE</I> reports a sense key of not ready together with an additional sense code of "target port in unavailable state". <DT><B>14</B> <DD> the <I>DEVICE</I> reports a miscompare sense key. VERIFY and COMPARE AND WRITE commands may report this. <DT><B>15</B> <DD> the utility is unable to open, close or use the given <I>DEVICE</I> or some other file. The given file name could be incorrect or there may be permission problems. Adding the '-v' option may give more information. <DT><B>16</B> <DD> the <I>DEVICE</I> reports a NOT READY sense code with an additional sense code of 0x4 with a qualifier value whose text contains "in progress". These include: Format in progress, Rebuild in progress, Sanitize in progress and Depopulation in progress. An exception is [0x2, 0x4, 0x1a] which is: LU not ready, Start stop unit command in progress. This is based on the assumption that SSU should complete in 30 seconds or less. Most of the other "in progress" messages could flag a wait of tens of hours. In most cases 'sg_requests --progress <I>DEVICE</I>' should quantify just how long the wait will be. <DT><B>17</B> <DD> a SCSI "Illegal request" sense code received with a flag indicating the Info field is valid. This is often a LBA but its meaning is command specific. <DT><B>18</B> <DD> the <I>DEVICE</I> reports a medium or hardware error (or a blank check) with a flag indicating the Info field is valid. This is often a LBA (of the first encountered error) but its meaning is command specific. <DT><B>19</B> <DD> the <I>DEVICE</I> reports an illegal request with an additional sense code of "invalid field in parameter list". The parameter list is extra data sent to the <I>DEVICE</I> for some commands (e.g. FORMAT UNIT) <DT><B>20</B> <DD> the <I>DEVICE</I> reports it has a check condition but "no sense" and non-zero information in its additional sense codes. Some polling commands (e.g. REQUEST SENSE) can receive this response. There may be useful information in the sense data such as a progress indication. <DT><B>21</B> <DD> the <I>DEVICE</I> reports a "recovered error". The requested command was successful. Most likely a utility will report a recovered error to stderr and continue, probably leaving the utility with an exit status of 0 . <DT><B>22</B> <DD> the <I>DEVICE</I> reports that the current command or its parameters imply a logical block address (LBA) that is out of range. This happens surprisingly often when trying to access the last block on a storage device; either a classic "off by one" logic error or a misreading of the response from READ CAPACITY(10 or 16) in which the address of the last block rather than the number of blocks on the <I>DEVICE</I> is returned. Since LBAs are origin zero they range from 0 to n-1 where n is the number of blocks on the <I>DEVICE</I>, so the LBA of the last block is one less than the total number of blocks. <DT><B>24</B> <DD> the <I>DEVICE</I> reports a SCSI status of "reservation conflict". This means access to the <I>DEVICE</I> with the current command has been blocked because another machine (HBA or SCSI "initiator") holds a reservation on this <I>DEVICE</I>. On modern SCSI systems this is related to the use of the PERSISTENT RESERVATION family of commands. <DT><B>25</B> <DD> the <I>DEVICE</I> reports a SCSI status of "condition met". Currently only the PRE-FETCH command (see SBC-4) yields this status. <DT><B>26</B> <DD> the <I>DEVICE</I> reports a SCSI status of "busy". SAM-6 defines this status as the logical unit is temporarily unable to process a command. It is recommended to re-issue the command. <DT><B>27</B> <DD> the <I>DEVICE</I> reports a SCSI status of "task set full". <DT><B>28</B> <DD> the <I>DEVICE</I> reports a SCSI status of "ACA active". ACA is "auto contingent allegiance" and is seldom used. <DT><B>29</B> <DD> the <I>DEVICE</I> reports a SCSI status of "task aborted". SAM-5 says: "This status shall be returned if a command is aborted by a command or task management function on another I_T nexus and the Control mode page TAS bit is set to one". <DT><B>31</B> <DD> error involving two or more command line options. They may be contradicting, select an unsupported mode, or a required option (given the context) is missing. <DT><B>32</B> <DD> there is a logic error in the utility. It corresponds to code comments like "shouldn't/can't get here". Perhaps the author should be informed. <DT><B>33</B> <DD> the command sent to <I>DEVICE</I> has timed out. <DT><B>34</B> <DD> this is a Windows only exit status and indicates that the Windows error number (32 bits) cannot meaningfully be mapped to an equivalent Unix error number returned as the exit status (7 bits). <DT><B>35</B> <DD> a transport error has occurred. This will either be in the driver (e.g. HBA driver) or in the interconnect between the host (initiator) and the device (target). For example in SAS an expander can run out of paths and thus be unable to return the user data from a READ command. <DT><B>36</B> <DD> no error has occurred plus the utility wants to convey a boolean value of false. The corresponding true value is conveyed by a 0 exit status. <DT><B>40</B> <DD> the command sent to <I>DEVICE</I> has received an "aborted command" sense key with an additional sense code of 0x10. This value is related to problems with protection information (PI or DIF). For example this error may occur when reading a block on a drive that has never been written (or is unmapped) if that drive was formatted with type 1, 2 or 3 protection. <DT><B>41</B> <DD> the command sent to <I>DEVICE</I> has received an "aborted command" sense key with an additional sense code of 0x10 (as with error code) plus a flag indicating the Info field is valid. <DT><B>47</B> <DD> flock (Unix system call) error. <DT><B>48</B> <DD> this is an internal message indicating a NVMe status field (SF) is other than zero after a command has been executed (i.e. something went wrong). Work in this area is currently experimental. <DT><B>49</B> <DD> low level driver reports a response's residual count (i.e. number of bytes actually received by HBA is 'requested_bytes - residual_count') that is nonsensical. <DT><B>50</B> <DD> OS system calls that fail often return a small integer number to help. In Unix these are called "errno" values where 0 implies no error. These error codes set aside 51 to 96 for mapping these errno values but that may not be sufficient. Higher errno values that cannot be mapped are all mapped to this value (i.e. 50). <BR> Note that an errno value of 0 is mapped to error code 0. <DT><B>50 + <os_error_number></B> <DD> OS system calls that fail often return a small integer number to help indicate what the error is. For example in Unix the inability of a system call to allocate memory returns (in 'errno') ENOMEM which often is associated with the integer 12. So 62 (i.e. '50 + 12') may be returned by a utility in this case. It is also possible that a utility in this package reports 50+ENOMEM when it can't allocate memory, not necessarily from an OS system call. In recent versions of Linux the file showing the mapping between symbolic constants (e.g. ENOMEM) and the corresponding integer is in the kernel source code file: include/uapi/asm-generic/errno-base.h <BR> Note that errno values that are greater than or equal to 47 cannot fit in range provided. Instead they are all mapped to 50 as discussed in the previous entry. <DT><B>97</B> <DD> a SCSI command response failed sanity checks. <DT><B>98</B> <DD> the <I>DEVICE</I> reports it has a check condition but the error doesn't fit into any of the above categories. <DT><B>99</B> <DD> any errors that can't be categorized into values 1 to 98 may yield this value. This includes transport and operating system errors after the command has been sent to the device. <DT><B>100-125</B> <DD> these error codes are used by the ddpt utility which uses the sg3_utils library. They are mainly specialized error codes associated with offloaded copies. <DT><B>126</B> <DD> the utility was found but could not be executed. That might occur if the executable does not have execute permissions. <DT><B>127</B> <DD> This is the exit status for utility not found. That might occur when a script calls a utility in this package but the PATH environment variable has not been properly set up, so the script cannot find the executable. <DT><B>128 + <signum></B> <DD> If a signal kills a utility then the exit status is 128 plus the signal number. For example if a segmentation fault occurs then a utility is typically killed by SIGSEGV which according to 'man 7 signal' has an associated signal number of 11; so the exit status will be 139 . <DT><B>255</B> <DD> the utility tried to yield an exit status of 255 or larger. That should not happen; given here for completeness. </DL> <P> Most of the error conditions reported above will be repeatable (an example of one that is not is "unit attention") so the utility can be run again with the '-v' option (or several) to obtain more information. <A NAME="lbAM"> </A> <H2>COMMON OPTIONS</H2> Arguments to long options are mandatory for short options as well. In the short form an argument to an option uses zero or more spaces as a separator (i.e. the short form does not use "=" as a separator). <P> If an option takes a numeric argument then that argument is assumed to be decimal unless otherwise indicated (e.g. with a leading "0x", a trailing "h" or as noted in the usage message). <P> Some options are used uniformly in most of the utilities in this package. Those options are listed below. Note that there are some exceptions. <DL COMPACT> <DT><B>-d</B>, <B>--dry-run</B><DD> utilities that can cause lots of user data to be lost or overwritten sometimes have a <I>--dry-run</I> option. Device modifying actions are typically bypassed (or skipped) to implement a policy of "do no harm". This allows complex command line invocations to be tested before the action required (e.g. format a disk) is performed. The <I>--dry-run</I> option has become a common feature of many command line utilities (e.g. the Unix 'patch' command), not just those from this package. <BR> Note that most hyphenated option names in this package also can be given with an underscore rather than a hyphen (e.g. <I>--dry_run</I>). <DT><B>-e</B>, <B>--enumerate</B><DD> some utilities (e.g. sg_ses and sg_vpd) store a lot of information in internal tables. This option will output that information in some readable form (e.g. sorted by an acronym or by page number) then exit. Note that with this option <I>DEVICE</I> is ignored (as are most other options) and no SCSI IO takes place, so the invoker does not need any elevated permissions. <DT><B>-h</B>, <B>-?</B>, <B>--help</B><DD> output the usage message then exit. In a few older utilities the '-h' option requests hexadecimal output. In these cases the '-?' option will output the usage message then exit. <DT><B>-H</B>, <B>--hex</B><DD> for SCSI commands that yield a non-trivial response, print out that response as a series ASCII hexadecimal bytes. When used once, 16 bytes are printed on each line, prefixed by a relative address, starting at 0 (hex). When used twice, an ASCII rendering of the 16 bytes is appended to each line, with non-printable characters replaced by a '.' . When used three times only the 16 hex bytes are printed on each line (hence no address prefix nor ASCII appended). To produce hexadecimal that can be parsed by other utilities use this option three, four or five times. <BR> The use of this option, especially when used multiple times has evolved over time as it has become more useful (at least in testing and fixing bugs). So check this option's description in the manpage of the utility being used. An example of a corner case with the '-HHH' option is with the sg_inq utility and the ATA Information VPD page. The sg_inq utility does not decode the ATA IDENTIFY DEVICE command response but instead outputs hexadecimal 16 bit words (little endian) suitable for the hdparm utility to decode. <DT><B>-i</B>, <B>--in</B>=<I>FN</I><DD> many SCSI commands fetch a significant amount of data (returned in the data-in buffer) which several of these utilities decode (e.g. sg_vpd and sg_logs). To separate the two steps of fetching the data from a SCSI device and then decoding it, this option has been added. The first step (fetching the data) can be done using the <I>--hex</I> or <I>--raw</I> option and redirecting the command line output to a file (often done with ">" in Unix based operating systems). The difference between <I>--hex</I> and <I>--raw</I> is that the former produces output in ASCII hexadecimal while <I>--raw</I> produces its output in "raw" binary. <BR> The second step (i.e. decoding the SCSI response data now held in a file) can be done using this <I>--in=FN</I> option where the file name is <I>FN</I>. If "-" is used for <I>FN</I> then stdin is assumed, again this allows for command line redirection (or piping). That file (or stdin) is assumed to contain ASCII hexadecimal unless the <I>--raw</I> option is also given in which case it is assumed to be binary. Notice that the meaning of the <I>--raw</I> option is "flipped" when used with <I>--in=FN</I> to act on the input, typically it acts on the output data. <BR> Since the structure of the data returned by SCSI commands varies considerably then the usage information or the manpage of the utility being used should be checked. In some cases <I>--hex</I> may need to be used multiple times (and is more conveniently given as '-HH' or '-HHH). <DT><B>-i</B>, <B>--inhex</B>=<I>FN</I><DD> This option has the same or similar functionality as <I>--in=FN</I>. And perhaps 'inhex' is more descriptive since by default, ASCII hexadecimal is expected in the contents of file: <I>FN</I>. Alternatively the short form option may be <I>-I</I> or <I>-X</I>. See the "HEX, BINARY AND JSON FORMATS" section below for more information. <DT><B>--json</B>[=<I>JO</I>]<DD> The default output of most utilities that decode information returned from SCSI devices is designed for human readability. Sometimes a more parseable form of output is required and JSON is a popular way to do this. Only utilities that decode a significant amount of SCSI data support this option. <BR> The corresponding short option is usually <I>-j[JO]</I> but maybe <I>-J[JO]</I> if <I>-j</I> is already in use. Note that in all cases <I>JO</I> argument is itself optional. See the sg3_utils_json manpage for more information. <DT><B>-J</B>, <B>--js-file</B>=<I>JFN</I><DD> output is in JSON format and it is sent to a file named <I>JFN</I>. If that file exists then it is truncated. By default, the JSON output is sent to stdout. <BR> When this option is given, the <I>--json[=JO]</I> option is implied and need not be given. The <I>--json[=JO]</I> option may still be needed to set the <I>JO</I> parameter to non-default values. <DT><B>-m</B>, <B>--maxlen</B>=<I>LEN</I><DD> several important SCSI commands (e.g. INQUIRY and MODE SENSE) have response lengths that vary depending on many factors, only some of which these utilities take into account. The maximum response length is typically specified in the 'allocation length' field of the cdb. In the absence of this option, several utilities use a default allocation length (sometimes recommended in the SCSI draft standards) or a "double fetch" strategy. See <A HREF="../man8/sg_logs.8.html">sg_logs</A>(8) for its description of a "double fetch" strategy. These techniques are imperfect and in the presence of faulty SCSI targets can cause problems (e.g. some USB mass storage devices freeze if they receive an INQUIRY allocation length other than 36). Also use of this option disables any "double fetch" strategy that may have otherwise been used. <BR> To head off a class of degenerate bugs, if <I>LEN</I> is less than 16 then it is ignored (usually with a warning message) and the default value is used instead. Some utilities use 4 (bytes), rather than 16, as the cutoff value. <DT><B>-r</B>, <B>--raw</B><DD> many SCSI commands yield 'data-in' (i.e. data moving from the LU (Logical Unit) toward the initiator (e.g. HBA)). Examples are the SCSI INQUIRY and READ commands. This option causes any decoding of the data-in to be skipped and instead the data is sent, in binary, to stdout. If any error messages or warning are produced they are usually sent to stderr so as to not interfere with the output from this option. <DT><DD> Alternatively the <I>--in=FN</I> option causes <I>DEVICE</I> to be ignored and the response data (to be decoded) fetched from a file named <I>FN</I>. In these cases this option may indicate that binary data can be read from stdin or from a nominated file (e.g. <I>FN</I>). <DT><B>-t</B>, <B>--timeout</B>=<I>SECS</I><DD> utilities that issue potentially long-running SCSI commands often have a <I>--timeout=SECS</I> option. This typically instructs the operating system to abort the SCSI command in question once the timeout expires. Aborting SCSI commands is typically a messy business and in the case of format like commands may leave the device in a "format corrupt" state requiring another long-running re-initialization command to be sent. The argument, <I>SECS</I>, is usually in seconds and the short form of the option may be something other than <I>-t</I> since the timeout option was typically added later as storage devices grew in size and initialization commands took longer. Since many utilities had relatively long internal command timeouts before this option was introduced, the actual command timeout given to the operating systems is the higher of the internal timeout and <I>SECS</I>. <BR> Many long running SCSI commands have an IMMED bit in the CDB which causes the command to return relatively quickly but the requested operation has only been commenced, not completed. In such cases the REQUEST SENSE command can be used to monitor progress via its progress indication field (see the sg_requests and sg_turs utilities). Utilities that send such SCSI commands either have an <I>--immed</I> option or a <I>--wait</I> option which is the logical inverse of the "immediate" action. Sending a SCSI data access command to a device which is busy performing a long running SCSI command (in background) will typically receive a NOT READY sense key with an additional sense code of 0x04 and a qualifier code which indicates which long running operation the device is performing. <BR> An alternate long option form is <I>--tmo=SECS</I>. <DT><B>-v</B>, <B>--verbose</B><DD> increase the level of verbosity, (i.e. debug output). Can be used multiple times to further increase verbosity. The additional output caused by this option is almost always sent to stderr. <DT><B>-V</B>, <B>--version</B><DD> print the version string and then exit. Each utility has its own version number and date of last code change. <DT><B>--</B><DD> in common with many Unix commands, these utilities will accept the '--' option to indicate that everything following that point on the command line is an argument and <B>not</B> an option. This becomes important if and when a filename starts with a '-' character. Doing that is not a good idea but fingers slip and bad things happens. </DL> <A NAME="lbAN"> </A> <H2>NUMERIC ARGUMENTS</H2> Many utilities have command line options that take numeric arguments. These numeric arguments can be large values (e.g. a logical block address (LBA) on a disk) and can be inconvenient to enter in the default decimal representation. So various other representations are permitted. <P> Multiplicative suffixes are accepted. They are one, two or three letter strings appended directly after the number to which they apply: <P> <BR> c C *1 <BR> w W *2 <BR> b B *512 <BR> k K KiB *1024 <BR> KB kB *1000 <BR> m M MiB *1048576 <BR> MB mB *1000000 <BR> g G GiB *(2^30) <BR> GB gB *(10^9) <BR> t T TiB *(2^40) <BR> TB *(10^12) <BR> p P PiB *(2^50) <BR> PB *(10^15) <P> An example is "2k" for 2048. The large tera and peta suffixes are only available for numeric arguments that might require 64 bits to represent internally. <P> These multiplicative suffixes are compatible with GNU's dd command (since 2002) which claims compliance with SI and with IEC 60027-2. <P> A suffix of the form "x<n>" multiplies the preceding number by <n>. An example is "2x33" for "66". The left argument cannot be '0' as '0x' will be interpreted as hexadecimal number prefix (see below). The left argument to the multiplication must end in a hexadecimal digit (i.e. 0 to f) and the whole expression cannot have any embedded whitespace (e.g. spaces). An ugly example: "0xfx0x2" for 30. <P> A suffix of the form "+<n>" adds the preceding number to <n>. An example is "3+1k" for "1027". The left argument to the addition must end in a hexadecimal digit (i.e. 0 to f) and the whole expression cannot have any embedded whitespace (e.g. spaces). Another example: "0xf+0x2" for 17. <P> Alternatively numerical arguments can be given in hexadecimal. There are two syntaxes. The number can be preceded by either "0x" or "0X" as found in the C programming language. The second hexadecimal representation is a trailing "h" or "H" as found in (storage) standards. When hex numbers are given, multipliers cannot be used. For example the decimal value "256" can be given as "0x100" or "100h". <A NAME="lbAO"> </A> <H2>HEX, BINARY AND JSON FORMATS</H2> Both hexadecimal (ASCII) and binary can be used as low level output of SCSI responses that can be redirected to a file and decoded with a later invocation. JSON output is typically generated after decoding and considered a high level output that may be parsed by some other tool. No utilities in sg3_utils can parse JSON as input. <P> JSON output is outlined in its own manpage, see: <A HREF="../man8/sg3_utils_json.8.html">sg3_utils_json</A>(8). <P> Binary and especially hexadecimal output can have two roles: for viewing immediately (e.g. when debugging) or as a parseable output that can be decoded later. The latter action uses the <I>--inhex=FN</I> option. For viewing hexadecimal output (e.g. from a SCSI command response) the <I>--hex</I> option can be used once or twice. Both have output similar to: <BR> $ hexdump --no-squeezing -C <P> When there is a single <I>--hex</I> option, the ASCII rendering to the right of each line is not shown; when used twice the ASCII rendering is shown. <BR> When the <I>--hex</I> option is used three or more times then the assumption is that it will be parsed later. Only hex data bytes are output, up to 16 per line. This is close to this hexdump invocation: <BR> $ hexdump --no-squeezing -C | cut -s -d ' ' -f 2-19 <P> This should remove the first column (i.e. the count) and remove the ASCII rendering to the right. This makes parsing the hexadecimal a lot easier. When used 4 or more times the <I>--hex</I> options adds comments (i.e. lines starting with the hash symbol): <BR> $ sg_logs -HHHH -A /dev/sdc <P> <BR> # Supported log pages log page [0x0]: <BR> 00 00 00 03 00 0d 2f <P> <BR> # Supported log pages and subpages log page [0x0, 0xff]: <BR> 40 ff 00 0e 00 00 00 ff 0d 00 0d 01 0d ff 2f 00 <BR> 2f ff <BR> ... <P> Comments can also be placed after hex in a line, everything from the '#' to the end of a line is ignored. <P> There are many examples of hex files suitable for the <I>--inhex=FN</I> option in the 'inhex' directory. The naming of files in that directory is the name of the utility that will decode it with the "sg_" prefix removed and a ".hex" suffix added. Those hex files can be viewed with a text editor and most contain an example of invoking it in the leading comments. <P> The hexadecimal format can be converted into binary using the sg_decode_sense utility with these options: "<I>--inhex=HFN --nodecode --write=WFN</I>". The input (in hex) is in the <I>HFN</I> file while the output is placed in the <I>WFN</I> file. <P> To convert a binary file into a hexadecimal form that can be given as input to various sg3_utils utilities, the sg_decode_sense utility can also be used with these options: "<I>--binary=BFN --nodecode -HHH</I>" and the hex output will be sent to the console (stdout). <A NAME="lbAP"> </A> <H2>MICROCODE AND FIRMWARE</H2> There are two standardized methods for downloading microcode (i.e. device firmware) to a SCSI device. The more general way is with the SCSI WRITE BUFFER command, see the sg_write_buffer utility. SCSI enclosures have their own method based on the Download microcode control/status diagnostic page, see the sg_ses_microcode utility. <A NAME="lbAQ"> </A> <H2>SCRIPTS, EXAMPLES and UTILS</H2> There are several bash shell scripts in the 'scripts' subdirectory that invoke compiled utilities (e.g. sg_readcap). Several of the scripts start with 'scsi_' rather than 'sg_'. One purpose of these scripts is to call the same utility (e.g. sg_readcap) on multiple devices. Most of the basic compiled utilities only allow one device as an argument. Some distributions install these scripts in a more visible directory (e.g. /usr/bin). Some of these scripts have man page entries. See the README file in the 'scripts' subdirectory. <P> There is some example C code plus examples of complex invocations in the 'examples' subdirectory. There is also a README file. The example C may be a simpler example of how to use a SCSI pass-through in Linux than the main utilities (found in the 'src' subdirectory). This is due to the fewer abstraction layers (e.g. they don't worry the MinGW in Windows may open a file in text rather than binary mode). <P> Some utilities that the author has found useful have been placed in the 'utils' subdirectory. <A NAME="lbAR"> </A> <H2>DEBUGGING</H2> Each utility and most scripts have a <I>--verbose</I> option (short form: <I>-v</I>) that can be used multiple times to increase the verbosity of the output to aid debugging. Normal output (if any) is sent to stdout while verbose output (and error output) is sent to stderr. This may be important when the (normal output) of a utility is being piped to another command (e.g. the grep command to find a particular field in the output). <P> The Linux SCSI subsystem has a pseudo file for getting and changing the SCSI logging level: /proc/sys/dev/scsi/logging_level . The scsi_logging_level script in this package can be used to manipulate the logging level in a command line friendly way. See its manpage. <P> The logging level runs from 0 (no logging and the default) to 7 (lots of logging) and applies to all storage devices that use the SCSI subsystem. The logging output goes to "the log" which is often the /var/log/syslog file. <P> The Linux SCSI generic (sg) driver is often used under the utilities in this package. It uses a seldom (otherwise) used logging type of SCSI_LOG_TIMEOUT. An example of its use to turn on full debugging is: <P> <BR> scsi_logging_level -s -T 7 <P> To reduce the amount of output to only error paths, the following is suggested: <P> <BR> scsi_logging_level -s -T 3 <P> And to turn off logging in the sg driver: <P> <BR> scsi_logging_level -s -T 0 <P> For analyzing machine crashes associated with a SCSI command, nothing beats a real serial port. By "real" means that it is _not_ a USB serial port. The reason is that like SCSI, USB needs a functioning software stack within the OS kernel, the very thing that may be crippled during a machine crash. <P> Modern laptops do not have real serial ports and many server machines don't either (or it is an optional extra). In Linux the netconsole module does a pretty good job by sending log entries to another machine (on the same sub-net)) using the UDP ("fire and forget") network protocol . <A NAME="lbAS"> </A> <H2>WEB SITE</H2> There is a web page discussing this package at <A HREF="https://sg.danny.cz/sg/sg3_utils.html">https://sg.danny.cz/sg/sg3_utils.html</A> and also at <A HREF="https://doug-gilbert.github.io/sg3_utils.html">https://doug-gilbert.github.io/sg3_utils.html</A> . The device naming used by this package on various operating systems is discussed at: <A HREF="https://sg.danny.cz/sg/device_name.html">https://sg.danny.cz/sg/device_name.html</A> . There is a git code mirror at <A HREF="https://github.com/hreinecke/sg3_utils">https://github.com/hreinecke/sg3_utils</A> . The principle code repository uses subversion and is on the author's equipment. The author keeps track of this via the subversion revision number which is an ascending integer (currently at 922 for this package). The github mirror gets updated periodically from the author's repository. Depending on the time of update, the above Downloads section at sg.danny.cz may be more up to date than the github mirror. <A NAME="lbAT"> </A> <H2>AUTHORS</H2> Written by Douglas Gilbert. Some utilities have been contributed, see the CREDITS file and individual source files (in the 'src' directory). <A NAME="lbAU"> </A> <H2>REPORTING BUGS</H2> Report bugs to <dgilbert at interlog dot com>. <A NAME="lbAV"> </A> <H2>COPYRIGHT</H2> Copyright © 1999-2026 Douglas Gilbert <BR> Some utilities are distributed under a GPL version 2 license while others, usually more recent ones, are under a BSD-2-Clause license. The files that are common to almost all utilities and thus contain the most reusable code, namely sg_lib.[hc], sg_cmds_basic.[hc] and sg_cmds_extra.[hc] are under a BSD-2-Clause license. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. <A NAME="lbAW"> </A> <H2>SEE ALSO</H2> <B>sg3_utils_json, sg_decode_sense(sg3_utils), sdparm(sdparm), ddpt(ddpt),</B> <B>lsscsi(lsscsi), <A HREF="../man1/dmesg.1.html">dmesg</A>(1), <A HREF="../man1/mt.1.html">mt</A>(1), hdparm(hdparm), hexdump(util-linux)</B> <BR> The format of this section is: <utility_name>(<package_containing_utility>) or <utility_name>(<manpage_section_number_containing_utility>) . <P> <HR> <A NAME="index"> </A><H2>Index</H2> <DL> <DT><A HREF="#lbAB">NAME</A><DD> <DT><A HREF="#lbAC">SYNOPSIS</A><DD> <DT><A HREF="#lbAD">DESCRIPTION</A><DD> <DT><A HREF="#lbAE">COMMAND DESCRIPTOR BLOCK</A><DD> <DT><A HREF="#lbAF">ENVIRONMENT VARIABLES</A><DD> <DT><A HREF="#lbAG">LINUX DEVICE NAMING</A><DD> <DT><A HREF="#lbAH">WINDOWS DEVICE NAMING</A><DD> <DT><A HREF="#lbAI">FREEBSD DEVICE NAMING</A><DD> <DT><A HREF="#lbAJ">SOLARIS DEVICE NAMING</A><DD> <DT><A HREF="#lbAK">NVME SUPPORT</A><DD> <DT><A HREF="#lbAL">EXIT STATUS</A><DD> <DT><A HREF="#lbAM">COMMON OPTIONS</A><DD> <DT><A HREF="#lbAN">NUMERIC ARGUMENTS</A><DD> <DT><A HREF="#lbAO">HEX, BINARY AND JSON FORMATS</A><DD> <DT><A HREF="#lbAP">MICROCODE AND FIRMWARE</A><DD> <DT><A HREF="#lbAQ">SCRIPTS, EXAMPLES and UTILS</A><DD> <DT><A HREF="#lbAR">DEBUGGING</A><DD> <DT><A HREF="#lbAS">WEB SITE</A><DD> <DT><A HREF="#lbAT">AUTHORS</A><DD> <DT><A HREF="#lbAU">REPORTING BUGS</A><DD> <DT><A HREF="#lbAV">COPYRIGHT</A><DD> <DT><A HREF="#lbAW">SEE ALSO</A><DD> </DL> <HR> This document was created by <A HREF="/cgi-bin/man/man2html">man2html</A>, using the manual pages.<BR> Time: 01:03:13 GMT, July 22, 2026 </BODY> </HTML>