NAS-130877 / 25.04 / Remove acltype normalization for datasets #14377
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
bugclerk
changed the title
yocalebo
approved these changes
Aug 30, 2024
anodos325
force-pushed
the
remove-hack-toplevel-acl-inherit
branch
from
af59fe1 to
eee5b35
Compare
When we were doing initial development of TrueNAS SCALE (pre-angelfish) a bug was encountered when importing pools from FreeBSD in which the acltype could be set to OFF. Eventually this bug was fixed on the ZFS side, but it was still early days of testing NFSv4 acltype on SCALE and so we didn't want early-adopter migrators to have broken shares. Thus we normalized creation of new datasets to be POSIX acltype if user had selected to inherit from top-level dataset. After a few releases and user bug reports we realized we needed to have specific ZFS dataset properties set depending on acltype. For example that the aclmode should be DISCARD (discard internal ZFS acl) because changing acltype does not remove the internal ZFS ACL and it could still be evaluated in zfs_zaccess calls. This eventually necessitated raising ValidationErrors on dataset creation for invalid parameter combinations. Unfortunately, the normalization behavior introduced in angelfish testing was overlooked at this point, which could cause a situation in which pools from FreeBSD where acltype now imports as nfsv4acl on top-level datasets would get flagged as needing normalization. This led to an invalid combination of acltype POSIX and aclmode PASSTHROUGH being set on new datasets when they were created to inherit from the top-level dataset. Rather than fixing the normalization, at this point we're pretty confident about NFSv4 ACL type behavior in SCALE and can just remove the normalization logic entirely.
anodos325
force-pushed
the
remove-hack-toplevel-acl-inherit
branch
from
eee5b35 to
da357b8
Compare
|
Unable to perform the following backports: backport-24.04.2.1. |
|
This PR has been merged and conversations have been locked. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When we were doing initial development of TrueNAS SCALE (pre-angelfish) a bug was encountered when importing pools from FreeBSD in which the acltype could be set to OFF. Eventually this bug was fixed on the ZFS side, but it was still early days of testing NFSv4 acltype on SCALE and so we didn't want early-adopter migrators to have broken shares. Thus we normalized creation of new datasets to be POSIX acltype if user had selected to inherit from top-level dataset.
After a few releases and user bug reports we realized we needed to have specific ZFS dataset properties set depending on acltype. For example that the aclmode should be DISCARD (discard internal ZFS acl) because changing acltype does not remove the internal ZFS ACL and it could still be evaluated in zfs_zaccess calls. This eventually necessitated raising ValidationErrors on dataset creation for invalid parameter combinations. Unfortunately, the normalization behavior introduced in angelfish testing was overlooked at this point, which could cause a situation in which pools from FreeBSD where acltype now imports as nfsv4acl on top-level datasets would get flagged as needing normalization. This led to an invalid combination of acltype POSIX and aclmode PASSTHROUGH being set on new datasets when they were created to inherit from the top-level dataset.
Rather than fixing the normalization, at this point we're pretty confident about NFSv4 ACL type behavior in SCALE and can just remove the normalization logic entirely.