rust/neon
1
0
Fork 0
mirror of https://github.com/neondatabase/neon.git synced 2026-01-13 16:32:56 +00:00
Code Issues Packages Projects Releases Wiki Activity

pageserver: make compaction walredo errors critical (#10884)

Browse Source 
Mark walredo errors as critical too.

Also pull the pattern matching out into the outer `match`.

Follows #10872.
This commit is contained in:
Erik Grinaker
2025-02-20 13:08:54 +01:00
committed by GitHub
parent f7edcf12e3
commit 07bee60037
2 changed files with 20 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
pageserver/src/tenant/timeline.rs
View File

@@ -5344,12 +5344,6 @@ impl From<OffloadError> for CompactionError {
}
}
impl CompactionError {
pub fn is_cancelled(&self) -> bool {
matches!(self, CompactionError::ShuttingDown)
}
}
impl From<CollectKeySpaceError> for CompactionError {
fn from(err: CollectKeySpaceError) -> Self {
match err {

42
pageserver/src/tenant/timeline/compaction.rs
View File

@@ -26,7 +26,7 @@ use pageserver_api::models::CompactInfoResponse;
use pageserver_api::shard::{ShardCount, ShardIdentity, TenantShardId};
use serde::Serialize;
use tokio_util::sync::CancellationToken;
use tracing::{debug, info, info_span, trace, warn, Instrument};
use tracing::{debug, error, info, info_span, trace, warn, Instrument};
use utils::critical;
use utils::id::TimelineId;
@@ -775,27 +775,25 @@ impl Timeline {
return Ok(CompactionOutcome::YieldForL0);
}
}
Err(err) => {
// no partitioning? This is normal, if the timeline was just created
// as an empty timeline. Also in unit tests, when we use the timeline
// as a simple key-value store, ignoring the datadir layout. Log the
// error but continue.
//
// Suppress error when it's due to cancellation
if !self.cancel.is_cancelled() && !err.is_cancelled() {
if let CompactionError::CollectKeySpaceError(
CollectKeySpaceError::Decode(_)
| CollectKeySpaceError::PageRead(PageReconstructError::MissingKey(_)),
) = err
{
critical!("could not compact, repartitioning keyspace failed: {err:?}");
} else {
tracing::error!(
"could not compact, repartitioning keyspace failed: {err:?}"
);
}
}
}
// Suppress errors when cancelled.
Err(_) if self.cancel.is_cancelled() => {}
Err(CompactionError::ShuttingDown) => {}
// Alert on critical errors that indicate data corruption.
Err(
err @ CompactionError::CollectKeySpaceError(
CollectKeySpaceError::Decode(_)
| CollectKeySpaceError::PageRead(
PageReconstructError::MissingKey(_) | PageReconstructError::WalRedo(_),
),
),
) => critical!("could not compact, repartitioning keyspace failed: {err:?}"),
// Log other errors. No partitioning? This is normal, if the timeline was just created
// as an empty timeline. Also in unit tests, when we use the timeline as a simple
// key-value store, ignoring the datadir layout. Log the error but continue.
Err(err) => error!("could not compact, repartitioning keyspace failed: {err:?}"),
};
let partition_count = self.partitioning.read().0 .0.parts.len();