Merge pull request #5375 from neondatabase/releases/2023-09-26

Release 2023-09-26

.github/workflows/build_and_test.yml

@@ -834,7 +834,7 @@ jobs:
       run:
         shell: sh -eu {0}
     env:
-      VM_BUILDER_VERSION: v0.17.10
+      VM_BUILDER_VERSION: v0.17.11
 
     steps:
       - name: Checkout
@@ -1091,8 +1091,9 @@ jobs:
           GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
         run: |
           if [[ "$GITHUB_REF_NAME" == "main" ]]; then
-            gh workflow --repo neondatabase/aws run deploy-dev.yml --ref main -f branch=main -f dockerTag=${{needs.tag.outputs.build-tag}}
+            gh workflow --repo neondatabase/aws run deploy-dev.yml --ref main -f branch=main -f dockerTag=${{needs.tag.outputs.build-tag}} -f deployPreprodRegion=false
           elif [[ "$GITHUB_REF_NAME" == "release" ]]; then
+            gh workflow --repo neondatabase/aws run deploy-dev.yml --ref main -f branch=main -f dockerTag=${{needs.tag.outputs.build-tag}} -f deployPreprodRegion=true
             gh workflow --repo neondatabase/aws run deploy-prod.yml --ref main -f branch=main -f dockerTag=${{needs.tag.outputs.build-tag}} -f disclamerAcknowledged=true
           else
             echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release'"

Cargo.toml

@@ -78,7 +78,7 @@ hostname = "0.3.1"
 humantime = "2.1"
 humantime-serde = "1.1.1"
 hyper = "0.14"
-hyper-tungstenite = "0.9"
+hyper-tungstenite = "0.11"
 inotify = "0.10.2"
 itertools = "0.10"
 jsonwebtoken = "8"

Dockerfile.compute-node

@@ -589,8 +589,7 @@ RUN case "${PG_VERSION}" in \
     echo "${PG_EMBEDDING_CHECKSUM} pg_embedding.tar.gz" | sha256sum --check && \
     mkdir pg_embedding-src && cd pg_embedding-src && tar xvzf ../pg_embedding.tar.gz --strip-components=1 -C . && \
     make -j $(getconf _NPROCESSORS_ONLN) && \
-    make -j $(getconf _NPROCESSORS_ONLN) install && \
-    echo 'trusted = true' >> /usr/local/pgsql/share/extension/embedding.control
+    make -j $(getconf _NPROCESSORS_ONLN) install
 
 #########################################################################################
 #

Makefile

@@ -153,18 +153,6 @@ neon-pg-ext-%: postgres-%
 		-C $(POSTGRES_INSTALL_DIR)/build/neon-utils-$* \
 		-f $(ROOT_PROJECT_DIR)/pgxn/neon_utils/Makefile install
 
-# pg_embedding was temporarily released as hnsw from this repo, when we only
-# supported PostgreSQL 14 and 15
-neon-pg-ext-v14: neon-pg-ext-hnsw-v14
-neon-pg-ext-v15: neon-pg-ext-hnsw-v15
-
-neon-pg-ext-hnsw-%: postgres-headers-% postgres-%
-	+@echo "Compiling hnsw $*"
-	mkdir -p $(POSTGRES_INSTALL_DIR)/build/hnsw-$*
-	$(MAKE) PG_CONFIG=$(POSTGRES_INSTALL_DIR)/$*/bin/pg_config CFLAGS='$(PG_CFLAGS) $(COPT)' \
-		-C $(POSTGRES_INSTALL_DIR)/build/hnsw-$* \
-		-f $(ROOT_PROJECT_DIR)/pgxn/hnsw/Makefile install
-
 .PHONY: neon-pg-ext-clean-%
 neon-pg-ext-clean-%:
 	$(MAKE) PG_CONFIG=$(POSTGRES_INSTALL_DIR)/$*/bin/pg_config \
@@ -179,9 +167,6 @@ neon-pg-ext-clean-%:
 	$(MAKE) PG_CONFIG=$(POSTGRES_INSTALL_DIR)/$*/bin/pg_config \
 	-C $(POSTGRES_INSTALL_DIR)/build/neon-utils-$* \
 	-f $(ROOT_PROJECT_DIR)/pgxn/neon_utils/Makefile clean
-	$(MAKE) PG_CONFIG=$(POSTGRES_INSTALL_DIR)/$*/bin/pg_config \
-	-C $(POSTGRES_INSTALL_DIR)/build/hnsw-$* \
-	-f $(ROOT_PROJECT_DIR)/pgxn/hnsw/Makefile clean
 
 .PHONY: neon-pg-ext
 neon-pg-ext: \

README.md

@@ -29,13 +29,13 @@ See developer documentation in [SUMMARY.md](/docs/SUMMARY.md) for more informati
 ```bash
 apt install build-essential libtool libreadline-dev zlib1g-dev flex bison libseccomp-dev \
 libssl-dev clang pkg-config libpq-dev cmake postgresql-client protobuf-compiler \
-libcurl4-openssl-dev openssl python-poetry lsof
+libcurl4-openssl-dev openssl python-poetry lsof libicu-dev
 ```
 * On Fedora, these packages are needed:
 ```bash
 dnf install flex bison readline-devel zlib-devel openssl-devel \
   libseccomp-devel perl clang cmake postgresql postgresql-contrib protobuf-compiler \
-  protobuf-devel libcurl-devel openssl poetry lsof
+  protobuf-devel libcurl-devel openssl poetry lsof libicu-devel
 ```
 * On Arch based systems, these packages are needed:
 ```bash

pageserver/ctl/src/layer_map_analyzer.rs

@@ -3,6 +3,8 @@
 //! Currently it only analyzes holes, which are regions within the layer range that the layer contains no updates for. In the future it might do more analysis (maybe key quantiles?) but it should never return sensitive data.
 
 use anyhow::Result;
+use pageserver::context::{DownloadBehavior, RequestContext};
+use pageserver::task_mgr::TaskKind;
 use pageserver::tenant::{TENANTS_SEGMENT_NAME, TIMELINES_SEGMENT_NAME};
 use std::cmp::Ordering;
 use std::collections::BinaryHeap;
@@ -96,9 +98,9 @@ pub(crate) fn parse_filename(name: &str) -> Option<LayerFile> {
 }
 
 // Finds the max_holes largest holes, ignoring any that are smaller than MIN_HOLE_LENGTH"
-async fn get_holes(path: &Path, max_holes: usize) -> Result<Vec<Hole>> {
+async fn get_holes(path: &Path, max_holes: usize, ctx: &RequestContext) -> Result<Vec<Hole>> {
     let file = FileBlockReader::new(VirtualFile::open(path).await?);
-    let summary_blk = file.read_blk(0).await?;
+    let summary_blk = file.read_blk(0, ctx).await?;
     let actual_summary = Summary::des_prefix(summary_blk.as_ref())?;
     let tree_reader = DiskBtreeReader::<_, DELTA_KEY_SIZE>::new(
         actual_summary.index_start_blk,
@@ -125,6 +127,7 @@ async fn get_holes(path: &Path, max_holes: usize) -> Result<Vec<Hole>> {
                 prev_key = Some(curr.next());
                 true
             },
+            ctx,
         )
         .await?;
     let mut holes = heap.into_vec();
@@ -135,6 +138,7 @@ async fn get_holes(path: &Path, max_holes: usize) -> Result<Vec<Hole>> {
 pub(crate) async fn main(cmd: &AnalyzeLayerMapCmd) -> Result<()> {
     let storage_path = &cmd.path;
     let max_holes = cmd.max_holes.unwrap_or(DEFAULT_MAX_HOLES);
+    let ctx = RequestContext::new(TaskKind::DebugTool, DownloadBehavior::Error);
 
     // Initialize virtual_file (file desriptor cache) and page cache which are needed to access layer persistent B-Tree.
     pageserver::virtual_file::init(10);
@@ -163,7 +167,7 @@ pub(crate) async fn main(cmd: &AnalyzeLayerMapCmd) -> Result<()> {
                     parse_filename(&layer.file_name().into_string().unwrap())
                 {
                     if layer_file.is_delta {
-                        layer_file.holes = get_holes(&layer.path(), max_holes).await?;
+                        layer_file.holes = get_holes(&layer.path(), max_holes, &ctx).await?;
                         n_deltas += 1;
                     }
                     layers.push(layer_file);

pageserver/ctl/src/layers.rs

@@ -2,6 +2,8 @@ use std::path::{Path, PathBuf};
 
 use anyhow::Result;
 use clap::Subcommand;
+use pageserver::context::{DownloadBehavior, RequestContext};
+use pageserver::task_mgr::TaskKind;
 use pageserver::tenant::block_io::BlockCursor;
 use pageserver::tenant::disk_btree::DiskBtreeReader;
 use pageserver::tenant::storage_layer::delta_layer::{BlobRef, Summary};
@@ -44,12 +46,12 @@ pub(crate) enum LayerCmd {
     },
 }
 
-async fn read_delta_file(path: impl AsRef<Path>) -> Result<()> {
+async fn read_delta_file(path: impl AsRef<Path>, ctx: &RequestContext) -> Result<()> {
     let path = path.as_ref();
     virtual_file::init(10);
     page_cache::init(100);
     let file = FileBlockReader::new(VirtualFile::open(path).await?);
-    let summary_blk = file.read_blk(0).await?;
+    let summary_blk = file.read_blk(0, ctx).await?;
     let actual_summary = Summary::des_prefix(summary_blk.as_ref())?;
     let tree_reader = DiskBtreeReader::<_, DELTA_KEY_SIZE>::new(
         actual_summary.index_start_blk,
@@ -67,18 +69,20 @@ async fn read_delta_file(path: impl AsRef<Path>) -> Result<()> {
                 all.push((curr, BlobRef(value_offset)));
                 true
             },
+            ctx,
         )
         .await?;
     let cursor = BlockCursor::new_fileblockreader(&file);
     for (k, v) in all {
-        let value = cursor.read_blob(v.pos()).await?;
+        let value = cursor.read_blob(v.pos(), ctx).await?;
         println!("key:{} value_len:{}", k, value.len());
     }
     // TODO(chi): special handling for last key?
     Ok(())
 }
 
 pub(crate) async fn main(cmd: &LayerCmd) -> Result<()> {
+    let ctx = RequestContext::new(TaskKind::DebugTool, DownloadBehavior::Error);
     match cmd {
         LayerCmd::List { path } => {
             for tenant in fs::read_dir(path.join(TENANTS_SEGMENT_NAME))? {
@@ -153,7 +157,7 @@ pub(crate) async fn main(cmd: &LayerCmd) -> Result<()> {
                         );
 
                         if layer_file.is_delta {
-                            read_delta_file(layer.path()).await?;
+                            read_delta_file(layer.path(), &ctx).await?;
                         } else {
                             anyhow::bail!("not supported yet :(");
                         }

pageserver/src/context.rs

@@ -94,6 +94,18 @@ pub struct RequestContext {
     task_kind: TaskKind,
     download_behavior: DownloadBehavior,
     access_stats_behavior: AccessStatsBehavior,
+    page_content_kind: PageContentKind,
+}
+
+/// The kind of access to the page cache.
+#[derive(Clone, Copy, PartialEq, Eq, Debug, enum_map::Enum, strum_macros::IntoStaticStr)]
+pub enum PageContentKind {
+    Unknown,
+    DeltaLayerBtreeNode,
+    DeltaLayerValue,
+    ImageLayerBtreeNode,
+    ImageLayerValue,
+    InMemoryLayer,
 }
 
 /// Desired behavior if the operation requires an on-demand download
@@ -137,6 +149,7 @@ impl RequestContextBuilder {
                 task_kind,
                 download_behavior: DownloadBehavior::Download,
                 access_stats_behavior: AccessStatsBehavior::Update,
+                page_content_kind: PageContentKind::Unknown,
             },
         }
     }
@@ -149,6 +162,7 @@ impl RequestContextBuilder {
                 task_kind: original.task_kind,
                 download_behavior: original.download_behavior,
                 access_stats_behavior: original.access_stats_behavior,
+                page_content_kind: original.page_content_kind,
             },
         }
     }
@@ -167,6 +181,11 @@ impl RequestContextBuilder {
         self
     }
 
+    pub(crate) fn page_content_kind(mut self, k: PageContentKind) -> Self {
+        self.inner.page_content_kind = k;
+        self
+    }
+
     pub fn build(self) -> RequestContext {
         self.inner
     }
@@ -263,4 +282,8 @@ impl RequestContext {
     pub(crate) fn access_stats_behavior(&self) -> AccessStatsBehavior {
         self.access_stats_behavior
     }
+
+    pub(crate) fn page_content_kind(&self) -> PageContentKind {
+        self.page_content_kind
+    }
 }

pageserver/src/import_datadir.rs

@@ -75,12 +75,12 @@ pub async fn import_timeline_from_postgres_datadir(
             {
                 pg_control = Some(control_file);
             }
-            modification.flush().await?;
+            modification.flush(ctx).await?;
         }
     }
 
     // We're done importing all the data files.
-    modification.commit().await?;
+    modification.commit(ctx).await?;
 
     // We expect the Postgres server to be shut down cleanly.
     let pg_control = pg_control.context("pg_control file not found")?;
@@ -359,7 +359,7 @@ pub async fn import_basebackup_from_tar(
                     // We found the pg_control file.
                     pg_control = Some(res);
                 }
-                modification.flush().await?;
+                modification.flush(ctx).await?;
             }
             tokio_tar::EntryType::Directory => {
                 debug!("directory {:?}", file_path);
@@ -377,7 +377,7 @@ pub async fn import_basebackup_from_tar(
     // sanity check: ensure that pg_control is loaded
     let _pg_control = pg_control.context("pg_control file not found")?;
 
-    modification.commit().await?;
+    modification.commit(ctx).await?;
     Ok(())
 }