Skip to content

Conversation

Keno
Copy link
Member

@Keno Keno commented Aug 9, 2025

As discussed in #3717, ZFS has known bugs around SEEK_HOLE/SEEK_DATA causing it to indicate spurious holes. We detect the case where it (inconsistently) ends up indicating the hole to be zero sized, but depending on the sequencing of events, the related upstream issues indicate that the file system could indicate a non-zero-sized spurious hole on these versions. To avoid trace corruption on such file systems, we must skip our SEEK_HOLE/SEEK_DATA optimizations.

As discussed in #3717, ZFS
has known bugs around SEEK_HOLE/SEEK_DATA causing it to indicate
spurious holes. We detect the case where it (inconsistently) ends
up indicating the hole to be zero sized, but depending on the
sequencing of events, the related upstream issues indicate that
the file system could indicate a non-zero-sized spurious hole on
these versions. To avoid trace corruption on such file systems, we
must skip our SEEK_HOLE/SEEK_DATA optimizations.
@Keno Keno changed the title Skip hold detection on broken ZFS filesystems Skip hole detection on broken ZFS filesystems Aug 9, 2025
Keno added a commit to JuliaPackaging/Yggdrasil that referenced this pull request Aug 9, 2025
Keno added a commit to JuliaPackaging/Yggdrasil that referenced this pull request Aug 11, 2025
@rocallahan rocallahan merged commit c9349cc into master Aug 23, 2025
9 of 10 checks passed
@rocallahan
Copy link
Collaborator

Thanks!!! Sorry about the delay. What a hideous bug.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants