mkdir library call causes Overflow

[Forthoney]

Originally posted in MLton, YarwarRaza has provided a simpler C reproduction of the issue and a pointer to a potential cause.

This also happens in C, so it's not a MLton issue.

---

EDIT: To figure out the real issue, this might be a good place to start.

On 64bit systems, even if all overlay layers are not on the same underlying filesystem, the same compliant behavior could be achieved with the “xino” feature. The “xino” feature composes a unique object identifier from the real object st_ino and an underlying fsid number. The “xino” feature uses the high inode number bits for fsid, because the underlying filesystems rarely use the high inode number bits. In case the underlying inode number does overflow into the high xino bits, overlay filesystem will fall back to the non xino behavior for that inode

Originally posted by @YawarRaza7349 in #607

[ezrizhu]

That's interesting, but not entirely surprising as we get EOVERFLOW on some other overlayfs operations that we have incorrectly invoked in the past, e.g., accessing files owned by an unmapped GID.

[ezrizhu]

https://github.com/Forthoney/overflow-reproduction/pull/1 should fix the CI issue. I am not sure how this will apply to your underlying issue, but I'll first start by seeing if any of the dir you're executing your commands in is owned by another group.

For more context: please see #145