pageserver: guard against WAL gaps in the interpreted protocol (#10858)

## Problem

The interpreted SK <-> PS protocol does not guard against gaps (neither
does the Vanilla one, but that's beside the point).

## Summary of changes

Extend the protocol to include the start LSN of the PG WAL section from
which the records were interpreted.
Validation is enabled via a config flag on the pageserver and works as
follows:

**Case 1**: `raw_wal_start_lsn` is smaller than the requested LSN
There can't be gaps here, but we check that the shard received records
which it hasn't seen before.

**Case 2**: `raw_wal_start_lsn` is equal to the requested LSN
This is the happy case. No gap and nothing to check

**Case 3**: `raw_wal_start_lsn` is greater than the requested LSN
This is a gap.

To make Case 3 work I had to bend the protocol a bit.
We read record chunks of WAL which aren't record aligned and feed them
to the decoder.
The picture below shows a shard which subscribes at a position somewhere
within Record 2.
We already have a wal reader which is below that position so we wait to
catch up.
We read some wal in Read 1 (all of Record 1 and some of Record 2). The
new shard doesn't
need Record 1 (it has already processed it according to the starting
position), but we read
past it's starting position. When we do Read 2, we decode Record 2 and
ship it off to the shard,
but the starting position of Read 2 is greater than the starting
position the shard requested.
This looks like a gap.


![image](https://github.com/user-attachments/assets/8aed292e-5d62-46a3-9b01-fbf9dc25efe0)

To make it work, we extend the protocol to send an empty
`InterpretedWalRecords` to shards
if the WAL the records originated from ends the requested start
position. On the pageserver,
that just updates the tracking LSNs in memory (no-op really). This gives
us a workaround for
the fake gap.

As a drive by, make `InterpretedWalRecords::next_record_lsn` mandatory
in the application level definition.
It's always included.

Related: neondatabase/cloud#23935

libs/pageserver_api/src/config.rs

@@ -122,6 +122,8 @@ pub struct ConfigToml {
     pub page_service_pipelining: PageServicePipeliningConfig,
     pub get_vectored_concurrent_io: GetVectoredConcurrentIo,
     pub enable_read_path_debugging: Option<bool>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub validate_wal_contiguity: Option<bool>,
 }
 
 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -521,6 +523,7 @@ impl Default for ConfigToml {
             } else {
                 None
             },
+            validate_wal_contiguity: None,
         }
     }
 }

libs/wal_decoder/proto/interpreted_wal.proto

@@ -5,6 +5,7 @@ package interpreted_wal;
 message InterpretedWalRecords {
   repeated InterpretedWalRecord records = 1;
   optional uint64 next_record_lsn = 2;
+  optional uint64 raw_wal_start_lsn = 3;
 }
 
 message InterpretedWalRecord {

libs/wal_decoder/src/models.rs

@@ -60,7 +60,11 @@ pub struct InterpretedWalRecords {
     pub records: Vec<InterpretedWalRecord>,
     // Start LSN of the next record after the batch.
     // Note that said record may not belong to the current shard.
-    pub next_record_lsn: Option<Lsn>,
+    pub next_record_lsn: Lsn,
+    // Inclusive start LSN of the PG WAL from which the interpreted
+    // WAL records were extracted. Note that this is not necessarily the
+    // start LSN of the first interpreted record in the batch.
+    pub raw_wal_start_lsn: Option<Lsn>,
 }
 
 /// An interpreted Postgres WAL record, ready to be handled by the pageserver

libs/wal_decoder/src/wire_format.rs

@@ -167,7 +167,8 @@ impl TryFrom<InterpretedWalRecords> for proto::InterpretedWalRecords {
             .collect::<Result<Vec<_>, _>>()?;
         Ok(proto::InterpretedWalRecords {
             records,
-            next_record_lsn: value.next_record_lsn.map(|l| l.0),
+            next_record_lsn: Some(value.next_record_lsn.0),
+            raw_wal_start_lsn: value.raw_wal_start_lsn.map(|l| l.0),
         })
     }
 }
@@ -254,7 +255,11 @@ impl TryFrom<proto::InterpretedWalRecords> for InterpretedWalRecords {
 
         Ok(InterpretedWalRecords {
             records,
-            next_record_lsn: value.next_record_lsn.map(Lsn::from),
+            next_record_lsn: value
+                .next_record_lsn
+                .map(Lsn::from)
+                .expect("Always provided"),
+            raw_wal_start_lsn: value.raw_wal_start_lsn.map(Lsn::from),
         })
     }
 }

pageserver/src/bin/pageserver.rs

@@ -134,6 +134,7 @@ fn main() -> anyhow::Result<()> {
     info!(?conf.virtual_file_io_engine, "starting with virtual_file IO engine");
     info!(?conf.virtual_file_io_mode, "starting with virtual_file IO mode");
     info!(?conf.wal_receiver_protocol, "starting with WAL receiver protocol");
+    info!(?conf.validate_wal_contiguity, "starting with WAL contiguity validation");
     info!(?conf.page_service_pipelining, "starting with page service pipelining config");
     info!(?conf.get_vectored_concurrent_io, "starting with get_vectored IO concurrency config");
 

pageserver/src/config.rs

@@ -197,6 +197,10 @@ pub struct PageServerConf {
     /// Enable read path debugging. If enabled, read key errors will print a backtrace of the layer
     /// files read.
     pub enable_read_path_debugging: bool,
+
+    /// Interpreted protocol feature: if enabled, validate that the logical WAL received from
+    /// safekeepers does not have gaps.
+    pub validate_wal_contiguity: bool,
 }
 
 /// Token for authentication to safekeepers
@@ -360,6 +364,7 @@ impl PageServerConf {
             page_service_pipelining,
             get_vectored_concurrent_io,
             enable_read_path_debugging,
+            validate_wal_contiguity,
         } = config_toml;
 
         let mut conf = PageServerConf {
@@ -446,6 +451,7 @@ impl PageServerConf {
             virtual_file_io_mode: virtual_file_io_mode.unwrap_or(virtual_file::IoMode::preferred()),
             no_sync: no_sync.unwrap_or(false),
             enable_read_path_debugging: enable_read_path_debugging.unwrap_or(false),
+            validate_wal_contiguity: validate_wal_contiguity.unwrap_or(false),
         };
 
         // ------------------------------------------------------------

pageserver/src/tenant/timeline.rs

@@ -2874,6 +2874,7 @@ impl Timeline {
                 auth_token: crate::config::SAFEKEEPER_AUTH_TOKEN.get().cloned(),
                 availability_zone: self.conf.availability_zone.clone(),
                 ingest_batch_size: self.conf.ingest_batch_size,
+                validate_wal_contiguity: self.conf.validate_wal_contiguity,
             },
             broker_client,
             ctx,

pageserver/src/tenant/timeline/walreceiver.rs

@@ -56,6 +56,7 @@ pub struct WalReceiverConf {
     pub auth_token: Option<Arc<String>>,
     pub availability_zone: Option<String>,
     pub ingest_batch_size: u64,
+    pub validate_wal_contiguity: bool,
 }
 
 pub struct WalReceiver {

pageserver/src/tenant/timeline/walreceiver/connection_manager.rs

@@ -537,6 +537,7 @@ impl ConnectionManagerState {
         let connect_timeout = self.conf.wal_connect_timeout;
         let ingest_batch_size = self.conf.ingest_batch_size;
         let protocol = self.conf.protocol;
+        let validate_wal_contiguity = self.conf.validate_wal_contiguity;
         let timeline = Arc::clone(&self.timeline);
         let ctx = ctx.detached_child(
             TaskKind::WalReceiverConnectionHandler,
@@ -558,6 +559,7 @@ impl ConnectionManagerState {
                     ctx,
                     node_id,
                     ingest_batch_size,
+                    validate_wal_contiguity,
                 )
                 .await;
 
@@ -1563,6 +1565,7 @@ mod tests {
                 auth_token: None,
                 availability_zone: None,
                 ingest_batch_size: 1,
+                validate_wal_contiguity: false,
             },
             wal_connection: None,
             wal_stream_candidates: HashMap::new(),

pageserver/src/tenant/timeline/walreceiver/walreceiver_connection.rs

@@ -120,6 +120,7 @@ pub(super) async fn handle_walreceiver_connection(
     ctx: RequestContext,
     safekeeper_node: NodeId,
     ingest_batch_size: u64,
+    validate_wal_contiguity: bool,
 ) -> Result<(), WalReceiverError> {
     debug_assert_current_span_has_tenant_and_timeline_id();
 
@@ -274,6 +275,7 @@ pub(super) async fn handle_walreceiver_connection(
         } => Some((format, compression)),
     };
 
+    let mut expected_wal_start = startpoint;
     while let Some(replication_message) = {
         select! {
             _ = cancellation.cancelled() => {
@@ -340,13 +342,49 @@ pub(super) async fn handle_walreceiver_connection(
                     )
                     })?;
 
+                // Guard against WAL gaps. If the start LSN of the PG WAL section
+                // from which the interpreted records were extracted, doesn't match
+                // the end of the previous batch (or the starting point for the first batch),
+                // then kill this WAL receiver connection and start a new one.
+                if validate_wal_contiguity {
+                    if let Some(raw_wal_start_lsn) = batch.raw_wal_start_lsn {
+                        match raw_wal_start_lsn.cmp(&expected_wal_start) {
+                            std::cmp::Ordering::Greater => {
+                                let msg = format!(
+                                    "Gap in streamed WAL: [{}, {})",
+                                    expected_wal_start, raw_wal_start_lsn
+                                );
+                                critical!("{msg}");
+                                return Err(WalReceiverError::Other(anyhow!(msg)));
+                            }
+                            std::cmp::Ordering::Less => {
+                                // Other shards are reading WAL behind us.
+                                // This is valid, but check that we received records
+                                // that we haven't seen before.
+                                if let Some(first_rec) = batch.records.first() {
+                                    if first_rec.next_record_lsn < last_rec_lsn {
+                                        let msg = format!(
+                                            "Received record with next_record_lsn multiple times ({} < {})",
+                                            first_rec.next_record_lsn, expected_wal_start
+                                        );
+                                        critical!("{msg}");
+                                        return Err(WalReceiverError::Other(anyhow!(msg)));
+                                    }
+                                }
+                            }
+                            std::cmp::Ordering::Equal => {}
+                        }
+                    }
+                }
+
                 let InterpretedWalRecords {
                     records,
                     next_record_lsn,
+                    raw_wal_start_lsn: _,
                 } = batch;
 
                 tracing::debug!(
-                    "Received WAL up to {} with next_record_lsn={:?}",
+                    "Received WAL up to {} with next_record_lsn={}",
                     streaming_lsn,
                     next_record_lsn
                 );
@@ -423,12 +461,11 @@ pub(super) async fn handle_walreceiver_connection(
                 // need to advance last record LSN on all shards. If we've not ingested the latest
                 // record, then set the LSN of the modification past it. This way all shards
                 // advance their last record LSN at the same time.
-                let needs_last_record_lsn_advance = match next_record_lsn {
-                    Some(lsn) if lsn > modification.get_lsn() => {
-                        modification.set_lsn(lsn).unwrap();
-                        true
-                    }
-                    _ => false,
+                let needs_last_record_lsn_advance = if next_record_lsn > modification.get_lsn() {
+                    modification.set_lsn(next_record_lsn).unwrap();
+                    true
+                } else {
+                    false
                 };
 
                 if uncommitted_records > 0 || needs_last_record_lsn_advance {
@@ -446,9 +483,8 @@ pub(super) async fn handle_walreceiver_connection(
                     timeline.get_last_record_lsn()
                 );
 
-                if let Some(lsn) = next_record_lsn {
-                    last_rec_lsn = lsn;
-                }
+                last_rec_lsn = next_record_lsn;
+                expected_wal_start = streaming_lsn;
 
                 Some(streaming_lsn)
             }

safekeeper/src/send_interpreted_wal.rs

@@ -295,14 +295,18 @@ impl InterpretedWalReader {
 
         let mut wal_decoder = WalStreamDecoder::new(start_pos, self.pg_version);
 
+        // Tracks the start of the PG WAL LSN from which the current batch of
+        // interpreted records originated.
+        let mut current_batch_wal_start_lsn: Option<Lsn> = None;
+
         loop {
             tokio::select! {
                 // Main branch for reading WAL and forwarding it
                 wal_or_reset = self.wal_stream.next() => {
                     let wal = wal_or_reset.map(|wor| wor.get_wal().expect("reset handled in select branch below"));
                     let WalBytes {
                         wal,
-                        wal_start_lsn: _,
+                        wal_start_lsn,
                         wal_end_lsn,
                         available_wal_end_lsn,
                     } = match wal {
@@ -315,6 +319,12 @@ impl InterpretedWalReader {
                         }
                     };
 
+                    // We will already have a value if the previous chunks of WAL
+                    // did not decode into anything useful.
+                    if current_batch_wal_start_lsn.is_none() {
+                        current_batch_wal_start_lsn = Some(wal_start_lsn);
+                    }
+
                     wal_decoder.feed_bytes(&wal);
 
                     // Deserialize and interpret WAL records from this batch of WAL.
@@ -363,7 +373,9 @@ impl InterpretedWalReader {
 
                     let max_next_record_lsn = match max_next_record_lsn {
                         Some(lsn) => lsn,
-                        None => { continue; }
+                        None => {
+                            continue;
+                        }
                     };
 
                     // Update the current position such that new receivers can decide
@@ -377,21 +389,38 @@ impl InterpretedWalReader {
                         }
                     }
 
+                    let batch_wal_start_lsn = current_batch_wal_start_lsn.take().unwrap();
+
                     // Send interpreted records downstream. Anything that has already been seen
                     // by a shard is filtered out.
                     let mut shard_senders_to_remove = Vec::new();
                     for (shard, states) in &mut self.shard_senders {
                         for state in states {
-                            if max_next_record_lsn <= state.next_record_lsn {
-                                continue;
-                            }
-
                             let shard_sender_id = ShardSenderId::new(*shard, state.sender_id);
-                            let records = records_by_sender.remove(&shard_sender_id).unwrap_or_default();
 
-                            let batch = InterpretedWalRecords {
-                                records,
-                                next_record_lsn: Some(max_next_record_lsn),
+                            let batch = if max_next_record_lsn > state.next_record_lsn {
+                                // This batch contains at least one record that this shard has not
+                                // seen yet.
+                                let records = records_by_sender.remove(&shard_sender_id).unwrap_or_default();
+
+                                InterpretedWalRecords {
+                                    records,
+                                    next_record_lsn: max_next_record_lsn,
+                                    raw_wal_start_lsn: Some(batch_wal_start_lsn),
+                                }
+                            } else if wal_end_lsn > state.next_record_lsn {
+                                // All the records in this batch were seen by the shard
+                                // However, the batch maps to a chunk of WAL that the
+                                // shard has not yet seen. Notify it of the start LSN
+                                // of the PG WAL chunk such that it doesn't look like a gap.
+                                InterpretedWalRecords {
+                                    records: Vec::default(),
+                                    next_record_lsn: state.next_record_lsn,
+                                    raw_wal_start_lsn: Some(batch_wal_start_lsn),
+                                }
+                            } else {
+                                // The shard has seen this chunk of WAL before. Skip it.
+                                continue;
                             };
 
                             let res = state.tx.send(Batch {
@@ -403,7 +432,7 @@ impl InterpretedWalReader {
                             if res.is_err() {
                                 shard_senders_to_remove.push(shard_sender_id);
                             } else {
-                                state.next_record_lsn = max_next_record_lsn;
+                                state.next_record_lsn = std::cmp::max(state.next_record_lsn, max_next_record_lsn);
                             }
                         }
                     }

test_runner/fixtures/neon_fixtures.py

@@ -1167,15 +1167,15 @@ def __init__(self, config: NeonEnvBuilder):
                 "max_batch_size": 32,
             }
 
-            # Concurrent IO (https://github.com/neondatabase/neon/issues/9378):
-            # enable concurrent IO by default in tests and benchmarks.
-            # Compat tests are exempt because old versions fail to parse the new config.
-            get_vectored_concurrent_io = self.pageserver_get_vectored_concurrent_io
             if config.test_may_use_compatibility_snapshot_binaries:
                 log.info(
-                    "Forcing use of binary-built-in default to avoid forward-compatibility related test failures"
+                    "Skipping WAL contiguity validation to avoid forward-compatibility related test failures"
                 )
-                get_vectored_concurrent_io = None
+            else:
+                # Look for gaps in WAL received from safekeepeers
+                ps_cfg["validate_wal_contiguity"] = True
+
+            get_vectored_concurrent_io = self.pageserver_get_vectored_concurrent_io
             if get_vectored_concurrent_io is not None:
                 ps_cfg["get_vectored_concurrent_io"] = {
                     "mode": self.pageserver_get_vectored_concurrent_io,