public inbox for pbs-devel@lists.proxmox.com
 help / color / mirror / Atom feed
From: Dietmar Maurer <dietmar@proxmox.com>
To: pbs-devel@lists.proxmox.com
Subject: [pbs-devel] [PATCH proxmox-backup v3 3/4] tape: media_catalog: add snapshot list cache for catalog
Date: Wed, 28 Jul 2021 11:11:17 +0200	[thread overview]
Message-ID: <20210728091118.581161-4-dietmar@proxmox.com> (raw)
In-Reply-To: <20210728091118.581161-1-dietmar@proxmox.com>

For some parts of the ui, we only need the snapshot list from the catalog,
and reading the whole catalog (can be multiple hundred MiB) is not
really necessary.

Instead, we write the list of snapshots into a seperate .index file. This file
is generated on demand and is much smaller and thus faster to read.
---
 src/tape/media_catalog_cache.rs | 108 ++++++++++++++++++++++++++++++++
 src/tape/mod.rs                 |   3 +
 2 files changed, 111 insertions(+)
 create mode 100644 src/tape/media_catalog_cache.rs

diff --git a/src/tape/media_catalog_cache.rs b/src/tape/media_catalog_cache.rs
new file mode 100644
index 00000000..bf298e65
--- /dev/null
+++ b/src/tape/media_catalog_cache.rs
@@ -0,0 +1,108 @@
+use std::path::Path;
+use std::io::{BufRead, BufReader};
+
+use anyhow::{format_err, bail, Error};
+
+use proxmox::tools::fs::CreateOptions;
+
+use crate::tape::{MediaCatalog, MediaId};
+
+/// Returns a list of (store, snapshot) for a given MediaId
+///
+/// To speedup things for large catalogs, we cache the list of
+/// snapshots into a separate file.
+pub fn media_catalog_snapshot_list(
+    base_path: &Path,
+    media_id: &MediaId,
+) -> Result<Vec<(String, String)>, Error> {
+
+    let uuid = &media_id.label.uuid;
+
+    let mut cache_path = base_path.to_owned();
+    cache_path.push(uuid.to_string());
+    let mut catalog_path = cache_path.clone();
+    cache_path.set_extension("index");
+    catalog_path.set_extension("log");
+
+    let stat = match nix::sys::stat::stat(&catalog_path) {
+        Ok(stat) => stat,
+        Err(err) => bail!("unable to stat media catalog {:?} - {}", catalog_path, err),
+    };
+
+    let cache_id = format!("{:016X}-{:016X}-{:016X}", stat.st_ino, stat.st_size as u64, stat.st_mtime as u64);
+
+    match std::fs::OpenOptions::new().read(true).open(&cache_path) {
+        Ok(file) => {
+            let mut list = Vec::new();
+            let file = BufReader::new(file);
+            let mut lines = file.lines();
+            match lines.next() {
+                Some(Ok(id)) => {
+                    if id != cache_id { // cache is outdated - rewrite
+                        return write_snapshot_cache(base_path, media_id, &cache_path, &cache_id);
+                    }
+                }
+                _ => bail!("unable to read catalog cache firstline {:?}", cache_path),
+            }
+
+            for line in lines {
+                let mut line = line?;
+
+                let idx = line
+                    .find(':')
+                    .ok_or_else(|| format_err!("invalid line format (no store found)"))?;
+
+                let snapshot = line.split_off(idx + 1);
+                line.truncate(idx);
+                list.push((line, snapshot));
+            }
+
+            Ok(list)
+        }
+        Err(err) if err.kind() == std::io::ErrorKind::NotFound => {
+            write_snapshot_cache(base_path, media_id, &cache_path, &cache_id)
+        }
+        Err(err) => bail!("unable to open catalog cache - {}", err),
+    }
+}
+
+fn write_snapshot_cache(
+    base_path: &Path,
+    media_id: &MediaId,
+    cache_path: &Path,
+    cache_id: &str,
+) ->  Result<Vec<(String, String)>, Error> {
+
+    // open normal catalog and write cache
+    let catalog = MediaCatalog::open(base_path, media_id, false, false)?;
+
+    let mut data = String::new();
+    data.push_str(cache_id);
+    data.push('\n');
+
+    let mut list = Vec::new();
+    for (store, content) in catalog.content() {
+        for snapshot in content.snapshot_index.keys() {
+            list.push((store.to_string(), snapshot.to_string()));
+            data.push_str(store);
+            data.push(':');
+            data.push_str(snapshot);
+            data.push('\n');
+        }
+    }
+
+    let backup_user = crate::backup::backup_user()?;
+    let mode = nix::sys::stat::Mode::from_bits_truncate(0o0640);
+    let options = CreateOptions::new()
+        .perm(mode)
+        .owner(backup_user.uid)
+        .group(backup_user.gid);
+
+    proxmox::tools::fs::replace_file(
+        cache_path,
+        data.as_bytes(),
+        options,
+    )?;
+
+    Ok(list)
+}
diff --git a/src/tape/mod.rs b/src/tape/mod.rs
index 8190e141..93c24719 100644
--- a/src/tape/mod.rs
+++ b/src/tape/mod.rs
@@ -42,6 +42,9 @@ pub use media_pool::*;
 mod media_catalog;
 pub use media_catalog::*;
 
+mod media_catalog_cache;
+pub use media_catalog_cache::*;
+
 mod pool_writer;
 pub use pool_writer::*;
 
-- 
2.30.2





  parent reply	other threads:[~2021-07-28  9:11 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-28  9:11 [pbs-devel] [PATCH proxmox-backup v3 0/4] improve catalog handling Dietmar Maurer
2021-07-28  9:11 ` [pbs-devel] [PATCH proxmox-backup v3 1/4] cleanup: factor out tape catalog path helpers Dietmar Maurer
2021-07-28  9:11 ` [pbs-devel] [PATCH proxmox-backup v3 2/4] tape: lock media_catalog file to to get a consistent view with load_catalog Dietmar Maurer
2021-07-28  9:11 ` Dietmar Maurer [this message]
2021-07-28  9:11 ` [pbs-devel] [PATCH proxmox-backup v3 4/4] api2: tape: media: use MediaCatalog::snapshot_list for content listing Dietmar Maurer
2021-07-29 12:30 ` [pbs-devel] applied: [PATCH proxmox-backup v3 0/4] improve catalog handling Dietmar Maurer

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=20210728091118.581161-4-dietmar@proxmox.com \
    --to=dietmar@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