SG_SAFTE(8)

✨ NAME ✨

okay bestie, this is the SG_SAFTE(8) page, and it's all about sg_safte, which lets you access a SCSI Accessed Fault-Tolerant Enclosure (SAF-TE) device 💅

🎀 SYNOPSIS 🎀

sg_safte [--config] [--devstatus] [--encstatus] [--flags] [--help]
[--hex] [--insertions] [--raw] [--usage] [--verbose] [--version] DEVICE

🌸 DESCRIPTION 🌸

okay so sg_safte fetches the enclosure status by sending a SCSI READ BUFFER command 🫶 The DEVICE should be a SAF-TE device, and that can be either a storage array controller (INQUIRY peripheral device type 0xc) or a generic processor device (INQUIRY peripheral device type 0x3). no cap, it's giving hardware nerd ✨

If you don't give any options and just pass the DEVICE argument, you get the overall enclosure status, as reported by the option --config ®is reported. main character energy 👑

---

💜 OPTIONS 💜

Quick vibe check: arguments to long options are mandatory for short options too. The options are also sorted alphabetically by their long name, so they're super easy to find 🧚

• -c, --config — this one issues a Read Enclosure Configuration ®(READ BUFFER ID 0) cdb to the device, and it gets you a list of the enclosure hardware resources 🌈
• -d, --devstatus — slays. It sends a Read Device Slot Status ®(READ BUFFER ID 4) cdb, which tells you the current state of each drive or slot 💖
• -s, --encstatus — sends a Read Enclosure Status ®(READ BUFFER ID 1) cdb, which tells you how the components are operating 🔥
• -f, --flags — sends a Read Global Flags ®(READ BUFFER ID 5) cdb and reads the most recent state of the global flags on the RAID processor device 🦋
• -h, --help — prints the usage message, then exits 🫶
• -H, --hex — prints the response to a READ BUFFER command in ASCII hex to stdout. Use it once and you get the response to the first READ BUFFER command (buffer_id=0), which should be the enclosure configuration. Use it twice (or more) and you get the responses to the later READ BUFFER commands too 🍓
• -i, --insertions — sends a Read Device Insertions ®(READ BUFFER ID 3) cdb, which counts how many times devices got inserted while the RAID system was powered on ✨
• -r, --raw — prints the response to a READ BUFFER command in binary to stdout. Once gives you the first response (buffer_id=0, the enclosure configuration), and twice or more gives you the later responses as well 🌸
• -u, --usage — sends a Read Usage Statistics ®(READ BUFFER ID 2) cdb to the device, and you get the total usage time and number of power-on cycles of the RAID device ⭐
• -v, --verbose — turns up the verbosity (aka debug output) 🙈
• -V, --version — prints the version string, then exits 💅

---

🦋 NOTES 🦋

ngl, this implementation is based on the intermediate review document dated 19970414, named "SR041497.pdf". So yeah, it's pretty old 💀 Intel and nStor are the authors. Intel has a zip archive with this and related documents in the "SAF-TE: SCSI Accessed Fault Tolerant Enclosures Interface Specification" section of this page: https://www.intel.com/content/www/us/en/servers/ipmi/ipmi-technical-resources.html

Lowkey, there's similar functionality in SPC-4 SCSI Enclosure Services (SES) devices (Peripheral device type 0xd), which you can query with the sg_ses utility 🧚

🍓 EXAMPLES 🍓

To view the configuration:

sg_safte /dev/sg1

To view the device slot status:

sg_safte --devstatus /dev/sg1

🌈 EXIT STATUS 🌈

If it works, sg_safte exits with 0 🎉 If not, check the sg3_utils(8) man page for the rest of the exit codes 🫶

👑 AUTHORS 👑

Written by Hannes Reinecke and Douglas Gilbert 💖

💌 REPORTING BUGS 💌

Found something broken? Report bugs to <dgilbert at interlog dot com> 🙈

🔥 COPYRIGHT 🔥

Copyright © 2004-2016 Hannes Reinecke and Douglas Gilbert. This software is distributed under a BSD-2-Clause license. There is NO warranty, not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE 💀

🧚 SEE ALSO 🧚

sg_inq, sg_ses (in sg3_utils package); safte-monitor (internet) ✨

---

okay that's the whole tea ☕ this page came from sg3_utils-1.43, April 2016. stay iconic, bestie 🌸💜🦋