|
SMARTMon-UX may be invoked as follows:
smartmon-ux
smartmon-ux [options] [device_list]
smartmon-ux -h
(If you are on an Apple, you must either run from root or invoke the software with sudo, as in sudo ./smartmon-ux [options] [device_list])
If you launch smartmon-ux without any options, the program will discover and report all devices, enable S.M.A.R.T. on all drives that support this feature, set the polling interval to 10 minutes, and run in the background. Status messages be recorded in the system log file, unless overridden by using the -L option as described below. All command details are below, listed in alphabetical order.
With only a few exceptions, all operating systems support all of these commands. The most notable exception is that Windows platforms do not support the -O command, and that the -16 or -12 command may require a service pack, kernel patch, or update.
Case-sensitive options (grouped alphabetically). Some commands, such as the -Mail function are specific to certain operating systems.:
| -A | Displays a hex dump of all mode pages for all devices (or devices in device list) and terminates the program. |
| -B C|S Hlist | Invokes the mode page editor feature to program revised mode page data for the selected disk(s). The -C flag is used to tell the program to change the current settings, and the -S flag instructs it to change the saved, or permanent settings. The saved settings will make the new mode page non-volatile, so they will be in force when the disk goes through a power cycle. The current mode page setting will be effective immediately and will be lost when the disk recycles. Never change the S page unless you are 100% sure you know what you are doing, as you could render your disk drive invisible to the operating system, or even cause data loss. |
The Hlist is the hexadecimal list of bytes that you want programmed into the disk. The program checks for valid syntax and byte count, but it does not protect you against programming the disk drive with settings which may be inappropriate for your particular environment.
Example smartmon-ux -B C 1A,A,0,1,0,0,0,0,0,0,8c,a0 /dev/sga would instruct the selected disk to automatically spin down after 60 minutes of inactivity.
| -bmsd | Disables background media scanning (Available on certain Seagate disk drives) |
| -bmse n | Enables background media scanning, and sets interval to n hours |
| -bmsr | Reports background media scanning state and provides detailed report |
| -capacity n | Reprograms / resizes the disk programmatically, so that it reports a user-defined capacity of n blocks. Send -capacity 0 to reset the disk to maximum capacity. |
| -confirm | Automatically responds "Y" to the are-you-sure type messages you get before running potentially destructive functions. |
| -C | Dump statistical device information (Log pages - in decoded ASCII text) see notes below |
| -Cx | Dump statistical device information (Log pages - in decoded ASCII text). This is an improved syntax that suppresses a trailing field that indicates the number of bytes that the peripheral allocates to the field. |
| -C+ | Same as the -C, only do brute-force log page discovery. Use this to force the program to manually poll every possible log page. Use this for devices which have log pages that are not reported due to the device not meeting ANSI compliance. |
| -d | Specifies that the remainder of the command-line contains a device list and/or device wild-card expressions |
| -E | Poll SES/SAF-TE information (fans, power supply, enclosure temperature), etc. This requires your disks to be mounted in either a SES-compliant enclosure, or SAF-TE enclosure. |
| -E+ | This is the verbose mode of the SES query. It displays additional details on many models of enclosures that are vendor-specific extensions to the ANSI SES specification. If you are addressing a SAF-TE enclosure, no additional information will be displayed. |
| -EF | Add this command to any -E family command to address situations where no SES data is reported, but you *know* the enclosure supports SES. This instructs the program to perform a brute-force SES discovery rather than query the enclosure's capability. As some enclosures and enclosure firmware are not fully ANSI compliant, we were forced to add this command to address the situation. |
| -EH | Print hex dump of all enclosure pages (includes both ANSI defined and vendor-specific pages) |
The "EP" functions allow you to program characteristics of your SES-enabled enclosure. Not all SES enclosures support all of these commands. Further details on these commands can be found in the section Enclosure Services Configurator.
| -EPAmn | Un-mute (turn on) audible alarm #n |
| -EPARn | Set alarm #n to reminder mode |
| -EPArn | Clear alarm #n from reminder mode |
| -EPATxn | Set alarm tone urgency control for alarm #n to x, where x is hex value 0 - F |
| -EPDFn | Enable visual fault indicator for device in slot #n |
| -EPDfn | Disable visual fault indicator for device in slot #n |
| -EPDIn | Identifies device in Slot #n |
| -EPDin | Disable identification for device in Slot #n |
| -EPLFn | Enable visual fault indicator for array device in slot #n |
| -EPLfn | Disable visual fault indicator for array device in slot #n |
| -EPLIn | Identifies array device in Slot #n |
| -EPLin | Disable identification for array device in Slot #n |
| -EPLRn | Enable visual rebuild indicator for array device in slot #n |
| -EPLrn | Disable visual rebuild indicator for array device in slot #n |
| -EPLSn | Enable visual remove indicator for array device in slot #n |
| -EPLsn | Disable visual remove indicator for array device in slot #n |
| -EP2ttnnwwxxyy | Provides complete programmability of all SES control page fields, whether ANSI defined or vendor-unique. This sends bytes ww xx yy to SES control page 2, for element type tt, element number nn. The section Enclosure Services Reprogrammingcontains further information. All command options must be 2-character hex numbers. |
| -F freq | Sets the default polling frequency from 600 seconds (10 minutes) to any number of seconds. (This option can now be added in combination with dump-type options such as -I+ to cause program to wait until exiting. You would ordinarily need this under Windows only, if you were using a .BAT script). Setting the freq value to 0 instructs the program to poll once and then exit. |
| -fc | Dumps additional fibre channel information (SAN discovery, frame-level statistics and errors, fabric and switch information, etc...) |
| -fchbainfo | Report Fibre Channel HBA information (make, model, firmware, driver, etc...) and exit. |
| -fciostat [options ...] [ <interval> ] [ <count> ] | Equivalent of UNIX iostat function, but for fibre channel HBAs. |
| -fciostat [ -help | -? ] | Reports option and usage info specific to this function |
| -fcping WWN LUN [n] | Pings a fibre channel port WWN and LUN, n times. This will verify connectivity as well as report return time in thousandths of a second. (If n=0, then ping indefinitely) |
| Flash firmware to SES-compatible enclosure |
| Flash firmware to SES-compatible enclosure (uses alternate "mode 7" technique, in event the enclosure does not support -flashses command) |
| -format | Format disk (perform a low-level format / i.e., issue FORMAT UNIT command) to the selected SAS/SCSI/FC/USB disk. |
| -formatb | Format disk (perform a low-level format / i.e., issue FORMAT UNIT command) to the selected SAS/SCSI/FC/USB disk in background (most disks support this, the program will not lock up and wait until the formatting has completed until it returns. |
| -formatconf | Disables the safety are-you-sure message when accompanied by a -format family command. |
| -formatg | Same as the -format command, but this will instruct software to automatically clear the grown defect list. You would generally accompany this with the -formatconf command to automate formatting within a script or batch file. |
| -G temp | Sets the thermal temperature warning in degrees Centigrade. If not specified, the default is 45 degrees. |
| -h | Displays all of the above usage information, and terminates the program. (Many UNIX shells will substitute the ? character, so best to use this instead of -?. |
| -H | Dump statistical device information (Log pages - full hex dump) see notes below |
| -H+ | Same as -H, above, but uses brute-force discovery of all log pages 0 - 3E. Added for devices that do not properly report log page 0. |
| -HEALTH | General disk / tape health report (short format). No other command-line options are required. |
| -i | International localization. Use this flag as part of any command-line to instruct the software to display date/time fields in the format native to your particular country. |
| -I | Displays a hex dump of all inquiry information for all devices (or devices in device list) and terminates the program. If the selected device uses a SCSI or Fibre Channel interface, this is the standard SCSI inquiry output. If the disk is an IDE disk, the resulting output is from the Identify Device command. |
| -I+ | This is the verbose mode of the inquiry command, and it instructs the program to also display hex dumps of all Extended Vital Product Data Pages (EVPD pages. These extended pages display additional information such as serial numbers and vendor-unique information. |
| -IS | Returns the serial number of installed media for tape drives. This command is not supported by all tape drives. |
| -J | Mode page viewer - Decodes ANSI-standard mode pages settings and displays in readable text with descriptions |
| -L | Instructs program to send logger output to /var/log/smartmon-ux, or /var/adm/smartmon-ux depending on what O/S you are running. (OS X and LINUX default to /var/log, UNIXWARE, IRIX, SOLARIS, AIX, and HPUX go to /var/adm).VMS uses a log file SMARTMON.LOG in the currently selected directory. Syslog, file, and windows event logging are discussed in more detail in the System Event Log chapter. |
| -LRemote Host | Sends messages to the remote system event log. (This flag only supported in Windows). Example -LRemote \\NOCSUPPORT2.or -LRemote \\12.18.1.25 |
| -LB <Scriptfile> | Launches the program or script, <Scriptfile>, in event of a predictive failure alert or in conjunction with the test message (-T option) |
| -link | Reports current interface speed (U320, U160, U80 ...) of SCSI / FC device at polling time. Not all devices have this capability. Use this for enclosure and cable testing. |
| -Mail | Interactively configures email account settings for SMTP servers that require authentication (Windows-specific flag). |
| -M <EMAIL> | Instructs program to send an alert via email to the email address supplied. Example: smartmon-ux -M david@xyz.com. The sendmail (or other mailer) daemon must be properly configured for your machine in order for this to work. Of course, the email address could also be that of a paging service or an alias list which would send the message to as many people as you desire. If you are running windows, you must configure the SMTP mail server. |
If you are using the Windows version, you must format the line as follows and use the -N option to supply the IP name of your email server. In addition, you can add up to 8 email addresses. If you do not supply either the -N or -M options, you will get a command-line error. Use the -Mail command to define your email server.
smartmon-ux -N mail.gte.net -M "<sysadmin@gte.net>,"<mypager@gte.net>" ... {This command is not supported under VMS}
| -mpexport FILE | Exports all mode pages for selected device to an ASCII text file that you may edit. Use the -mpimport command to burn the saved mode pages onto the same or equivalent device. (Example: -mpexport seagate.txt /dev/rdsk/c0d0s0 ) |
| -mpimport FILE | Imports mode pages from FILE and burns them onto selected device. |
Example: -mpimport seagate.txt /dev/rdsk/c0d0s0 /dev/rdsk/c0d[3-5]s0
| -N SMTPAcct | This windows-specific flags let you assign the desired EMAIL server for sending messages. The SMTPAccount must be the full network name, rather than an IP number. See the example above. |
| -O | Dumps detailed ATA/SATA disk drive error log report on supported operating systems. |
| -P | Enable the performance (PERF) bit. This disables S.M.A.R.T. tests which could cause delays. Not all disk drives support this feature. |
| -p | Disable S.M.A.R.T. for selected disks and exit. The disks are programmed via the mode page editor to turn feature off in current (volatile) settings. The saved (non-volatile) pages are not affected. You must use the mode page editor feature to permanently disable S.M.A.R.T. |
| -pp | Disable S.M.A.R.T. for selected disks, and save it, so that the only way to revert is use the mode page editor. |
| -ping | Report if device has been removed or does not respond to a poll (after initial discovery). |
| -Q | Displays partition information and file system types, then the program is terminated. (This option is available on the LINUX, SPARC, OS X and WindowsTM family operating systems). |
| -random n | Sets every bit on the selected SAS, SCSI, USB or fibre channel disk to random data. Then -n refers to the number of desired passes. This does not use the secure erase function, but it is rather fast. |
| -rb BlockNo | Reassign block #BlockNo on selected SCSI, FC, SSA, or SAS disk. This feature is not yet supported on ATA or SATA devices. The Block number must be decimal). |
| -rb BlockNoh | Reassign block #BlockNo, but BlockNo is in hex, i.e. -rb f7d01h |
| -rc BlockNo | Corrupt block #BlockNo on selected SCSI, FC, SSA or SAS disk. The ECC information will be incorrect so the next read on that block will generate an unrecovered read error. |
| -read s,n,FILE | Reads n blocks from random access device starting at block #s and saves to binary file. (Block size can be from 512 - 528) |
| -scrub | Fitness test (full I/O test with detailed error reporting - usually takes hours |
| -scrubv | Fitness test, same as above but verbose. Reports errors as discovered and percent complete. |
| -scrubq | Quick fitness test. Reads 32 blocks at a time for faster completion, but sacrifices granularity. |
| -scrubs | Sequential seek fitness test. |
| -scrubr | Pseudo-random seek fitness test. |
| -scrubt | Instructs program to terminate any "scrub" family fitness test upon first error with return code 11. |
| Destroys all data on the disk by sending n triple-pass iterations of all zeros, ones, and random bits. |
| -securecheck n | Analyze data on device to confirm randomness and/or erasure patterns. The -n parameter sets the maximum time in minutes you want it to run. Enter 0 to check entire disk, or use -securecheckall. |
| -securecheckall | This analyzes the entire disk to validate and report randomness. |
-spinq Report whether drive is spun up, down, or in a transitional state.
-spindown Spin the drive down (same as SCSI STOP UNIT command) and wait for drive to spin down before returning.
-spindowni Spin the drive down (same as SCSI STOP UNIT command) and return immediately after issuing command.
-spinup Spin the drive up (same as SCSI START UNIT command) and wait for drive to spin up before returning.
-spinupi Spin the drive up (same as SCSI START UNIT command) and return immediately after issuing command.
| -steb | Initiates self-test, extended, background for selected SCSI, SAS, and fibre channel devices. |
| -steba | ATA/SATA Extended Background Self-Test. Note: This will typically take 1-2 hours, but it does not lock up the disk. |
| -stfd | Initiates factory default self-test. (Which is vendor/product specific, but generally completes in one or two minutes). |
| -stefa | ATA/SATA Extended Foreground Self-Test. Note: This will lock up the drive while it runs, do not perform it on a disk that is mounted by the O/S. |
| -stsb | Initiates self-test, short, background for selected SCSI, SAS, and fibre channel devices. |
| -stsba | ATA/SATA Short Self-Test. (Must complete in under 2 minutes per ANSI specification) |
| -sta | Aborts current self-test for selected SCSI, SAS, and fibre channel devices |
| -staa | Aborts current self-test for selected ATA / SATA disk device. |
| -str | Reports results and status of current and last self-tests for selected SCSI, SAS, and fibre channel devices. |
| -stra | Reports results and status of current and last self-tests for selected SCSI, SAS, and fibre channel devices. |
| -T <EMAIL> | Instructs program to send out a test predictive failure alert. The <EMAIL> address is optional. This may be used with the -LB option. |
(Windows users will normally add the -N flag and SMTP server name to specify an account to use).
| -sq | Suppress logging successful polling messages in the system event log. |
| -sqq | Suppress all logging into the system event log. |
| -wsbyte hexbytevalue | Writes the hex byte value to every block on the selected device using the efficient WRITE SAME command. |
| -wsbyteconfirm hexbytevalue | Writes the hex byte value to every block on the selected device using the efficient WRITE SAME command. The -wsbyteconfirm command does not ask an are-you-sure question, so it can easily be scripted. |
| -wsc | This optional command can be used with both -wsbyte and -wsbyteconfirm, and it instructs the program to immediately terminate after a write error on the disk drive. |
| -verify | Instructs disks to read/verify all sectors on the device. Bad blocks will be reported. (This runs mostly within the disk firmware so it is very fast). The command is supported under all SAS/FC/SCSI disks as well as SATA/ATA disks under windows only. |
| -Wfilename | Enables threshold monitoring, using parameters defined in filename. Combine this command with the -F option and a list of desired SCSI/FC devices. The configuration file, filename created interactively with the -K command. Example: -F 60 -WUnrecoveredWriteDaemon.cfg |
Important: there must NOT be any white space between the -W and the filename. (If you leave white space, then the program will incorrectly interpret the next option as a physical device name)
| -wcd | Disable write cache. This disables a SCSI/SAS/SSA or Fibre channel disk drive's write cache. (The function is currently not supported on IDE or SATA disks). |
| -wce | Enable write cache. This enables a SCSI/SAS/SSA or Fibre channel disk drive's write cache. (The function is currently not supported on IDE or SATA disks). |
| -wp | Write-protect test. This performs a test to see if media (typically tape) is write protected. |
| -X | Polls selected tape devices that support the TapeAlert feature. (This can be equated to S.M.A.R.T. for disk drives). |
| -XT | TapeAlert Test. This enables test mode, does single poll, disables test mode, then exits. It should not be run on tapes/auto changers currently in use. |
| -X+ | Reports all TapeAlert components that the selected tapes are capable of reporting. Note: Not all tape drives can be queried to learn exactly what TapeAlert flags it supports. Program terminates after displaying this information. |
| -z | Report physical and logical drive information for selected IBM, SGI and Engenio (formerly LSI) RAID engines. |
| -Z | Report physical and logical drive status for subsystems using Mylex fibre channel external RAID engines. Supported engines are FF, FFX, FF2, FFx2, also known as the SANArray Pro family. All engines must be running FW 7.0 or higher. [Mylex RAID] |
| -ZA start# n | Display n RAID event log entries >= starting# (if n=0, display all events) [Mylex RAID] |
Example: smartmon-ux -ZA 3440 32 /dev/sgj would dump up to 32 events starting at event #3440
| -ZL | Display all RAID event log entries. [Mylex RAID]. This option may be run without the -Z flag). |
| -ZM | Report Mylex SAN-Mapping table |
| -zd[x] | Report physical and logical drive info for selected LSI-MPT family RAID engines. The -x suffice reports extended information. |
| -zdL | Report LSI-MPT RAID event log |
| -zdq | Report LSI-MPT physical disk status and serial numbers |
| -zi | Report physical and logical drive information for Infortrend-family RAID engines. |
| -zie | Display enclosure state summary and full event log [Infortrend RAID]. |
| -ziL | Display all RAID event log entries [Infortrend RAID]. |
| -zm | Continuously monitor Infortrend event log rather than just dump and exit. |
| -zix | Report detailed back-end drive information [Infortrend RAID]. This should be run during a maintenance window. |
| -ziA start# n | Display n RAID event log entries >= starting# (if n=0, display all events) [Infortrend RAID] |
| -z3[x] | Report physical and logical drive info for selected 3ware (AMCC) family RAID engines. The optional "x" suffix reports extended information. |
| -z3d | Report controller (3ware (AMCC)) diagnostic dump (this is very cryptic but useful to RAID controller experts and OEMs who imbed the controller |
| -z3L | Report controller (3ware (AMCC)) event log |
| -z3m | Monitor 3ware / AMCC health in background (or as a Windows service) |
| -? | Displays all of the above usage information and terminates the program. (Many UNIX shells will substitute the ? character, so best to use the -h flag instead. |
| -16 | Forces the -ws, -wsbyte, and all "scrub" family commands to send READ(16) and WRITE(16) CDBs instead of 10-byte CDBs. Note that your O/S, drivers, and target peripheral must all support these extended SCSI commands. (Windows uses need Win2003 with SP1, and LINUX users will require the 2.6 kernel). |
| -12 | Forces the "scrub" family commands to use the READ(12) and WRITE(12) commands instead of the READ(10) and WRITE(10) CDBs. |
Unless the debug parameter is sent, the program will run in the background. This has the same effect as entering a trailing ampersand (&). i.e., smartmon-ux -F 3000 has the same effect as smartmon-ux -F 3000 &. This is by design, to automate running SMARTMon-UX at boot time.
Some examples:
| 1) smartmon-ux | Scan for all disk drives. If any disk drives that support S.M.A.R.T. are found, then the program re-launches itself in the background with a 10-minute polling period, and sends the results to the system log file |
| 2) smartmon-ux -M admin@xyz.com | Same as above, but alerts are sent to email address supplied. |
| 3) smartmon-ux -I -S /dev/sd0 /dev/sd3 | Dumps inquiry data and mode pages for the two disks, /dev/sd0 and /dev/sd3 and terminates the program. |
Notes on Statistical Device Information
The statistical information options (-C & -H) are applicable to SCSI, Fibre Channel, and IBM SSA disk drives only. IDE disks do not maintain these fields. Most of the data is non-volatile, and they are stored in what is called Log Pages. Some fields are defined by the ANSI SCSI specifications, and others are vendor/drive specific. There is a lot to discuss here, so we have dedicated a chapter called Log Page Viewer to this subject.
Notes on Device List and using Wild Cards
The [device_list] is used to supply a list of physical devices which you want the command-line options to be executed on. If you do not supply a device list, all devices will be acted upon. So, if you were to enter smartmon-ux -I, it will display inquiry information for all devices it discovers.
By using wild cards, you can quickly enter multiple devices rather then entering them individually. The * matches any string of characters or numbers, any length from that point onward. The [list] matches any single character in the list. i.e, /dev/rdsk/c[236]d* means it will match /dev/rdsk/c2d*, /dev/rdsk/c3d*, or /dev/rdsk/c6d*.
You may also combine devices that use wild cards, and those that do not, as in "./smartmon-ux /dev/sga /dev/sgc /dev/rmt/*".
Apple users will use device numbers, as in ./smartmon-ux 0 3 8
Commands by Function Type
Flag
|
Description
|
Destructive
|
Notes
|
Polling Commands (Program continues to run if only these commands are supplied)
|
-E
|
Enclosure check
|
No
|
This command can be run in foreground also
|
-F
|
Polling frequency
|
No
|
Default if -F not supplied is 10 minutes
|
-G
|
Thermal warning
|
No
|
Adds temperature to polling log if supported on device.
|
-i
|
International date/time
|
No
|
Returns date/time in local language & format
|
-link
|
SCSI/FC link speed
|
No
|
Adds current interface speed to poll
|
-L
|
Logging
|
No
|
Sends messages to smartmon-ux file instead of syslog
|
-LRemote
|
Logging
|
No
|
Specifies remote host (For Windows version, Active Directory implementations)
|
-sq
|
Logging
|
No
|
Suppress logging successful polling messages. (Only messages that indicate a problem will be logged).
|
-M
|
E-Mail address
|
No
|
SMART Alerts, Tape Alerts and threshold warnings generate email.
|
-P
|
PERFormance bit
|
No
|
SCSI/FC drives prioritize application I/O over S.M.A.R.T. tests
|
-Wfilename
|
Statistical alerting
|
No
|
Combine with the -F to set minimum time between polls and to issue custom threshold monitoring scripts using config file supplied with -W
(replace filename with the name of your file, as in -F 60 -WUnrecoveredWriteDaemon.cfg [You must not have a space between the -W and the file name)
|
-X
|
TapeAlert monitoring
|
No
|
Like S.M.A.R.T., but for tape drives and auto changers.
|
General reporting commands (Program terminates after reporting information on first pass)
|
Flag
|
Description
|
Destructive
|
Notes
|
-A
|
Mode page hex dump
|
No
|
|
-J
|
Mode page text dump
|
No
|
|
-C
|
Log page text dump
|
No
|
|
-Cx
|
Log page text dump
|
No
|
You will most likely prefer output of -Cx over -C as this suppresses the trailing [x] on each field that reports the field size. As a large part of our customers have scripted commands to parse the output, we chose to implement the improved results with -Cx rather than break any scripts by modifying the syntax of -C
|
-H
-H+
|
Log page hex dump
|
No
|
The -H+ provides same output as the -H, but the -H+ does a brute force discovery. This is necessary because some peripherals are not fully ANSI compliant in that they do not provide a list of log pages. As such the -H attempts to read every possible log page. This results in a large amount of I/Os that are likely unnecessary. You would only use the -H+ command if the -H command doesn't report any log pages.
|
-E
|
Enclosure status
|
No
|
|
-E+
|
Extended enclosure status
|
No
|
-EF and -EH are related commands.
|
-I
|
SCSI Inquiry dump
|
No
|
|
-I+
|
Extended SCSI inquiry
|
No
|
|
-IS
|
Return serial number
|
No
|
Returns serial number of removable media. (Generally for auto changers and tape libraries).
|
-O
|
IDE (ATA/SATA) inquiry
|
Possibly
|
The -O option could take several seconds to complete, so this might be disruptive. We have seen this command take almost 30 seconds on disks that have problems.
|
-Q
|
Disk/CD partition dump
|
No
|
|
-S
|
IDE, SATA, ATA S.M.A.R.T. dump
|
No
|
|
-V
|
Display version level
|
No
|
|
-V+
|
Display vendor-unique log/inquiry details
|
No
|
(This is a rather long report that shows all reportable vendor unique device information)
|
-X+
|
TapeAlert capability
|
No
|
|
-XT
|
TapeAlert test
|
No
|
Temporarily enables Test function which is not supported on all tape changers/drives.
|
-Y
|
Factory/Grown defects
|
No
|
|
SES Enclosure commands (Applicable to SES-compliant FC-attached enclosures only)
|
Flag
|
Description
|
Destructive
|
Notes
|
-EPDF
|
Fault light on
|
No
|
|
-EPDf
|
Fault light off
|
No
|
|
-EPDI
|
Identity light on
|
No
|
|
-EPDi
|
Identity light off
|
No
|
|
-EPAM
|
Mute alarm
|
No
|
|
-EPARm
|
Clear alarm
|
No
|
|
-EPArn
|
Set alarm to reminder
|
No
|
|
-EPAT
|
Clear alarm reminder
|
No
|
|
-EPLFn
|
Fault light on, for array device
|
No
|
Some enclosures classify individual drive slots as Array Slots. If the LED does not light with the -EPDF command, try this instead. This is the same for the identify LED (-EPDI)
|
-EPLfn
|
Fault light off, for array device
|
No
|
|
-EPLIn
|
Identity light on, for array device
|
No
|
|
-EPLin
|
Identity light off, for array device
|
No
|
|
-EPLRn
|
Array device rebuild indicator on
|
No
|
|
-EPLrn
|
Array device rebuild indicator off
|
No
|
|
-EPLSn
|
Array device remove indicator on
|
No
|
|
-EPLsn
|
Array device remove indicator off
|
No
|
|
-EP2
|
User defined SES
|
Possibly
|
May be destructive, if your enclosure lets you send commands to turn off all fans, or example.
|
Mode Page Programming (Applicable to SCSI / FC / SAS / USB devices)
|
Flag
|
Description
|
Destructive
|
Notes
|
-B
|
Single line editor
|
Possibly
|
In general, misconfiguring mode pages can render device invisible to O/S
|
-wcd
|
Disable write cache
|
No
|
|
-wce
|
Enable write cache
|
Possibly
|
You should have the device on a UPS or you risk data loss if power is lost before the disk flushes pending I/Os in the cache.
|
-mpexport
|
Export mode pages to file
|
No
|
|
-mpimport
|
Import mode pages from file
|
Possibly
|
Very convenient for cloning all mode pages for multiple devices.
|
Background Media Scanning (Applicable to SCSI / FC / SAS devices that support BGMS function)
|
Flag
|
Description
|
Destructive
|
Notes
|
-bmsd
|
Disable background media scanning
|
No
|
|
-bmse n
|
Enable automated background media scanning every n hours
|
No
|
We strongly recommend you enable this feature.
|
-bmsr
|
Report background media scanning status and bad block list
|
No
|
The report will complete in a few seconds or less.
|
Secure Erase Family Commands
|
Flag
|
Description
|
Destructive
|
Notes
|
-secure n
|
Destroys all data on the disk
|
Yes
|
Set n to 1 for one iteration. This is normally sufficient. The official Department of Defense specification states that you must use 3 full passes for compliance to their spec.
|
-securecheck n
|
Check to see if disk has "data" on it
|
No
|
The -n parameter sets the maximum time in minutes you want it to run. You will generally set the n value to 1, unless the disk is partitioned. If that is the case, set n to zero so it will test entire disk.
|
-securecheckall
|
Look for data on entire disk
|
No
|
This produces chart of how many times each byte is used on the disk, and whether or not there are any repeating patterns that could indicate there is live data.
|
Spin up, down, query (Also referred to START / STOP UNIT)
|
Flag
|
Description
|
Destructive
|
Notes
|
-spinq
|
Inquire spin status
|
No
|
Set n to 1 for one iteration. This is normally sufficient. The official Department of Defense specification states that you must use 3 full passes for compliance to their spec.
|
-spindown
|
Check to see if disk has "data" on it
|
Possibly
|
Never spin a disk down a mounted disk with live data, unless it is your intention to simulate a drive failure. The software does not test to see if the disk is used in any way.
|
-spinup
|
Spin disk up and wait for confirmation
|
No
|
|
-spindowni
|
Spin disk down (immediate)
|
Possibly
|
Same warning as stated above about spinning mounted disks down
|
-spinupi
|
Spin disk up (immediate)
|
No
|
The immediate bit, as defined by ANSI, basically means to return an OK status immediately after the command has been sent, rather than pausing the program while waiting for disk to start up.
|
Miscellaneous Programming
|
Flag
|
Description
|
Destructive
|
Notes
|
-capacity nBlocks
|
Changes drive capacity (resizes disk)
|
No
|
Not destructive as it is reversible, but it can hide usable storage
|
-capacitybs Blocksize
|
Changes block size
|
Yes
|
All data will be lost and drive must be reformatted. Block sizes are normally 512, but some RAID systems, such as NetApp and EMC use 520 block sizes
|
-confirm
|
Automatic affirmative response
|
No
|
Responds "Y" to any are-you-sure type messages that are typically associated with destructive commands such as running a destructive write data integrity test.
|
-flash
|
Flashes device firmware
|
Possibly
|
If you use wrong firmware image then device may have to go back to factory to get recovered.
|
-format
|
Low level format
|
Yes
|
All data will be lost (unless the data was all zeros)
|
-p
|
Disable S.M.A.R.T.
|
No
|
S.M.A.R.T. setting will revert to previous value after power cycle
|
-pp
|
Disable S.M.A.R.T. , permanently
|
No
|
This turns off S.M.A.R.T. so it stays off after power cycle. You must enable it with the mode page editor function.
|
-rb
|
Reassign block
|
Possibly
|
|
-rc
|
Corrupt block
|
Yes
|
Destroys contents of this block by corrupting ECC data. Use this to test to see that a corrupted block is handled properly by your RAID engine or path/data fail over redundancy hardware or software.
|
-wsbyte
|
Write SAME
|
Yes
|
Sends same byte to every block on the random-access device
|
-wsbyteconfirm
|
Write SAME
|
Yes
|
Same as -wsbyte, but no are-you-sure
|
-16
|
16-byte CDB
|
Possibly
|
This forces program to use the 16-byte SCSI command instead of the 10-byte SCSI commands for the -ws and -scrub family functions. Your host O/S and drivers, and target devices must all support 16-byte commands.
|
-12
|
12-byte CDB
|
Possibly
|
This forces using the READ(12) and WRITE(12) commands for the -scrub family commands. Like the 16-byte commands, it is not necessarily going to be supported by your O/S or your storage hardware.
|
RAID Engine Reporting Commands
|
Flag(s)
|
Description
|
Destructive
|
Notes
|
-z
|
Physical drive status (LSI & Mylex engines)
|
No
|
Command ignored if an unsupported RAID engine
|
-Z
|
Physical & logical drive status (Mylex engines)
|
No
|
If you know you have a Mylex (external RAID) engine, then no need to combine with -Z
|
-ZL
|
Full Mylex event log
|
No
|
|
-ZA
|
Selective Mylex event log
|
No
|
Same as -ZL, but you can start at a particular event number
|
-ZM
|
Report Mylex SAN-Mapping table
|
No
|
|
-z3
|
Physical and logical drive status (3ware (AMCC) RAID engines)
|
No
|
Command will be rejected if it is sent to something other than a 3ware 7xxx, 8xxx, or 9xxx family controller.
|
-z3d
|
Reports 3ware (AMCC) diagnostic dump
|
No
|
|
-z3L
|
Reports 3ware (AMCC) event log
|
No
|
|
-zd, -zdx, -zdL
|
Reports Dell (LSI MPT family) RAID controller information
|
No
|
In general, all RAID commands will be rejected by the target device if they are sent to the wrong type of controller, or sent to something other than the RAID controller.
|
-zi
|
Physical and logical drive status (Infortrend RAID engines)
|
No
|
In general, all RAID commands will be rejected by the target device if they are sent to the wrong type of controller, or sent to something other than the RAID controller.
|
-zie, -zil
-ziA, -ziL, -zm
|
Event-log related reporting commands.
|
No
|
The RAID controller will acknowledge that the event log has been reported, but SANTOOLS instructs the RAID engine to not delete them after they have been reported.
|
-zix
|
Detailed physical device information and controller data.
|
Possibly
|
You should not run this command if the system is actively processing I/Os. Some of the commands that it generates could cause a time-out which might affect your host O/S or an application.
|
Miscellaneous Commands
|
Flag
|
Description
|
Destructive
|
Notes
|
-LB
|
Specifies batch program to run in event of predictive failure
|
No
|
Run program with -T and -LB options to test proper launching of the batch program.
|
-read
|
Reads raw device info and saves into file
|
No
|
|
-T
|
Test predictive failure system alerting.
|
No
|
If you pass an email address with this option, you must make sure that your host is configured to send email. Windows users can use the imbedded -Mail command to set up email settings, other operating systems require properly configured sendmail. You may also combine -T with -LB
|
-K
|
Interactive mode for configuring threshold monitoring
|
No
|
Program goes into interactive wizard mode to assist setting up threshold monitoring.
|
-Mail
|
Interactive command for configuring mail server settings
|
No
|
Windows specific option for configuring client PC to be able to send messages to the SMTP server (username, password, IP name, etc..)
|
-N
|
SMTP Server name
|
No
|
Windows specific option, combine with -T and/or -M
|
Service Management Commands
The Windows version of the program has several commands that deal with installing, starting, stoping, and controlling the program when it runs as a Windows service routine. See the Running as a Windows Service chapter for additional details.
Command Syntax:
You may use either smartmon-ux -h or smartmon-ux -? to get help and usage information. As many UNIX/LINUX shells substitute the '?' character for a single-byte wild-card, you should just enter smartmon-ux -h for help, which will work for all operating systems and shells.
|