* [pbs-devel] applied: [PATCH backup] backup verify: re-check if we can skip a chunk in the actual verify loop
@ 2021-04-15 8:13 Thomas Lamprecht
0 siblings, 0 replies; only message in thread
From: Thomas Lamprecht @ 2021-04-15 8:13 UTC (permalink / raw)
To: pbs-devel
Fixes a non-negligible performance regression from commit
7f394c807bca3f451e77b6a1cf7de7c6e7df5f92
While we skip known-verified chunks in the stat-and-inode-sort loop,
those are only the ones from previous indexes. If there's a repeated
chunk in one index they would get re-verified more often as required.
So, add the check again explicitly to the read+verify loop.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
src/backup/verify.rs | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/src/backup/verify.rs b/src/backup/verify.rs
index ed959813..1192821e 100644
--- a/src/backup/verify.rs
+++ b/src/backup/verify.rs
@@ -214,6 +214,12 @@ fn verify_index_chunks(
let info = index.chunk_info(pos).unwrap();
+ // we must always recheck this here, the parallel worker below alter it!
+ // Else we miss skipping repeated chunks from the same index, and re-verify them all
+ if verify_worker.verified_chunks.lock().unwrap().contains(&info.digest) {
+ continue; // already verified
+ }
+
match verify_worker.datastore.load_chunk(&info.digest) {
Err(err) => {
verify_worker.corrupt_chunks.lock().unwrap().insert(info.digest);
--
2.29.2
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2021-04-15 8:13 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-04-15 8:13 [pbs-devel] applied: [PATCH backup] backup verify: re-check if we can skip a chunk in the actual verify loop Thomas Lamprecht
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox