libnvme_set_owner(9)

💅 libnvme_set_owner — the libnvme API manual one, bestie

okay so picture this: you're out here setting the orchestrator identity for the registry, and this is literally the function that does it. come on, slay.

---

✨ NAME ✨

libnvme_set_owner — set the orchestrator identity for the registry 🎀

---

🦋 SYNOPSIS 🦋

int libnvme_set_owner (struct libnvme_global_ctx *ctx , const char
*owner );

like, that's it. two args. very demure, very mindful.

---

🌸 ARGUMENTS 🌸

• ctx — your struct libnvme_global_ctx object 🫶 that's the whole vibe, the context everything hangs off of
• owner — the Orchestrator identity string, fr fr (e.g. "stas", "nbft").

---

💖 DESCRIPTION 💖

So what does it actually do. It records the orchestrator identity used when claiming registry ownership of connections made through ctx. No cap.

A later call overwrites the previous value — so the previous one is gone, periodt. Treating the identity as immutable? That's a policy decision left to the caller, not something this function enforces. You want it locked, you lock it. 💅

And a process that doesn't participate in the registry just… never calls this. iconic behaviour honestly. not every process needs to be a main character here.

the way that this is the supported way to record the registry owner — and libnvme_create_global_ctx deliberately takes no owner parameter. so don't go looking for it there, bestie, it's not there and that was on purpose 💀

see also, iykyk: libnvmf_get_owner_from_tid and libnvmf_get_owner_from_fctx (in fabrics.h) to look up who owns a candidate connection before connecting it, and libnvmf_registry_retrieve (in registry.h) to read a live controller's registry entry directly. 🌈

---

👑 RETURN 👑

• 0 on success ✨
• -EINVAL or -ENOMEM on error 😭

---

and that's the whole page, bestie — no crumbs left behind 🧚

🦋 cute little sign-off: this one's from the libnvme API manual, libnvme_set_owner page, dated October 2026. go forth and own the registry, queen 🎀⭐💜