public inbox for pbs-devel@lists.proxmox.com
 help / color / mirror / Atom feed
From: "Fabian Grünbichler" <f.gruenbichler@proxmox.com>
To: pbs-devel@lists.proxmox.com
Subject: [pbs-devel] [PATCH proxmox-backup v3 3/3] GC: S3: phase2: do not force delete for every list iteration
Date: Fri, 21 Nov 2025 13:38:25 +0100	[thread overview]
Message-ID: <20251121123829.701196-4-f.gruenbichler@proxmox.com> (raw)
In-Reply-To: <20251121123829.701196-1-f.gruenbichler@proxmox.com>

force delete if the first pending delete was inserted more than 5
minutes ago, and at the very end after the last iteration, instead of
after every processing every batch of 1000 listed objects. this reduces
the number of delete calls made to the backend, making regular garbage
collections that do not delete most objects cheaper, but means holding
the flocks for garbage chunks/objects longer.

Suggested-by: Christian Ebner <c.ebner@proxmox.com>
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
---

Notes:
    v1->v2: reworked to use age-based cutoff
    v2->v3:
    - prevent delete calls for empty lists (thanks Chris)
    - reset age correctly on first insert into delete list (thanks Chris)

 pbs-datastore/src/datastore.rs | 27 +++++++++++++++++++--------
 1 file changed, 19 insertions(+), 8 deletions(-)

diff --git a/pbs-datastore/src/datastore.rs b/pbs-datastore/src/datastore.rs
index e9d6b46f3..bc6f6b3ba 100644
--- a/pbs-datastore/src/datastore.rs
+++ b/pbs-datastore/src/datastore.rs
@@ -22,7 +22,7 @@ use proxmox_sys::error::SysError;
 use proxmox_sys::fs::{file_read_optional_string, replace_file, CreateOptions};
 use proxmox_sys::linux::procfs::MountInfo;
 use proxmox_sys::process_locker::{ProcessLockExclusiveGuard, ProcessLockSharedGuard};
-use proxmox_time::TimeSpan;
+use proxmox_time::{epoch_i64, TimeSpan};
 use proxmox_worker_task::WorkerTaskContext;
 
 use pbs_api_types::{
@@ -60,6 +60,8 @@ const NAMESPACE_MARKER_FILENAME: &str = ".namespace";
 const CHUNK_LOCK_TIMEOUT: Duration = Duration::from_secs(3 * 60 * 60);
 // s3 deletion batch size to avoid 1024 open files soft limit
 const S3_DELETE_BATCH_LIMIT: usize = 100;
+// max defer time for s3 batch deletions
+const S3_DELETE_DEFER_LIMIT_SECONDS: i64 = 60 * 5;
 
 /// checks if auth_id is owner, or, if owner is a token, if
 /// auth_id is the user of the token
@@ -1660,6 +1662,7 @@ impl DataStore {
                     .context("failed to list chunk in s3 object store")?;
 
             let mut delete_list = Vec::with_capacity(S3_DELETE_BATCH_LIMIT);
+            let mut delete_list_age = epoch_i64();
 
             let s3_delete_batch = |delete_list: &mut Vec<(S3ObjectKey, BackupLockGuard)>,
                                    s3_client: &Arc<S3Client>|
@@ -1730,6 +1733,11 @@ impl DataStore {
                                         std::fs::remove_file(chunk_path)?;
                                     }
                                 }
+
+                                // set age based on first insertion
+                                if delete_list.is_empty() {
+                                    delete_list_age = epoch_i64();
+                                }
                                 delete_list.push((content.key, _chunk_guard));
                                 Ok(())
                             },
@@ -1742,16 +1750,13 @@ impl DataStore {
                     drop(_guard);
 
                     // limit pending deletes to avoid holding too many chunk flocks
-                    if delete_list.len() >= S3_DELETE_BATCH_LIMIT {
+                    if delete_list.len() >= S3_DELETE_BATCH_LIMIT
+                        || (!delete_list.is_empty()
+                            && epoch_i64() - delete_list_age > S3_DELETE_DEFER_LIMIT_SECONDS)
+                    {
                         s3_delete_batch(&mut delete_list, s3_client)?;
                     }
                 }
-
-                // delete the last batch of objects, if there are any remaining
-                if !delete_list.is_empty() {
-                    s3_delete_batch(&mut delete_list, s3_client)?;
-                }
-
                 // Process next batch of chunks if there is more
                 if list_bucket_result.is_truncated {
                     list_bucket_result =
@@ -1764,6 +1769,12 @@ impl DataStore {
 
                 break;
             }
+
+            // delete the last batch of objects, if there are any remaining
+            if !delete_list.is_empty() {
+                s3_delete_batch(&mut delete_list, s3_client)?;
+            }
+
             info!("processed {chunk_count} total chunks");
 
             // Phase 2 GC of Filesystem backed storage is phase 3 for S3 backed GC
-- 
2.47.3



_______________________________________________
pbs-devel mailing list
pbs-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pbs-devel

  parent reply	other threads:[~2025-11-21 12:39 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-21 12:38 [pbs-devel] [PATCH proxmox-backup v3 0/3] reduce GC S3 locking Fabian Grünbichler
2025-11-21 12:38 ` [pbs-devel] [PATCH proxmox-backup v3 1/3] GC: S3: reduce number of open FDs for to-be-deleted objects Fabian Grünbichler
2025-11-21 12:38 ` [pbs-devel] [PATCH proxmox-backup v3 2/3] GC: S3: factor out batch object deletion Fabian Grünbichler
2025-11-21 12:38 ` Fabian Grünbichler [this message]
2025-11-21 14:14 ` [pbs-devel] applied: [PATCH proxmox-backup v3 0/3] reduce GC S3 locking Thomas Lamprecht

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=20251121123829.701196-4-f.gruenbichler@proxmox.com \
    --to=f.gruenbichler@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
Service provided by Proxmox Server Solutions GmbH | Privacy | Legal