NVME-REGISTRY-UPDA(1)

💅 nvme-registry-update — like, the page for writing an attribute into a controller's registry entry

okay so bestie, you're reading the nvme-registry-update(1) page from the NVMe Manual and honestly? it's giving "tiny command, big responsibility" ✨ let's get into it.

---

🌸 NAME 🌸

nvme-registry-update — write an attribute to a controller's registry entry

✨ SYNOPSIS ✨

nvme [<global-options>] registry update <device>
                        --attr=<ATTR> | -a <ATTR>
                        --value=<VAL> | -V <VAL>

that's the whole shape of it, no cap 🫶

🦋 DESCRIPTION 🦋

so basically you're writing an attribute into the ownership registry entry for an NVMe-oF controller. the registry lives under /run/nvme/registry/ and it's literally one directory per live controller, with one plain-text file per attribute. if the entry directory isn't there yet? it gets created. main character energy 💅

the <device> is the NVMe controller name (like nvme3), and it's given as a positional argument — the /dev/ prefix is optional, so you can include it or not, no stress.

okay now the important part, listen up 🙈 the owner attribute is normally recorded when the connection is created (like, by nvme connect --owner). changing it by hand can stop an orchestrator from protecting the controller, so updating owner prompts for confirmation when you run it interactively. non-interactive callers (hi, scripts 👋) just proceed without prompting. and before you ask — no, this is not access control, it's a guard against accidental mistakes. a user with root can just edit the files under /run/nvme/registry/ directly. iykyk.

writes are atomic, which is iconic: a temporary file gets written and synced, then renamed into place, so you don't get corruption under concurrent access. slay 🔥

🎀 OPTIONS 🎀

• -a <ATTR>, --attr=<ATTR> — the attribute name you wanna write (e.g. note, owner).
• -V <VAL>, --value=<VAL> — the new value for that attribute. periodt.

🌈 GLOBAL OPTIONS 🌈

these bad boys are defined at the top-level nvme command and they're available to this subcommand too:

• --dry-run — print the command that would be executed, but don't actually execute it. vibes only, no consequences ✨
• --no-ioctl-probing — disable probing for 64-bit IOCTL support.
• --no-retries — disable retry logic on transient errors.
• -o <fmt>, --output-format=<fmt> — set the reporting format to normal, tabular, 'json, or binary. only one output format may be used at a time.
• --output-format-version=<version> — select the output format version. version 1 uses the original field naming, while version 2 (the default) gives you more consistent and script-friendly field names. other versions are rejected, sorry bestie 💀
• --quiet — suppress informational messages. errors still get reported, and --quiet cancels any --verbose for the library log level.
• --set-options=<key=value>[,<key=value>...] — set libnvme library options for this invocation. supported keys are force-4k, mi-probe-enabled, test-base-dir and test-sysfs-dir.
• --timeout=<ms> — set the timeout for the command in milliseconds. infinite or -1 waits without a time limit (0xffffffff).
• -v, --verbose — increase the level of detail in the output. may be specified multiple times to further increase verbosity. the way that just keeps going 📈

these options can also be set as machine-wide defaults in nvme-cli.conf(5). a command-line flag always overrides the file — flag wins, always 👑

🍓 EXAMPLES 🍓

• attach a free-form note to nvme3:

# nvme registry update nvme3 --attr note --value "maintenance 2026-07"

• update a previously written note:

# nvme registry update nvme3 --attr note --value "released"

🧚 SEE ALSO 🧚

nvme-registry-list(1) nvme-registry-retrieve(1) nvme-registry-delete(1) nvme-disconnect-all(1)

💜 NVME 💜

part of the nvme-user suite 🫶

---

and that's the whole page bestie!! 💖 this little gem landed on 10/30/2026 and it's part of the nvme-user suite — go forth and update those registry attributes responsibly, mwah 😭💅