From: "Fabian Grünbichler" <f.gruenbichler@proxmox.com>
To: Christian Ebner <c.ebner@proxmox.com>, pbs-devel@lists.proxmox.com
Subject: Re: [pbs-devel] [PATCH proxmox-backup 4/4] chunk store: return chunk extension and check for used marker
Date: Wed, 10 Dec 2025 15:21:58 +0100 [thread overview]
Message-ID: <176537651811.371273.12546243384711800815@yuna.proxmox.com> (raw)
In-Reply-To: <20251126133419.570874-5-c.ebner@proxmox.com>
Quoting Christian Ebner (2025-11-26 14:34:19)
> Clearly distinguish the cases for `bad` and `using` extensions for
> items returned by the chunk iterator. Directory entries with
> filenames which matched the expected length but not the extension
> are now skipped over.
I think we could go a step further and also move the parsing/rendering to
ChunkExt everywhere we currently do it? which I guess is mostly
verification/GC? not sure whether ChunkExt or a new Chunk struct that also
contains the digest would be a better fit though..
>
> Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
> ---
> pbs-datastore/src/chunk_store.rs | 48 ++++++++++++++++++++++++--------
> 1 file changed, 37 insertions(+), 11 deletions(-)
>
> diff --git a/pbs-datastore/src/chunk_store.rs b/pbs-datastore/src/chunk_store.rs
> index 7980938ad..ad517391d 100644
> --- a/pbs-datastore/src/chunk_store.rs
> +++ b/pbs-datastore/src/chunk_store.rs
> @@ -280,7 +280,11 @@ impl ChunkStore {
> &self,
> ) -> Result<
> impl std::iter::FusedIterator<
> - Item = (Result<proxmox_sys::fs::ReadDirEntry, Error>, usize, bool),
> + Item = (
> + Result<proxmox_sys::fs::ReadDirEntry, Error>,
> + usize,
> + ChunkExt,
> + ),
> >,
> Error,
> > {
> @@ -318,14 +322,20 @@ impl ChunkStore {
>
> if bytes.len() == 64 && bytes.iter().take(64).all(u8::is_ascii_hexdigit)
> {
> - return Some((Ok(entry), percentage, false));
> + return Some((Ok(entry), percentage, ChunkExt::None));
> }
>
> if bytes.len() == 64 + ".0.bad".len()
> && bytes.iter().take(64).all(u8::is_ascii_hexdigit)
> {
> - let bad = bytes.ends_with(b".bad");
> - return Some((Ok(entry), percentage, bad));
> + let chunk_ext = if bytes.ends_with(b".bad") {
> + ChunkExt::Bad
> + } else if bytes.ends_with(USING_MARKER_FILENAME_EXT.as_bytes()) {
> + ChunkExt::UsedMarker
> + } else {
> + continue;
> + };
> + return Some((Ok(entry), percentage, chunk_ext));
> }
>
> continue;
> @@ -334,7 +344,7 @@ impl ChunkStore {
> // stop after first error
> done = true;
> // and pass the error through:
> - return Some((Err(err), percentage, false));
> + return Some((Err(err), percentage, ChunkExt::None));
> }
> None => (), // open next directory
> }
> @@ -367,7 +377,7 @@ impl ChunkStore {
> return Some((
> Err(format_err!("unable to read subdir '{subdir}' - {err}")),
> percentage,
> - false,
> + ChunkExt::None,
> ));
> }
> }
> @@ -402,7 +412,8 @@ impl ChunkStore {
> let mut last_percentage = 0;
> let mut chunk_count = 0;
>
> - for (entry, percentage, bad) in self.get_chunk_store_iterator()? {
> + for (entry, percentage, chunk_ext) in self.get_chunk_store_iterator()? {
> + let bad = chunk_ext.is_bad();
> if last_percentage != percentage {
> last_percentage = percentage;
> info!("processed {percentage}% ({chunk_count} chunks)");
> @@ -433,10 +444,7 @@ impl ChunkStore {
> drop(lock);
> continue;
> }
> - if filename
> - .to_bytes()
> - .ends_with(USING_MARKER_FILENAME_EXT.as_bytes())
> - {
> + if chunk_ext.is_used_marker() {
> unlinkat(Some(dirfd), filename, UnlinkatFlags::NoRemoveDir).map_err(|err| {
> format_err!("unlinking chunk using marker {filename:?} failed - {err}")
> })?;
> @@ -908,6 +916,24 @@ impl ChunkStore {
> }
> }
>
> +#[derive(PartialEq)]
> +/// Chunk iterator directory entry filename extension
> +enum ChunkExt {
> + None,
> + Bad,
> + UsedMarker,
> +}
> +
> +impl ChunkExt {
> + fn is_bad(&self) -> bool {
> + *self == Self::Bad
> + }
> +
> + fn is_used_marker(&self) -> bool {
> + *self == Self::UsedMarker
> + }
> +}
> +
> #[test]
> fn test_chunk_store1() {
> let mut path = std::fs::canonicalize(".").unwrap(); // we need absolute path
> --
> 2.47.3
>
>
>
> _______________________________________________
> pbs-devel mailing list
> pbs-devel@lists.proxmox.com
> https://lists.proxmox.com/cgi-bin/mailman/listinfo/pbs-devel
>
>
_______________________________________________
pbs-devel mailing list
pbs-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pbs-devel
next prev parent reply other threads:[~2025-12-10 14:21 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-26 13:34 [pbs-devel] [PATCH proxmox-backup 0/4] followups for garbage collection Christian Ebner
2025-11-26 13:34 ` [pbs-devel] [PATCH proxmox-backup 1/4] GC: Move S3 delete list state and logic to a dedicated struct Christian Ebner
2025-11-26 13:34 ` [pbs-devel] [PATCH proxmox-backup 2/4] chunk store: rename and limit scope for chunk store iterator Christian Ebner
2025-11-26 13:34 ` [pbs-devel] [PATCH proxmox-backup 3/4] chunk store: invert chunk filename checks in " Christian Ebner
2025-11-26 13:34 ` [pbs-devel] [PATCH proxmox-backup 4/4] chunk store: return chunk extension and check for used marker Christian Ebner
2025-12-10 14:21 ` Fabian Grünbichler [this message]
2025-12-11 11:28 ` Christian Ebner
2025-12-10 14:18 ` [pbs-devel] [PATCH proxmox-backup 0/4] followups for garbage collection Fabian Grünbichler
2025-12-11 15:39 ` [pbs-devel] superseded: " Christian Ebner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=176537651811.371273.12546243384711800815@yuna.proxmox.com \
--to=f.gruenbichler@proxmox.com \
--cc=c.ebner@proxmox.com \
--cc=pbs-devel@lists.proxmox.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox