From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from firstgate.proxmox.com (firstgate.proxmox.com [212.224.123.68]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits)) (No client certificate requested) by lists.proxmox.com (Postfix) with ESMTPS id DC8AB73780 for ; Fri, 18 Jun 2021 11:29:18 +0200 (CEST) Received: from firstgate.proxmox.com (localhost [127.0.0.1]) by firstgate.proxmox.com (Proxmox) with ESMTP id C92742558A for ; Fri, 18 Jun 2021 11:29:18 +0200 (CEST) Received: from proxmox-new.maurer-it.com (proxmox-new.maurer-it.com [94.136.29.106]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by firstgate.proxmox.com (Proxmox) with ESMTPS id A3FB725576 for ; Fri, 18 Jun 2021 11:29:17 +0200 (CEST) Received: from proxmox-new.maurer-it.com (localhost.localdomain [127.0.0.1]) by proxmox-new.maurer-it.com (Proxmox) with ESMTP id 6F99B44153 for ; Fri, 18 Jun 2021 11:29:11 +0200 (CEST) From: Dominik Csapak To: pbs-devel@lists.proxmox.com Date: Fri, 18 Jun 2021 11:29:09 +0200 Message-Id: <20210618092910.4518-1-d.csapak@proxmox.com> X-Mailer: git-send-email 2.20.1 MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-SPAM-LEVEL: Spam detection results: 0 AWL 0.877 Adjusted score from AWL reputation of From: address BAYES_00 -1.9 Bayes spam probability is 0 to 1% KAM_DMARC_STATUS 0.01 Test Rule for DKIM or SPF Failure with Strict Alignment SPF_HELO_NONE 0.001 SPF: HELO does not publish an SPF Record SPF_PASS -0.001 SPF: sender matches SPF record URIBL_BLOCKED 0.001 ADMINISTRATOR NOTICE: The query to URIBL was blocked. See http://wiki.apache.org/spamassassin/DnsBlocklists#dnsbl-block for more information. [datastore.rs, verify.rs] Subject: [pbs-devel] [PATCH proxmox-backup 1/2] backup/datastore: refactor chunk inode sorting to the datastore X-BeenThere: pbs-devel@lists.proxmox.com X-Mailman-Version: 2.1.29 Precedence: list List-Id: Proxmox Backup Server development discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 18 Jun 2021 09:29:18 -0000 so that we can reuse that information the removal of the adding to the corrupted list is ok, since 'get_chunks_in_order' returns them at the end of the list and we do the same if the loading fails later in 'verify_index_chunks' so we still mark them corrupt (assuming that the load will fail if the stat does) Signed-off-by: Dominik Csapak --- alternatively, we could return 2 lists, or check after the loop for the u64::MAX value, but imho it's ok this way src/backup/datastore.rs | 38 ++++++++++++++++++++++++++++++++++++++ src/backup/verify.rs | 38 +++++++------------------------------- 2 files changed, 45 insertions(+), 31 deletions(-) diff --git a/src/backup/datastore.rs b/src/backup/datastore.rs index a0cf50b2..116d2441 100644 --- a/src/backup/datastore.rs +++ b/src/backup/datastore.rs @@ -825,4 +825,42 @@ impl DataStore { pub fn verify_new(&self) -> bool { self.verify_new } + + /// returns a list of chunks sorted by their inode number on disk + /// chunks that could not be stat'ed are at the end of the list + pub fn get_chunks_in_order( + &self, + index: &Box, + skip_chunk: F, + check_abort: A, + ) -> Result, Error> + where + F: Fn(&[u8; 32]) -> bool, + A: Fn(usize) -> Result<(), Error>, + { + let index_count = index.index_count(); + let mut chunk_list = Vec::with_capacity(index_count); + use std::os::unix::fs::MetadataExt; + for pos in 0..index_count { + check_abort(pos)?; + + let info = index.chunk_info(pos).unwrap(); + + if skip_chunk(&info.digest) { + continue; + } + + let ino = match self.stat_chunk(&info.digest) { + Err(_) => u64::MAX, // could not stat, move to end of list + Ok(metadata) => metadata.ino(), + }; + + chunk_list.push((pos, ino)); + } + + // sorting by inode improves data locality, which makes it lots faster on spinners + chunk_list.sort_unstable_by(|(_, ino_a), (_, ino_b)| ino_a.cmp(&ino_b)); + + Ok(chunk_list) + } } diff --git a/src/backup/verify.rs b/src/backup/verify.rs index a1b1e6dd..74b0cfe7 100644 --- a/src/backup/verify.rs +++ b/src/backup/verify.rs @@ -179,42 +179,18 @@ fn verify_index_chunks( } }; - let index_count = index.index_count(); - let mut chunk_list = Vec::with_capacity(index_count); - - use std::os::unix::fs::MetadataExt; - - for pos in 0..index_count { + let check_abort = |pos: usize| -> Result<(), Error> { if pos & 1023 == 0 { verify_worker.worker.check_abort()?; crate::tools::fail_on_shutdown()?; } + Ok(()) + }; - let info = index.chunk_info(pos).unwrap(); - - if skip_chunk(&info.digest) { - continue; // already verified or marked corrupt - } - - match verify_worker.datastore.stat_chunk(&info.digest) { - Err(err) => { - verify_worker.corrupt_chunks.lock().unwrap().insert(info.digest); - task_log!(verify_worker.worker, "can't verify chunk, stat failed - {}", err); - errors.fetch_add(1, Ordering::SeqCst); - rename_corrupted_chunk( - verify_worker.datastore.clone(), - &info.digest, - &verify_worker.worker, - ); - } - Ok(metadata) => { - chunk_list.push((pos, metadata.ino())); - } - } - } - - // sorting by inode improves data locality, which makes it lots faster on spinners - chunk_list.sort_unstable_by(|(_, ino_a), (_, ino_b)| ino_a.cmp(&ino_b)); + let chunk_list = + verify_worker + .datastore + .get_chunks_in_order(&index, skip_chunk, check_abort)?; for (pos, _) in chunk_list { verify_worker.worker.check_abort()?; -- 2.20.1