fix workspace hack

Signed-off-by: Alex Chi Z <chi@neon.tech>
consolidate encryption_key into download opts
2026-05-16 20:50:37 +00:00 · 2025-04-17 14:40:00 -04:00 · 2025-04-16 18:28:34 -04:00 · 2025-04-16 15:34:17 -04:00 · 2025-04-16 14:19:43 -04:00 · 2025-04-16 14:16:43 -04:00
224 changed files with 11010 additions and 5134 deletions
--- a/.dockerignore
+++ b/.dockerignore
@@ -19,6 +19,7 @@
 !pageserver/
 !pgxn/
 !proxy/
+!object_storage/
 !storage_scrubber/
 !safekeeper/
 !storage_broker/
--- a/.github/actionlint.yml
+++ b/.github/actionlint.yml
@@ -6,6 +6,7 @@ self-hosted-runner:
    - small
    - small-metal
    - small-arm64
+    - unit-perf
    - us-east-2
 config-variables:
  - AWS_ECR_REGION
--- a/.github/actions/allure-report-generate/action.yml
+++ b/.github/actions/allure-report-generate/action.yml
@@ -70,6 +70,7 @@ runs:

    - name: Install Allure
      shell: bash -euxo pipefail {0}
+      working-directory: /tmp
      run: |
        if ! which allure; then
          ALLURE_ZIP=allure-${ALLURE_VERSION}.zip
--- a/.github/scripts/push_with_image_map.py
+++ b/.github/scripts/push_with_image_map.py
@@ -2,6 +2,9 @@ import json
 import os
 import subprocess

+RED = "\033[91m"
+RESET = "\033[0m"
+
 image_map = os.getenv("IMAGE_MAP")
 if not image_map:
    raise ValueError("IMAGE_MAP environment variable is not set")
@@ -29,9 +32,14 @@ while len(pending) > 0:
    result = subprocess.run(cmd, text=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT)

    if result.returncode != 0:
-        failures.append((" ".join(cmd), result.stdout))
+        failures.append((" ".join(cmd), result.stdout, target))
        pending.append((source, target))
+        print(
+            f"{RED}[RETRY]{RESET} Push failed for {target}. Retrying... (failure count: {len(failures)})"
+        )
+        print(result.stdout)

 if len(failures) > 0 and (github_output := os.getenv("GITHUB_OUTPUT")):
+    failed_targets = [target for _, _, target in failures]
    with open(github_output, "a") as f:
-        f.write("slack_notify=true\n")
+        f.write(f"push_failures={json.dumps(failed_targets)}\n")
--- a/.github/workflows/_create-release-pr.yml
+++ b/.github/workflows/_create-release-pr.yml
@@ -53,10 +53,13 @@ jobs:
            || inputs.component-name == 'Compute' && 'release-compute'
          }}
      run: |
-        today=$(date +'%Y-%m-%d')
-        echo "title=${COMPONENT_NAME} release ${today}" | tee -a ${GITHUB_OUTPUT}
-        echo "rc-branch=rc/${RELEASE_BRANCH}/${today}"  | tee -a ${GITHUB_OUTPUT}
-        echo "release-branch=${RELEASE_BRANCH}"         | tee -a ${GITHUB_OUTPUT}
+        now_date=$(date -u +'%Y-%m-%d')
+        now_time=$(date -u +'%H-%M-%Z')
+        {
+          echo "title=${COMPONENT_NAME} release ${now_date}"
+          echo "rc-branch=rc/${RELEASE_BRANCH}/${now_date}_${now_time}"
+          echo "release-branch=${RELEASE_BRANCH}"
+        } | tee -a ${GITHUB_OUTPUT}

    - name: Configure git
      run: |
--- a/.github/workflows/_push-to-container-registry.yml
+++ b/.github/workflows/_push-to-container-registry.yml
@@ -110,12 +110,19 @@ jobs:
          IMAGE_MAP: ${{ inputs.image-map }}

      - name: Notify Slack if container image pushing fails
-        if: steps.push.outputs.slack_notify == 'true' || failure()
+        if: steps.push.outputs.push_failures || failure()
        uses: slackapi/slack-github-action@485a9d42d3a73031f12ec201c457e2162c45d02d # v2.0.0
        with:
          method: chat.postMessage
          token: ${{ secrets.SLACK_BOT_TOKEN }}
          payload: |
            channel: ${{ vars.SLACK_ON_CALL_DEVPROD_STREAM }}
-            text: |
-              Pushing container images failed in <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}|GitHub Run>
+            text: >
+              *Container image pushing ${{
+                steps.push.outcome == 'failure' && 'failed completely' || 'succeeded with some retries'
+              }}* in
+              <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}|GitHub Run>
+
+              ${{ steps.push.outputs.push_failures && format(
+                '*Failed targets:*\n• {0}', join(fromJson(steps.push.outputs.push_failures), '\n• ')
+              ) || '' }}
--- a/.github/workflows/build_and_test.yml
+++ b/.github/workflows/build_and_test.yml
@@ -284,7 +284,7 @@ jobs:
      statuses: write
      contents: write
      pull-requests: write
-    runs-on: [ self-hosted, small-metal ]
+    runs-on: [ self-hosted, unit-perf ]
    container:
      image: ${{ needs.build-build-tools-image.outputs.image }}-bookworm
      credentials:
@@ -980,7 +980,7 @@ jobs:
          TEST_EXTENSIONS_TAG: >-
            ${{
              contains(fromJSON('["storage-rc-pr", "proxy-rc-pr"]'), needs.meta.outputs.run-kind)
-              && 'latest'
+              && needs.meta.outputs.previous-compute-release
              || needs.meta.outputs.build-tag
            }}
          TEST_VERSION_ONLY: ${{ matrix.pg_version }}
@@ -1271,7 +1271,7 @@ jobs:
          exit 1

  deploy:
-    needs: [ check-permissions, push-neon-image-dev, push-compute-image-dev, push-neon-image-prod, push-compute-image-prod, meta, build-and-test-locally, trigger-custom-extensions-build-and-wait ]
+    needs: [ check-permissions, push-neon-image-dev, push-compute-image-dev, push-neon-image-prod, push-compute-image-prod, meta, trigger-custom-extensions-build-and-wait ]
    # `!failure() && !cancelled()` is required because the workflow depends on the job that can be skipped: `push-neon-image-prod` and `push-compute-image-prod`
    if: ${{ contains(fromJSON('["push-main", "storage-release", "proxy-release", "compute-release"]'), needs.meta.outputs.run-kind) && !failure() && !cancelled() }}
    permissions:
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,4 @@
+/artifact_cache
 /pg_install
 /target
 /tmp_check
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -2837,6 +2837,7 @@ dependencies = [
 "utils",
 "uuid",
 "workspace_hack",
+ "x509-cert",
 ]

 [[package]]
@@ -3991,6 +3992,33 @@ dependencies = [
 "memchr",
 ]

+[[package]]
+name = "object_storage"
+version = "0.0.1"
+dependencies = [
+ "anyhow",
+ "axum",
+ "axum-extra",
+ "camino",
+ "camino-tempfile",
+ "futures",
+ "http-body-util",
+ "itertools 0.10.5",
+ "jsonwebtoken",
+ "prometheus",
+ "rand 0.8.5",
+ "remote_storage",
+ "serde",
+ "serde_json",
+ "test-log",
+ "tokio",
+ "tokio-util",
+ "tower 0.5.2",
+ "tracing",
+ "utils",
+ "workspace_hack",
+]
+
 [[package]]
 name = "once_cell"
 version = "1.20.2"
@@ -4693,7 +4721,7 @@ dependencies = [
 [[package]]
 name = "postgres-protocol"
 version = "0.6.6"
-source = "git+https://github.com/neondatabase/rust-postgres.git?branch=neon#1f21e7959a96a34dcfbfce1b14b73286cdadffe9"
+source = "git+https://github.com/neondatabase/rust-postgres.git?branch=neon#f3cf448febde5fd298071d54d568a9c875a7a62b"
 dependencies = [
 "base64 0.22.1",
 "byteorder",
@@ -4727,7 +4755,7 @@ dependencies = [
 [[package]]
 name = "postgres-types"
 version = "0.2.6"
-source = "git+https://github.com/neondatabase/rust-postgres.git?branch=neon#1f21e7959a96a34dcfbfce1b14b73286cdadffe9"
+source = "git+https://github.com/neondatabase/rust-postgres.git?branch=neon#f3cf448febde5fd298071d54d568a9c875a7a62b"
 dependencies = [
 "bytes",
 "chrono",
@@ -5467,6 +5495,16 @@ version = "1.9.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "c707298afce11da2efef2f600116fa93ffa7a032b5d7b628aa17711ec81383ca"

+[[package]]
+name = "remote_keys"
+version = "0.1.0"
+dependencies = [
+ "anyhow",
+ "rand 0.8.5",
+ "utils",
+ "workspace_hack",
+]
+
 [[package]]
 name = "remote_storage"
 version = "0.1.0"
@@ -5482,6 +5520,7 @@ dependencies = [
 "azure_identity",
 "azure_storage",
 "azure_storage_blobs",
+ "base64 0.13.1",
 "bytes",
 "camino",
 "camino-tempfile",
@@ -5492,6 +5531,7 @@ dependencies = [
 "humantime-serde",
 "hyper 1.4.1",
 "itertools 0.10.5",
+ "md5",
 "metrics",
 "once_cell",
 "pin-project-lite",
@@ -6925,6 +6965,28 @@ dependencies = [
 "syn 2.0.100",
 ]

+[[package]]
+name = "test-log"
+version = "0.2.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e7f46083d221181166e5b6f6b1e5f1d499f3a76888826e6cb1d057554157cd0f"
+dependencies = [
+ "env_logger",
+ "test-log-macros",
+ "tracing-subscriber",
+]
+
+[[package]]
+name = "test-log-macros"
+version = "0.2.17"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "888d0c3c6db53c0fdab160d2ed5e12ba745383d3e85813f2ea0f2b1475ab553f"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.100",
+]
+
 [[package]]
 name = "thiserror"
 version = "1.0.69"
@@ -7116,9 +7178,9 @@ dependencies = [

 [[package]]
 name = "tokio"
-version = "1.43.0"
+version = "1.43.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3d61fa4ffa3de412bfea335c6ecff681de2b609ba3c77ef3e00e521813a9ed9e"
+checksum = "492a604e2fd7f814268a378409e6c92b5525d747d10db9a229723f55a417958c"
 dependencies = [
 "backtrace",
 "bytes",
@@ -7172,7 +7234,7 @@ dependencies = [
 [[package]]
 name = "tokio-postgres"
 version = "0.7.10"
-source = "git+https://github.com/neondatabase/rust-postgres.git?branch=neon#1f21e7959a96a34dcfbfce1b14b73286cdadffe9"
+source = "git+https://github.com/neondatabase/rust-postgres.git?branch=neon#f3cf448febde5fd298071d54d568a9c875a7a62b"
 dependencies = [
 "async-trait",
 "byteorder",
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -30,6 +30,7 @@ members = [
    "libs/tenant_size_model",
    "libs/metrics",
    "libs/postgres_connection",
+    "libs/remote_keys",
    "libs/remote_storage",
    "libs/tracing-utils",
    "libs/postgres_ffi/wal_craft",
@@ -40,6 +41,7 @@ members = [
    "libs/proxy/postgres-protocol2",
    "libs/proxy/postgres-types2",
    "libs/proxy/tokio-postgres2",
+    "object_storage",
 ]

 [workspace.package]
@@ -183,7 +185,7 @@ test-context = "0.3"
 thiserror = "1.0"
 tikv-jemallocator = { version = "0.6", features = ["profiling", "stats", "unprefixed_malloc_on_supported_platforms"] }
 tikv-jemalloc-ctl = { version = "0.6", features = ["stats"] }
-tokio = { version = "1.41", features = ["macros"] }
+tokio = { version = "1.43.1", features = ["macros"] }
 tokio-epoll-uring = { git = "https://github.com/neondatabase/tokio-epoll-uring.git" , branch = "main" }
 tokio-io-timeout = "1.2.0"
 tokio-postgres-rustls = "0.12.0"
@@ -208,6 +210,7 @@ tracing-opentelemetry = "0.28"
 tracing-serde = "0.2.0"
 tracing-subscriber = { version = "0.3", default-features = false, features = ["smallvec", "fmt", "tracing-log", "std", "env-filter", "json"] }
 try-lock = "0.2.5"
+test-log = { version = "0.2.17", default-features = false, features = ["log"] }
 twox-hash = { version = "1.6.3", default-features = false }
 typed-json = "0.1"
 url = "2.2"
--- a/2
+++ b/2
@@ -89,6 +89,7 @@ RUN set -e \
      --bin storage_broker  \
      --bin storage_controller  \
      --bin proxy  \
+      --bin object_storage \
      --bin neon_local \
      --bin storage_scrubber \
      --locked --release
@@ -121,6 +122,7 @@ COPY --from=build --chown=neon:neon /home/nonroot/target/release/safekeeper
 COPY --from=build --chown=neon:neon /home/nonroot/target/release/storage_broker      /usr/local/bin
 COPY --from=build --chown=neon:neon /home/nonroot/target/release/storage_controller  /usr/local/bin
 COPY --from=build --chown=neon:neon /home/nonroot/target/release/proxy               /usr/local/bin
+COPY --from=build --chown=neon:neon /home/nonroot/target/release/object_storage      /usr/local/bin
 COPY --from=build --chown=neon:neon /home/nonroot/target/release/neon_local          /usr/local/bin
 COPY --from=build --chown=neon:neon /home/nonroot/target/release/storage_scrubber    /usr/local/bin

--- a/compute/compute-node.Dockerfile
+++ b/compute/compute-node.Dockerfile
@@ -1022,39 +1022,6 @@ RUN make -j $(getconf _NPROCESSORS_ONLN) && \
    make -j $(getconf _NPROCESSORS_ONLN) install && \
    echo 'trusted = true' >> /usr/local/pgsql/share/extension/semver.control

-#########################################################################################
-#
-# Layer "pg_embedding-build"
-# compile pg_embedding extension
-#
-#########################################################################################
-FROM build-deps AS pg_embedding-src
-ARG PG_VERSION
-
-# This is our extension, support stopped in favor of pgvector
-# TODO: deprecate it
-WORKDIR /ext-src
-RUN case "${PG_VERSION:?}" in \
-      "v14" | "v15") \
-        export PG_EMBEDDING_VERSION=0.3.5 \
-        export PG_EMBEDDING_CHECKSUM=0e95b27b8b6196e2cf0a0c9ec143fe2219b82e54c5bb4ee064e76398cbe69ae9 \
-        ;; \
-      *) \
-        echo "pg_embedding not supported on this PostgreSQL version. Use pgvector instead." && exit 0;; \
-    esac && \
-    wget https://github.com/neondatabase/pg_embedding/archive/refs/tags/${PG_EMBEDDING_VERSION}.tar.gz -O pg_embedding.tar.gz && \
-    echo "${PG_EMBEDDING_CHECKSUM} pg_embedding.tar.gz" | sha256sum --check && \
-    mkdir pg_embedding-src && cd pg_embedding-src && tar xzf ../pg_embedding.tar.gz --strip-components=1 -C .
-
-FROM pg-build AS pg_embedding-build
-COPY --from=pg_embedding-src /ext-src/ /ext-src/
-WORKDIR /ext-src/
-RUN  if [ -d pg_embedding-src ]; then \
-        cd pg_embedding-src && \
-        make -j $(getconf _NPROCESSORS_ONLN) && \
-        make -j $(getconf _NPROCESSORS_ONLN) install; \
-    fi
-
 #########################################################################################
 #
 # Layer "pg build with nonroot user and cargo installed"
@@ -1647,7 +1614,6 @@ COPY --from=rdkit-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg_uuidv7-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg_roaringbitmap-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg_semver-build /usr/local/pgsql/ /usr/local/pgsql/
-COPY --from=pg_embedding-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=wal2json-build /usr/local/pgsql /usr/local/pgsql
 COPY --from=pg_ivm-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg_partman-build /usr/local/pgsql/ /usr/local/pgsql/
@@ -1824,7 +1790,6 @@ COPY --from=pg_cron-src /ext-src/ /ext-src/
 COPY --from=pg_uuidv7-src /ext-src/ /ext-src/
 COPY --from=pg_roaringbitmap-src /ext-src/ /ext-src/
 COPY --from=pg_semver-src /ext-src/ /ext-src/
-#COPY --from=pg_embedding-src /ext-src/ /ext-src/
 #COPY --from=wal2json-src /ext-src/ /ext-src/
 COPY --from=pg_ivm-src /ext-src/ /ext-src/
 COPY --from=pg_partman-src /ext-src/ /ext-src/
--- a/compute/patches/cloud_regress_pg16.patch
+++ b/compute/patches/cloud_regress_pg16.patch
@@ -202,10 +202,10 @@ index cf0b80d616..e8e2a14a4a 100644
 COMMENT ON CONSTRAINT the_constraint ON constraint_comments_tbl IS 'no, the comment';
 ERROR:  must be owner of relation constraint_comments_tbl
 diff --git a/src/test/regress/expected/conversion.out b/src/test/regress/expected/conversion.out
-index 442e7aff2b..525f732b03 100644
+index d785f92561..16377e5ac9 100644
 --- a/src/test/regress/expected/conversion.out
 +++ b/src/test/regress/expected/conversion.out
-@@ -8,7 +8,7 @@
+@@ -15,7 +15,7 @@ SELECT FROM test_enc_setup();
 CREATE FUNCTION test_enc_conversion(bytea, name, name, bool, validlen OUT int, result OUT bytea)
     AS :'regresslib', 'test_enc_conversion'
     LANGUAGE C STRICT;
@@ -587,16 +587,15 @@ index f551624afb..57f1e432d4 100644
 SELECT *
    INTO TABLE ramp
 diff --git a/src/test/regress/expected/database.out b/src/test/regress/expected/database.out
-index 454db91ec0..01378d7081 100644
+index 4cbdbdf84d..573362850e 100644
 --- a/src/test/regress/expected/database.out
 +++ b/src/test/regress/expected/database.out
-@@ -1,8 +1,7 @@
+@@ -1,8 +1,6 @@
 CREATE DATABASE regression_tbd
 	ENCODING utf8 LC_COLLATE "C" LC_CTYPE "C" TEMPLATE template0;
 ALTER DATABASE regression_tbd RENAME TO regression_utf8;
 -ALTER DATABASE regression_utf8 SET TABLESPACE regress_tblspace;
 -ALTER DATABASE regression_utf8 RESET TABLESPACE;
-+WARNING:  you need to manually restart any running background workers after this command
 ALTER DATABASE regression_utf8 CONNECTION_LIMIT 123;
 -- Test PgDatabaseToastTable.  Doing this with GRANT would be slow.
 BEGIN;
@@ -700,7 +699,7 @@ index 6ed50fdcfa..caa00a345d 100644
 COMMENT ON FOREIGN DATA WRAPPER dummy IS 'useless';
 CREATE FOREIGN DATA WRAPPER postgresql VALIDATOR postgresql_fdw_validator;
 diff --git a/src/test/regress/expected/foreign_key.out b/src/test/regress/expected/foreign_key.out
-index 6b8c2f2414..8e13b7fa46 100644
+index 84745b9f60..4883c12351 100644
 --- a/src/test/regress/expected/foreign_key.out
 +++ b/src/test/regress/expected/foreign_key.out
@@ -1985,7 +1985,7 @@ ALTER TABLE fk_partitioned_fk_6 ATTACH PARTITION fk_partitioned_pk_6 FOR VALUES
@@ -1112,7 +1111,7 @@ index 8475231735..0653946337 100644
 DROP ROLE regress_passwd_sha_len1;
 DROP ROLE regress_passwd_sha_len2;
 diff --git a/src/test/regress/expected/privileges.out b/src/test/regress/expected/privileges.out
-index 5b9dba7b32..cc408dad42 100644
+index 620fbe8c52..0570102357 100644
 --- a/src/test/regress/expected/privileges.out
 +++ b/src/test/regress/expected/privileges.out
@@ -20,19 +20,19 @@ SELECT lo_unlink(oid) FROM pg_largeobject_metadata WHERE oid >= 1000 AND oid < 3
@@ -1174,8 +1173,8 @@ index 5b9dba7b32..cc408dad42 100644
 +CREATE GROUP regress_priv_group2 WITH ADMIN regress_priv_user1 PASSWORD NEON_PASSWORD_PLACEHOLDER USER regress_priv_user2;
 ALTER GROUP regress_priv_group1 ADD USER regress_priv_user4;
 GRANT regress_priv_group2 TO regress_priv_user2 GRANTED BY regress_priv_user1;
- SET SESSION AUTHORIZATION regress_priv_user1;
-@@ -239,12 +239,16 @@ GRANT regress_priv_role TO regress_priv_user1 WITH ADMIN OPTION GRANTED BY regre
+ SET SESSION AUTHORIZATION regress_priv_user3;
+@@ -246,12 +246,16 @@ GRANT regress_priv_role TO regress_priv_user1 WITH ADMIN OPTION GRANTED BY regre
 ERROR:  permission denied to grant privileges as role "regress_priv_role"
 DETAIL:  The grantor must have the ADMIN option on role "regress_priv_role".
 GRANT regress_priv_role TO regress_priv_user1 WITH ADMIN OPTION GRANTED BY CURRENT_ROLE;
@@ -1192,7 +1191,7 @@ index 5b9dba7b32..cc408dad42 100644
 DROP ROLE regress_priv_role;
 SET SESSION AUTHORIZATION regress_priv_user1;
 SELECT session_user, current_user;
-@@ -1776,7 +1780,7 @@ SELECT has_table_privilege('regress_priv_user1', 'atest4', 'SELECT WITH GRANT OP
+@@ -1783,7 +1787,7 @@ SELECT has_table_privilege('regress_priv_user1', 'atest4', 'SELECT WITH GRANT OP
 
 -- security-restricted operations
 \c -
@@ -1201,7 +1200,7 @@ index 5b9dba7b32..cc408dad42 100644
 -- Check that index expressions and predicates are run as the table's owner
 -- A dummy index function checking current_user
 CREATE FUNCTION sro_ifun(int) RETURNS int AS $$
-@@ -2668,8 +2672,8 @@ drop cascades to function testns.priv_testagg(integer)
+@@ -2675,8 +2679,8 @@ drop cascades to function testns.priv_testagg(integer)
 drop cascades to function testns.priv_testproc(integer)
 -- Change owner of the schema & and rename of new schema owner
 \c -
@@ -1212,7 +1211,7 @@ index 5b9dba7b32..cc408dad42 100644
 SET SESSION ROLE regress_schemauser1;
 CREATE SCHEMA testns;
 SELECT nspname, rolname FROM pg_namespace, pg_roles WHERE pg_namespace.nspname = 'testns' AND pg_namespace.nspowner = pg_roles.oid;
-@@ -2792,7 +2796,7 @@ DROP USER regress_priv_user7;
+@@ -2799,7 +2803,7 @@ DROP USER regress_priv_user7;
 DROP USER regress_priv_user8; -- does not exist
 ERROR:  role "regress_priv_user8" does not exist
 -- permissions with LOCK TABLE
@@ -1221,7 +1220,7 @@ index 5b9dba7b32..cc408dad42 100644
 CREATE TABLE lock_table (a int);
 -- LOCK TABLE and SELECT permission
 GRANT SELECT ON lock_table TO regress_locktable_user;
-@@ -2874,7 +2878,7 @@ DROP USER regress_locktable_user;
+@@ -2881,7 +2885,7 @@ DROP USER regress_locktable_user;
 -- pg_backend_memory_contexts.
 -- switch to superuser
 \c -
@@ -1230,7 +1229,7 @@ index 5b9dba7b32..cc408dad42 100644
 SELECT has_table_privilege('regress_readallstats','pg_backend_memory_contexts','SELECT'); -- no
  has_table_privilege 
 ---------------------
-@@ -2918,10 +2922,10 @@ RESET ROLE;
+@@ -2925,10 +2929,10 @@ RESET ROLE;
 -- clean up
 DROP ROLE regress_readallstats;
 -- test role grantor machinery
@@ -1245,7 +1244,7 @@ index 5b9dba7b32..cc408dad42 100644
 GRANT regress_group TO regress_group_direct_manager WITH INHERIT FALSE, ADMIN TRUE;
 GRANT regress_group_direct_manager TO regress_group_indirect_manager;
 SET SESSION AUTHORIZATION regress_group_direct_manager;
-@@ -2950,9 +2954,9 @@ DROP ROLE regress_group_direct_manager;
+@@ -2957,9 +2961,9 @@ DROP ROLE regress_group_direct_manager;
 DROP ROLE regress_group_indirect_manager;
 DROP ROLE regress_group_member;
 -- test SET and INHERIT options with object ownership changes
@@ -1841,7 +1840,7 @@ index 09a255649b..15895f0c53 100644
 CREATE TABLE ruletest_t2 (x int);
 CREATE VIEW ruletest_v1 WITH (security_invoker=true) AS
 diff --git a/src/test/regress/expected/security_label.out b/src/test/regress/expected/security_label.out
-index a8e01a6220..5a9cef4ede 100644
+index a8e01a6220..83543b250a 100644
 --- a/src/test/regress/expected/security_label.out
 +++ b/src/test/regress/expected/security_label.out
@@ -6,8 +6,8 @@ SET client_min_messages TO 'warning';
@@ -1855,34 +1854,6 @@ index a8e01a6220..5a9cef4ede 100644
 CREATE TABLE seclabel_tbl1 (a int, b text);
 CREATE TABLE seclabel_tbl2 (x int, y text);
 CREATE VIEW seclabel_view1 AS SELECT * FROM seclabel_tbl2;
-@@ -19,21 +19,21 @@ ALTER TABLE seclabel_tbl2 OWNER TO regress_seclabel_user2;
- -- Test of SECURITY LABEL statement without a plugin
- --
- SECURITY LABEL ON TABLE seclabel_tbl1 IS 'classified';			-- fail
-ERROR:  no security label providers have been loaded
-+ERROR:  must specify provider when multiple security label providers have been loaded
- SECURITY LABEL FOR 'dummy' ON TABLE seclabel_tbl1 IS 'classified';		-- fail
- ERROR:  security label provider "dummy" is not loaded
- SECURITY LABEL ON TABLE seclabel_tbl1 IS '...invalid label...';		-- fail
-ERROR:  no security label providers have been loaded
-+ERROR:  must specify provider when multiple security label providers have been loaded
- SECURITY LABEL ON TABLE seclabel_tbl3 IS 'unclassified';			-- fail
-ERROR:  no security label providers have been loaded
-+ERROR:  must specify provider when multiple security label providers have been loaded
- SECURITY LABEL ON ROLE regress_seclabel_user1 IS 'classified';			-- fail
-ERROR:  no security label providers have been loaded
-+ERROR:  must specify provider when multiple security label providers have been loaded
- SECURITY LABEL FOR 'dummy' ON ROLE regress_seclabel_user1 IS 'classified';		-- fail
- ERROR:  security label provider "dummy" is not loaded
- SECURITY LABEL ON ROLE regress_seclabel_user1 IS '...invalid label...';		-- fail
-ERROR:  no security label providers have been loaded
-+ERROR:  must specify provider when multiple security label providers have been loaded
- SECURITY LABEL ON ROLE regress_seclabel_user3 IS 'unclassified';			-- fail
-ERROR:  no security label providers have been loaded
-+ERROR:  must specify provider when multiple security label providers have been loaded
- -- clean up objects
- DROP FUNCTION seclabel_four();
- DROP DOMAIN seclabel_domain;
 diff --git a/src/test/regress/expected/select_into.out b/src/test/regress/expected/select_into.out
 index b79fe9a1c0..e29fab88ab 100644
 --- a/src/test/regress/expected/select_into.out
@@ -2413,10 +2384,10 @@ index e3e3bea709..fa86ddc326 100644
 COMMENT ON CONSTRAINT the_constraint ON constraint_comments_tbl IS 'no, the comment';
 COMMENT ON CONSTRAINT the_constraint ON DOMAIN constraint_comments_dom IS 'no, another comment';
 diff --git a/src/test/regress/sql/conversion.sql b/src/test/regress/sql/conversion.sql
-index 9a65fca91f..58431a3056 100644
+index b567a1a572..4d1ac2e631 100644
 --- a/src/test/regress/sql/conversion.sql
 +++ b/src/test/regress/sql/conversion.sql
-@@ -12,7 +12,7 @@ CREATE FUNCTION test_enc_conversion(bytea, name, name, bool, validlen OUT int, r
+@@ -17,7 +17,7 @@ CREATE FUNCTION test_enc_conversion(bytea, name, name, bool, validlen OUT int, r
     AS :'regresslib', 'test_enc_conversion'
     LANGUAGE C STRICT;
 
@@ -2780,7 +2751,7 @@ index ae6841308b..47bc792e30 100644
 
 SELECT *
 diff --git a/src/test/regress/sql/database.sql b/src/test/regress/sql/database.sql
-index 0367c0e37a..a23b98c4bd 100644
+index 46ad263478..eb05584ed5 100644
 --- a/src/test/regress/sql/database.sql
 +++ b/src/test/regress/sql/database.sql
@@ -1,8 +1,6 @@
@@ -2893,7 +2864,7 @@ index aa147b14a9..370e0dd570 100644
 CREATE FOREIGN DATA WRAPPER dummy;
 COMMENT ON FOREIGN DATA WRAPPER dummy IS 'useless';
 diff --git a/src/test/regress/sql/foreign_key.sql b/src/test/regress/sql/foreign_key.sql
-index 45c7a534cb..32dd26b8cd 100644
+index 9f4210b26e..620d3fc87e 100644
 --- a/src/test/regress/sql/foreign_key.sql
 +++ b/src/test/regress/sql/foreign_key.sql
@@ -1435,7 +1435,7 @@ ALTER TABLE fk_partitioned_fk_6 ATTACH PARTITION fk_partitioned_pk_6 FOR VALUES
@@ -3246,7 +3217,7 @@ index 53e86b0b6c..0303fdfe96 100644
 -- Check that the invalid secrets were re-hashed. A re-hashed secret
 -- should not contain the original salt.
 diff --git a/src/test/regress/sql/privileges.sql b/src/test/regress/sql/privileges.sql
-index 249df17a58..b258e7f26a 100644
+index 259f1aedd1..6e1a3d17b7 100644
 --- a/src/test/regress/sql/privileges.sql
 +++ b/src/test/regress/sql/privileges.sql
@@ -24,18 +24,18 @@ RESET client_min_messages;
@@ -3308,7 +3279,7 @@ index 249df17a58..b258e7f26a 100644
 
 ALTER GROUP regress_priv_group1 ADD USER regress_priv_user4;
 
-@@ -1157,7 +1157,7 @@ SELECT has_table_privilege('regress_priv_user1', 'atest4', 'SELECT WITH GRANT OP
+@@ -1160,7 +1160,7 @@ SELECT has_table_privilege('regress_priv_user1', 'atest4', 'SELECT WITH GRANT OP
 
 -- security-restricted operations
 \c -
@@ -3317,7 +3288,7 @@ index 249df17a58..b258e7f26a 100644
 
 -- Check that index expressions and predicates are run as the table's owner
 
-@@ -1653,8 +1653,8 @@ DROP SCHEMA testns CASCADE;
+@@ -1656,8 +1656,8 @@ DROP SCHEMA testns CASCADE;
 -- Change owner of the schema & and rename of new schema owner
 \c -
 
@@ -3328,7 +3299,7 @@ index 249df17a58..b258e7f26a 100644
 
 SET SESSION ROLE regress_schemauser1;
 CREATE SCHEMA testns;
-@@ -1748,7 +1748,7 @@ DROP USER regress_priv_user8; -- does not exist
+@@ -1751,7 +1751,7 @@ DROP USER regress_priv_user8; -- does not exist
 
 
 -- permissions with LOCK TABLE
@@ -3337,7 +3308,7 @@ index 249df17a58..b258e7f26a 100644
 CREATE TABLE lock_table (a int);
 
 -- LOCK TABLE and SELECT permission
-@@ -1836,7 +1836,7 @@ DROP USER regress_locktable_user;
+@@ -1839,7 +1839,7 @@ DROP USER regress_locktable_user;
 -- switch to superuser
 \c -
 
@@ -3346,7 +3317,7 @@ index 249df17a58..b258e7f26a 100644
 
 SELECT has_table_privilege('regress_readallstats','pg_backend_memory_contexts','SELECT'); -- no
 SELECT has_table_privilege('regress_readallstats','pg_shmem_allocations','SELECT'); -- no
-@@ -1856,10 +1856,10 @@ RESET ROLE;
+@@ -1859,10 +1859,10 @@ RESET ROLE;
 DROP ROLE regress_readallstats;
 
 -- test role grantor machinery
@@ -3361,7 +3332,7 @@ index 249df17a58..b258e7f26a 100644
 
 GRANT regress_group TO regress_group_direct_manager WITH INHERIT FALSE, ADMIN TRUE;
 GRANT regress_group_direct_manager TO regress_group_indirect_manager;
-@@ -1881,9 +1881,9 @@ DROP ROLE regress_group_indirect_manager;
+@@ -1884,9 +1884,9 @@ DROP ROLE regress_group_indirect_manager;
 DROP ROLE regress_group_member;
 
 -- test SET and INHERIT options with object ownership changes
--- a/compute/patches/cloud_regress_pg17.patch
+++ b/compute/patches/cloud_regress_pg17.patch
@@ -202,10 +202,10 @@ index cf0b80d616..e8e2a14a4a 100644
 COMMENT ON CONSTRAINT the_constraint ON constraint_comments_tbl IS 'no, the comment';
 ERROR:  must be owner of relation constraint_comments_tbl
 diff --git a/src/test/regress/expected/conversion.out b/src/test/regress/expected/conversion.out
-index 442e7aff2b..525f732b03 100644
+index d785f92561..16377e5ac9 100644
 --- a/src/test/regress/expected/conversion.out
 +++ b/src/test/regress/expected/conversion.out
-@@ -8,7 +8,7 @@
+@@ -15,7 +15,7 @@ SELECT FROM test_enc_setup();
 CREATE FUNCTION test_enc_conversion(bytea, name, name, bool, validlen OUT int, result OUT bytea)
     AS :'regresslib', 'test_enc_conversion'
     LANGUAGE C STRICT;
@@ -587,16 +587,15 @@ index f551624afb..57f1e432d4 100644
 SELECT *
    INTO TABLE ramp
 diff --git a/src/test/regress/expected/database.out b/src/test/regress/expected/database.out
-index 454db91ec0..01378d7081 100644
+index 4cbdbdf84d..573362850e 100644
 --- a/src/test/regress/expected/database.out
 +++ b/src/test/regress/expected/database.out
-@@ -1,8 +1,7 @@
+@@ -1,8 +1,6 @@
 CREATE DATABASE regression_tbd
 	ENCODING utf8 LC_COLLATE "C" LC_CTYPE "C" TEMPLATE template0;
 ALTER DATABASE regression_tbd RENAME TO regression_utf8;
 -ALTER DATABASE regression_utf8 SET TABLESPACE regress_tblspace;
 -ALTER DATABASE regression_utf8 RESET TABLESPACE;
-+WARNING:  you need to manually restart any running background workers after this command
 ALTER DATABASE regression_utf8 CONNECTION_LIMIT 123;
 -- Test PgDatabaseToastTable.  Doing this with GRANT would be slow.
 BEGIN;
@@ -700,7 +699,7 @@ index 6ed50fdcfa..caa00a345d 100644
 COMMENT ON FOREIGN DATA WRAPPER dummy IS 'useless';
 CREATE FOREIGN DATA WRAPPER postgresql VALIDATOR postgresql_fdw_validator;
 diff --git a/src/test/regress/expected/foreign_key.out b/src/test/regress/expected/foreign_key.out
-index 69994c98e3..129abcfbe8 100644
+index fe6a1015f2..614b387b7d 100644
 --- a/src/test/regress/expected/foreign_key.out
 +++ b/src/test/regress/expected/foreign_key.out
@@ -1985,7 +1985,7 @@ ALTER TABLE fk_partitioned_fk_6 ATTACH PARTITION fk_partitioned_pk_6 FOR VALUES
@@ -1147,7 +1146,7 @@ index 924d6e001d..7fdda73439 100644
 DROP ROLE regress_passwd_sha_len1;
 DROP ROLE regress_passwd_sha_len2;
 diff --git a/src/test/regress/expected/privileges.out b/src/test/regress/expected/privileges.out
-index 1296da0d57..f43fffa44c 100644
+index e8c668e0a1..03be5c2120 100644
 --- a/src/test/regress/expected/privileges.out
 +++ b/src/test/regress/expected/privileges.out
@@ -20,19 +20,19 @@ SELECT lo_unlink(oid) FROM pg_largeobject_metadata WHERE oid >= 1000 AND oid < 3
@@ -1209,8 +1208,8 @@ index 1296da0d57..f43fffa44c 100644
 +CREATE GROUP regress_priv_group2 WITH ADMIN regress_priv_user1 PASSWORD NEON_PASSWORD_PLACEHOLDER USER regress_priv_user2;
 ALTER GROUP regress_priv_group1 ADD USER regress_priv_user4;
 GRANT regress_priv_group2 TO regress_priv_user2 GRANTED BY regress_priv_user1;
- SET SESSION AUTHORIZATION regress_priv_user1;
-@@ -239,12 +239,16 @@ GRANT regress_priv_role TO regress_priv_user1 WITH ADMIN OPTION GRANTED BY regre
+ SET SESSION AUTHORIZATION regress_priv_user3;
+@@ -246,12 +246,16 @@ GRANT regress_priv_role TO regress_priv_user1 WITH ADMIN OPTION GRANTED BY regre
 ERROR:  permission denied to grant privileges as role "regress_priv_role"
 DETAIL:  The grantor must have the ADMIN option on role "regress_priv_role".
 GRANT regress_priv_role TO regress_priv_user1 WITH ADMIN OPTION GRANTED BY CURRENT_ROLE;
@@ -1227,7 +1226,7 @@ index 1296da0d57..f43fffa44c 100644
 DROP ROLE regress_priv_role;
 SET SESSION AUTHORIZATION regress_priv_user1;
 SELECT session_user, current_user;
-@@ -1776,7 +1780,7 @@ SELECT has_table_privilege('regress_priv_user1', 'atest4', 'SELECT WITH GRANT OP
+@@ -1783,7 +1787,7 @@ SELECT has_table_privilege('regress_priv_user1', 'atest4', 'SELECT WITH GRANT OP
 
 -- security-restricted operations
 \c -
@@ -1236,7 +1235,7 @@ index 1296da0d57..f43fffa44c 100644
 -- Check that index expressions and predicates are run as the table's owner
 -- A dummy index function checking current_user
 CREATE FUNCTION sro_ifun(int) RETURNS int AS $$
-@@ -2668,8 +2672,8 @@ drop cascades to function testns.priv_testagg(integer)
+@@ -2675,8 +2679,8 @@ drop cascades to function testns.priv_testagg(integer)
 drop cascades to function testns.priv_testproc(integer)
 -- Change owner of the schema & and rename of new schema owner
 \c -
@@ -1247,7 +1246,7 @@ index 1296da0d57..f43fffa44c 100644
 SET SESSION ROLE regress_schemauser1;
 CREATE SCHEMA testns;
 SELECT nspname, rolname FROM pg_namespace, pg_roles WHERE pg_namespace.nspname = 'testns' AND pg_namespace.nspowner = pg_roles.oid;
-@@ -2792,7 +2796,7 @@ DROP USER regress_priv_user7;
+@@ -2799,7 +2803,7 @@ DROP USER regress_priv_user7;
 DROP USER regress_priv_user8; -- does not exist
 ERROR:  role "regress_priv_user8" does not exist
 -- permissions with LOCK TABLE
@@ -1256,7 +1255,7 @@ index 1296da0d57..f43fffa44c 100644
 CREATE TABLE lock_table (a int);
 -- LOCK TABLE and SELECT permission
 GRANT SELECT ON lock_table TO regress_locktable_user;
-@@ -2888,7 +2892,7 @@ DROP USER regress_locktable_user;
+@@ -2895,7 +2899,7 @@ DROP USER regress_locktable_user;
 -- pg_backend_memory_contexts.
 -- switch to superuser
 \c -
@@ -1265,7 +1264,7 @@ index 1296da0d57..f43fffa44c 100644
 SELECT has_table_privilege('regress_readallstats','pg_backend_memory_contexts','SELECT'); -- no
  has_table_privilege 
 ---------------------
-@@ -2932,10 +2936,10 @@ RESET ROLE;
+@@ -2939,10 +2943,10 @@ RESET ROLE;
 -- clean up
 DROP ROLE regress_readallstats;
 -- test role grantor machinery
@@ -1280,7 +1279,7 @@ index 1296da0d57..f43fffa44c 100644
 GRANT regress_group TO regress_group_direct_manager WITH INHERIT FALSE, ADMIN TRUE;
 GRANT regress_group_direct_manager TO regress_group_indirect_manager;
 SET SESSION AUTHORIZATION regress_group_direct_manager;
-@@ -2964,9 +2968,9 @@ DROP ROLE regress_group_direct_manager;
+@@ -2971,9 +2975,9 @@ DROP ROLE regress_group_direct_manager;
 DROP ROLE regress_group_indirect_manager;
 DROP ROLE regress_group_member;
 -- test SET and INHERIT options with object ownership changes
@@ -1293,7 +1292,7 @@ index 1296da0d57..f43fffa44c 100644
 CREATE SCHEMA regress_roleoption;
 GRANT CREATE, USAGE ON SCHEMA regress_roleoption TO PUBLIC;
 GRANT regress_roleoption_donor TO regress_roleoption_protagonist WITH INHERIT TRUE, SET FALSE;
-@@ -2995,9 +2999,9 @@ DROP ROLE regress_roleoption_protagonist;
+@@ -3002,9 +3006,9 @@ DROP ROLE regress_roleoption_protagonist;
 DROP ROLE regress_roleoption_donor;
 DROP ROLE regress_roleoption_recipient;
 -- MAINTAIN
@@ -2433,10 +2432,10 @@ index e3e3bea709..fa86ddc326 100644
 COMMENT ON CONSTRAINT the_constraint ON constraint_comments_tbl IS 'no, the comment';
 COMMENT ON CONSTRAINT the_constraint ON DOMAIN constraint_comments_dom IS 'no, another comment';
 diff --git a/src/test/regress/sql/conversion.sql b/src/test/regress/sql/conversion.sql
-index 9a65fca91f..58431a3056 100644
+index b567a1a572..4d1ac2e631 100644
 --- a/src/test/regress/sql/conversion.sql
 +++ b/src/test/regress/sql/conversion.sql
-@@ -12,7 +12,7 @@ CREATE FUNCTION test_enc_conversion(bytea, name, name, bool, validlen OUT int, r
+@@ -17,7 +17,7 @@ CREATE FUNCTION test_enc_conversion(bytea, name, name, bool, validlen OUT int, r
     AS :'regresslib', 'test_enc_conversion'
     LANGUAGE C STRICT;
 
@@ -2800,7 +2799,7 @@ index ae6841308b..47bc792e30 100644
 
 SELECT *
 diff --git a/src/test/regress/sql/database.sql b/src/test/regress/sql/database.sql
-index 0367c0e37a..a23b98c4bd 100644
+index 46ad263478..eb05584ed5 100644
 --- a/src/test/regress/sql/database.sql
 +++ b/src/test/regress/sql/database.sql
@@ -1,8 +1,6 @@
@@ -2913,7 +2912,7 @@ index aa147b14a9..370e0dd570 100644
 CREATE FOREIGN DATA WRAPPER dummy;
 COMMENT ON FOREIGN DATA WRAPPER dummy IS 'useless';
 diff --git a/src/test/regress/sql/foreign_key.sql b/src/test/regress/sql/foreign_key.sql
-index 2e710e419c..89cd481a54 100644
+index 8c4e4c7c83..e946cd2119 100644
 --- a/src/test/regress/sql/foreign_key.sql
 +++ b/src/test/regress/sql/foreign_key.sql
@@ -1435,7 +1435,7 @@ ALTER TABLE fk_partitioned_fk_6 ATTACH PARTITION fk_partitioned_pk_6 FOR VALUES
@@ -3301,7 +3300,7 @@ index bb82aa4aa2..dd8a05e24d 100644
 -- Check that the invalid secrets were re-hashed. A re-hashed secret
 -- should not contain the original salt.
 diff --git a/src/test/regress/sql/privileges.sql b/src/test/regress/sql/privileges.sql
-index 5880bc018d..27aa952b18 100644
+index b7e1cb6cdd..6e5a2217f1 100644
 --- a/src/test/regress/sql/privileges.sql
 +++ b/src/test/regress/sql/privileges.sql
@@ -24,18 +24,18 @@ RESET client_min_messages;
@@ -3363,7 +3362,7 @@ index 5880bc018d..27aa952b18 100644
 
 ALTER GROUP regress_priv_group1 ADD USER regress_priv_user4;
 
-@@ -1157,7 +1157,7 @@ SELECT has_table_privilege('regress_priv_user1', 'atest4', 'SELECT WITH GRANT OP
+@@ -1160,7 +1160,7 @@ SELECT has_table_privilege('regress_priv_user1', 'atest4', 'SELECT WITH GRANT OP
 
 -- security-restricted operations
 \c -
@@ -3372,7 +3371,7 @@ index 5880bc018d..27aa952b18 100644
 
 -- Check that index expressions and predicates are run as the table's owner
 
-@@ -1653,8 +1653,8 @@ DROP SCHEMA testns CASCADE;
+@@ -1656,8 +1656,8 @@ DROP SCHEMA testns CASCADE;
 -- Change owner of the schema & and rename of new schema owner
 \c -
 
@@ -3383,7 +3382,7 @@ index 5880bc018d..27aa952b18 100644
 
 SET SESSION ROLE regress_schemauser1;
 CREATE SCHEMA testns;
-@@ -1748,7 +1748,7 @@ DROP USER regress_priv_user8; -- does not exist
+@@ -1751,7 +1751,7 @@ DROP USER regress_priv_user8; -- does not exist
 
 
 -- permissions with LOCK TABLE
@@ -3392,7 +3391,7 @@ index 5880bc018d..27aa952b18 100644
 CREATE TABLE lock_table (a int);
 
 -- LOCK TABLE and SELECT permission
-@@ -1851,7 +1851,7 @@ DROP USER regress_locktable_user;
+@@ -1854,7 +1854,7 @@ DROP USER regress_locktable_user;
 -- switch to superuser
 \c -
 
@@ -3401,7 +3400,7 @@ index 5880bc018d..27aa952b18 100644
 
 SELECT has_table_privilege('regress_readallstats','pg_backend_memory_contexts','SELECT'); -- no
 SELECT has_table_privilege('regress_readallstats','pg_shmem_allocations','SELECT'); -- no
-@@ -1871,10 +1871,10 @@ RESET ROLE;
+@@ -1874,10 +1874,10 @@ RESET ROLE;
 DROP ROLE regress_readallstats;
 
 -- test role grantor machinery
@@ -3416,7 +3415,7 @@ index 5880bc018d..27aa952b18 100644
 
 GRANT regress_group TO regress_group_direct_manager WITH INHERIT FALSE, ADMIN TRUE;
 GRANT regress_group_direct_manager TO regress_group_indirect_manager;
-@@ -1896,9 +1896,9 @@ DROP ROLE regress_group_indirect_manager;
+@@ -1899,9 +1899,9 @@ DROP ROLE regress_group_indirect_manager;
 DROP ROLE regress_group_member;
 
 -- test SET and INHERIT options with object ownership changes
@@ -3429,7 +3428,7 @@ index 5880bc018d..27aa952b18 100644
 CREATE SCHEMA regress_roleoption;
 GRANT CREATE, USAGE ON SCHEMA regress_roleoption TO PUBLIC;
 GRANT regress_roleoption_donor TO regress_roleoption_protagonist WITH INHERIT TRUE, SET FALSE;
-@@ -1926,9 +1926,9 @@ DROP ROLE regress_roleoption_donor;
+@@ -1929,9 +1929,9 @@ DROP ROLE regress_roleoption_donor;
 DROP ROLE regress_roleoption_recipient;
 
 -- MAINTAIN
--- a/compute_tools/src/bin/compute_ctl.rs
+++ b/compute_tools/src/bin/compute_ctl.rs
@@ -29,13 +29,12 @@
 //! ```sh
 //! compute_ctl -D /var/db/postgres/compute \
 //!             -C 'postgresql://cloud_admin@localhost/postgres' \
-//!             -S /var/db/postgres/specs/current.json \
+//!             -c /var/db/postgres/configs/config.json \
 //!             -b /usr/local/bin/postgres \
 //!             -r http://pg-ext-s3-gateway \
 //! ```
 use std::ffi::OsString;
 use std::fs::File;
-use std::path::Path;
 use std::process::exit;
 use std::sync::mpsc;
 use std::thread;
@@ -43,8 +42,7 @@ use std::time::Duration;

 use anyhow::{Context, Result};
 use clap::Parser;
-use compute_api::responses::ComputeCtlConfig;
-use compute_api::spec::ComputeSpec;
+use compute_api::responses::ComputeConfig;
 use compute_tools::compute::{
    BUILD_TAG, ComputeNode, ComputeNodeParams, forward_termination_signal,
 };
@@ -118,16 +116,21 @@ struct Cli {
    #[arg(long)]
    pub set_disk_quota_for_fs: Option<String>,

-    #[arg(short = 's', long = "spec", group = "spec")]
-    pub spec_json: Option<String>,
-
-    #[arg(short = 'S', long, group = "spec-path")]
-    pub spec_path: Option<OsString>,
+    // TODO(tristan957): remove alias after compatibility tests are no longer
+    // an issue
+    #[arg(short = 'c', long, alias = "spec-path")]
+    pub config: Option<OsString>,

    #[arg(short = 'i', long, group = "compute-id")]
    pub compute_id: String,

-    #[arg(short = 'p', long, conflicts_with_all = ["spec", "spec-path"], value_name = "CONTROL_PLANE_API_BASE_URL")]
+    #[arg(
+        short = 'p',
+        long,
+        conflicts_with = "config",
+        value_name = "CONTROL_PLANE_API_BASE_URL",
+        requires = "compute-id"
+    )]
    pub control_plane_uri: Option<String>,
 }

@@ -136,7 +139,7 @@ fn main() -> Result<()> {

    let scenario = failpoint_support::init();

-    // For historical reasons, the main thread that processes the spec and launches postgres
+    // For historical reasons, the main thread that processes the config and launches postgres
    // is synchronous, but we always have this tokio runtime available and we "enter" it so
    // that you can use tokio::spawn() and tokio::runtime::Handle::current().block_on(...)
    // from all parts of compute_ctl.
@@ -152,7 +155,7 @@ fn main() -> Result<()> {

    let connstr = Url::parse(&cli.connstr).context("cannot parse connstr as a URL")?;

-    let cli_spec = try_spec_from_cli(&cli)?;
+    let config = get_config(&cli)?;

    let compute_node = ComputeNode::new(
        ComputeNodeParams {
@@ -172,10 +175,8 @@ fn main() -> Result<()> {
            cgroup: cli.cgroup,
            #[cfg(target_os = "linux")]
            vm_monitor_addr: cli.vm_monitor_addr,
-            live_config_allowed: cli_spec.live_config_allowed,
        },
-        cli_spec.spec,
-        cli_spec.compute_ctl_config,
+        config,
    )?;

    let exit_code = compute_node.run()?;
@@ -200,37 +201,17 @@ async fn init() -> Result<()> {
    Ok(())
 }

-fn try_spec_from_cli(cli: &Cli) -> Result<CliSpecParams> {
-    // First, try to get cluster spec from the cli argument
-    if let Some(ref spec_json) = cli.spec_json {
-        info!("got spec from cli argument {}", spec_json);
-        return Ok(CliSpecParams {
-            spec: Some(serde_json::from_str(spec_json)?),
-            compute_ctl_config: ComputeCtlConfig::default(),
-            live_config_allowed: false,
-        });
+fn get_config(cli: &Cli) -> Result<ComputeConfig> {
+    // First, read the config from the path if provided
+    if let Some(ref config) = cli.config {
+        let file = File::open(config)?;
+        return Ok(serde_json::from_reader(&file)?);
    }

-    // Second, try to read it from the file if path is provided
-    if let Some(ref spec_path) = cli.spec_path {
-        let file = File::open(Path::new(spec_path))?;
-        return Ok(CliSpecParams {
-            spec: Some(serde_json::from_reader(file)?),
-            compute_ctl_config: ComputeCtlConfig::default(),
-            live_config_allowed: true,
-        });
-    }
-
-    if cli.control_plane_uri.is_none() {
-        panic!("must specify --control-plane-uri");
-    };
-
-    match get_spec_from_control_plane(cli.control_plane_uri.as_ref().unwrap(), &cli.compute_id) {
-        Ok(resp) => Ok(CliSpecParams {
-            spec: resp.0,
-            compute_ctl_config: resp.1,
-            live_config_allowed: true,
-        }),
+    // If the config wasn't provided in the CLI arguments, then retrieve it from
+    // the control plane
+    match get_config_from_control_plane(cli.control_plane_uri.as_ref().unwrap(), &cli.compute_id) {
+        Ok(config) => Ok(config),
        Err(e) => {
            error!(
                "cannot get response from control plane: {}\n\
@@ -242,14 +223,6 @@ fn try_spec_from_cli(cli: &Cli) -> Result<CliSpecParams> {
    }
 }

-struct CliSpecParams {
-    /// If a spec was provided via CLI or file, the [`ComputeSpec`]
-    spec: Option<ComputeSpec>,
-    #[allow(dead_code)]
-    compute_ctl_config: ComputeCtlConfig,
-    live_config_allowed: bool,
-}
-
 fn deinit_and_exit(exit_code: Option<i32>) -> ! {
    // Shutdown trace pipeline gracefully, so that it has a chance to send any
    // pending traces before we exit. Shutting down OTEL tracing provider may
--- a/compute_tools/src/catalog.rs
+++ b/compute_tools/src/catalog.rs
@@ -98,13 +98,15 @@ pub async fn get_database_schema(
        .kill_on_drop(true)
        .spawn()?;

-    let stdout = cmd.stdout.take().ok_or_else(|| {
-        std::io::Error::new(std::io::ErrorKind::Other, "Failed to capture stdout.")
-    })?;
+    let stdout = cmd
+        .stdout
+        .take()
+        .ok_or_else(|| std::io::Error::other("Failed to capture stdout."))?;

-    let stderr = cmd.stderr.take().ok_or_else(|| {
-        std::io::Error::new(std::io::ErrorKind::Other, "Failed to capture stderr.")
-    })?;
+    let stderr = cmd
+        .stderr
+        .take()
+        .ok_or_else(|| std::io::Error::other("Failed to capture stderr."))?;

    let mut stdout_reader = FramedRead::new(stdout, BytesCodec::new());
    let stderr_reader = BufReader::new(stderr);
@@ -128,8 +130,7 @@ pub async fn get_database_schema(
                }
            });

-            return Err(SchemaDumpError::IO(std::io::Error::new(
-                std::io::ErrorKind::Other,
+            return Err(SchemaDumpError::IO(std::io::Error::other(
                "failed to start pg_dump",
            )));
        }
--- a/compute_tools/src/compute.rs
+++ b/compute_tools/src/compute.rs
@@ -11,7 +11,7 @@ use std::{env, fs};
 use anyhow::{Context, Result};
 use chrono::{DateTime, Utc};
 use compute_api::privilege::Privilege;
-use compute_api::responses::{ComputeCtlConfig, ComputeMetrics, ComputeStatus};
+use compute_api::responses::{ComputeConfig, ComputeCtlConfig, ComputeMetrics, ComputeStatus};
 use compute_api::spec::{
    ComputeAudit, ComputeFeature, ComputeMode, ComputeSpec, ExtVersion, PgIdent,
 };
@@ -93,20 +93,6 @@ pub struct ComputeNodeParams {

    /// the address of extension storage proxy gateway
    pub ext_remote_storage: Option<String>,
-
-    /// We should only allow live re- / configuration of the compute node if
-    /// it uses 'pull model', i.e. it can go to control-plane and fetch
-    /// the latest configuration. Otherwise, there could be a case:
-    /// - we start compute with some spec provided as argument
-    /// - we push new spec and it does reconfiguration
-    /// - but then something happens and compute pod / VM is destroyed,
-    ///   so k8s controller starts it again with the **old** spec
-    ///
-    /// and the same for empty computes:
-    /// - we started compute without any spec
-    /// - we push spec and it does configuration
-    /// - but then it is restarted without any spec again
-    pub live_config_allowed: bool,
 }

 /// Compute node info shared across several `compute_ctl` threads.
@@ -317,11 +303,7 @@ struct StartVmMonitorResult {
 }

 impl ComputeNode {
-    pub fn new(
-        params: ComputeNodeParams,
-        cli_spec: Option<ComputeSpec>,
-        compute_ctl_config: ComputeCtlConfig,
-    ) -> Result<Self> {
+    pub fn new(params: ComputeNodeParams, config: ComputeConfig) -> Result<Self> {
        let connstr = params.connstr.as_str();
        let conn_conf = postgres::config::Config::from_str(connstr)
            .context("cannot build postgres config from connstr")?;
@@ -329,8 +311,8 @@ impl ComputeNode {
            .context("cannot build tokio postgres config from connstr")?;

        let mut new_state = ComputeState::new();
-        if let Some(cli_spec) = cli_spec {
-            let pspec = ParsedSpec::try_from(cli_spec).map_err(|msg| anyhow::anyhow!(msg))?;
+        if let Some(spec) = config.spec {
+            let pspec = ParsedSpec::try_from(spec).map_err(|msg| anyhow::anyhow!(msg))?;
            new_state.pspec = Some(pspec);
        }

@@ -341,7 +323,7 @@ impl ComputeNode {
            state: Mutex::new(new_state),
            state_changed: Condvar::new(),
            ext_download_progress: RwLock::new(HashMap::new()),
-            compute_ctl_config,
+            compute_ctl_config: config.compute_ctl_config,
        })
    }

@@ -537,11 +519,14 @@ impl ComputeNode {

        let pspec = compute_state.pspec.as_ref().expect("spec must be set");
        info!(
-            "starting compute for project {}, operation {}, tenant {}, timeline {}, features {:?}, spec.remote_extensions {:?}",
+            "starting compute for project {}, operation {}, tenant {}, timeline {}, project {}, branch {}, endpoint {}, features {:?}, spec.remote_extensions {:?}",
            pspec.spec.cluster.cluster_id.as_deref().unwrap_or("None"),
            pspec.spec.operation_uuid.as_deref().unwrap_or("None"),
            pspec.tenant_id,
            pspec.timeline_id,
+            pspec.spec.project_id.as_deref().unwrap_or("None"),
+            pspec.spec.branch_id.as_deref().unwrap_or("None"),
+            pspec.spec.endpoint_id.as_deref().unwrap_or("None"),
            pspec.spec.features,
            pspec.spec.remote_extensions,
        );
@@ -645,31 +630,28 @@ impl ComputeNode {
            });
        }

-        // Configure and start rsyslog for HIPAA if necessary
-        if let ComputeAudit::Hipaa = pspec.spec.audit_log_level {
-            let remote_endpoint = std::env::var("AUDIT_LOGGING_ENDPOINT").unwrap_or("".to_string());
-            if remote_endpoint.is_empty() {
-                anyhow::bail!("AUDIT_LOGGING_ENDPOINT is empty");
+        // Configure and start rsyslog for compliance audit logging
+        match pspec.spec.audit_log_level {
+            ComputeAudit::Hipaa | ComputeAudit::Extended | ComputeAudit::Full => {
+                let remote_endpoint =
+                    std::env::var("AUDIT_LOGGING_ENDPOINT").unwrap_or("".to_string());
+                if remote_endpoint.is_empty() {
+                    anyhow::bail!("AUDIT_LOGGING_ENDPOINT is empty");
+                }
+
+                let log_directory_path = Path::new(&self.params.pgdata).join("log");
+                let log_directory_path = log_directory_path.to_string_lossy().to_string();
+                configure_audit_rsyslog(log_directory_path.clone(), "hipaa", &remote_endpoint)?;
+
+                // Launch a background task to clean up the audit logs
+                launch_pgaudit_gc(log_directory_path);
            }
-
-            let log_directory_path = Path::new(&self.params.pgdata).join("log");
-            let log_directory_path = log_directory_path.to_string_lossy().to_string();
-            configure_audit_rsyslog(log_directory_path.clone(), "hipaa", &remote_endpoint)?;
-
-            // Launch a background task to clean up the audit logs
-            launch_pgaudit_gc(log_directory_path);
+            _ => {}
        }

        // Configure and start rsyslog for Postgres logs export
-        if self.has_feature(ComputeFeature::PostgresLogsExport) {
-            if let Some(ref project_id) = pspec.spec.cluster.cluster_id {
-                let host = PostgresLogsRsyslogConfig::default_host(project_id);
-                let conf = PostgresLogsRsyslogConfig::new(Some(&host));
-                configure_postgres_logs_export(conf)?;
-            } else {
-                warn!("not configuring rsyslog for Postgres logs export: project ID is missing")
-            }
-        }
+        let conf = PostgresLogsRsyslogConfig::new(pspec.spec.logs_export_host.as_deref());
+        configure_postgres_logs_export(conf)?;

        // Launch remaining service threads
        let _monitor_handle = launch_monitor(self);
@@ -1573,6 +1555,10 @@ impl ComputeNode {
            });
        }

+        // Reconfigure rsyslog for Postgres logs export
+        let conf = PostgresLogsRsyslogConfig::new(spec.logs_export_host.as_deref());
+        configure_postgres_logs_export(conf)?;
+
        // Write new config
        let pgdata_path = Path::new(&self.params.pgdata);
        config::write_postgres_conf(
--- a/compute_tools/src/config.rs
+++ b/compute_tools/src/config.rs
@@ -7,7 +7,7 @@ use std::io::prelude::*;
 use std::path::Path;

 use compute_api::responses::TlsConfig;
-use compute_api::spec::{ComputeAudit, ComputeFeature, ComputeMode, ComputeSpec, GenericOption};
+use compute_api::spec::{ComputeAudit, ComputeMode, ComputeSpec, GenericOption};

 use crate::pg_helpers::{
    GenericOptionExt, GenericOptionsSearch, PgOptionsSerialize, escape_conf_value,
@@ -89,6 +89,15 @@ pub fn write_postgres_conf(
            escape_conf_value(&s.to_string())
        )?;
    }
+    if let Some(s) = &spec.project_id {
+        writeln!(file, "neon.project_id={}", escape_conf_value(s))?;
+    }
+    if let Some(s) = &spec.branch_id {
+        writeln!(file, "neon.branch_id={}", escape_conf_value(s))?;
+    }
+    if let Some(s) = &spec.endpoint_id {
+        writeln!(file, "neon.endpoint_id={}", escape_conf_value(s))?;
+    }

    // tls
    if let Some(tls_config) = tls_config {
@@ -169,7 +178,7 @@ pub fn write_postgres_conf(
    // and don't allow the user or the control plane admin to change them.
    match spec.audit_log_level {
        ComputeAudit::Disabled => {}
-        ComputeAudit::Log => {
+        ComputeAudit::Log | ComputeAudit::Base => {
            writeln!(file, "# Managed by compute_ctl base audit settings: start")?;
            writeln!(file, "pgaudit.log='ddl,role'")?;
            // Disable logging of catalog queries to reduce the noise
@@ -193,16 +202,20 @@ pub fn write_postgres_conf(
            }
            writeln!(file, "# Managed by compute_ctl base audit settings: end")?;
        }
-        ComputeAudit::Hipaa => {
+        ComputeAudit::Hipaa | ComputeAudit::Extended | ComputeAudit::Full => {
            writeln!(
                file,
                "# Managed by compute_ctl compliance audit settings: begin"
            )?;
-            // This log level is very verbose
-            // but this is necessary for HIPAA compliance.
-            // Exclude 'misc' category, because it doesn't contain anythig relevant.
-            writeln!(file, "pgaudit.log='all, -misc'")?;
-            writeln!(file, "pgaudit.log_parameter=on")?;
+            // Enable logging of parameters.
+            // This is very verbose and may contain sensitive data.
+            if spec.audit_log_level == ComputeAudit::Full {
+                writeln!(file, "pgaudit.log_parameter=on")?;
+                writeln!(file, "pgaudit.log='all'")?;
+            } else {
+                writeln!(file, "pgaudit.log_parameter=off")?;
+                writeln!(file, "pgaudit.log='all, -misc'")?;
+            }
            // Disable logging of catalog queries
            // The catalog doesn't contain sensitive data, so we don't need to audit it.
            writeln!(file, "pgaudit.log_catalog=off")?;
@@ -255,7 +268,7 @@ pub fn write_postgres_conf(

    // We need Postgres to send logs to rsyslog so that we can forward them
    // further to customers' log aggregation systems.
-    if spec.features.contains(&ComputeFeature::PostgresLogsExport) {
+    if spec.logs_export_host.is_some() {
        writeln!(file, "log_destination='stderr,syslog'")?;
    }

--- a/compute_tools/src/http/middleware/authorize.rs
+++ b/compute_tools/src/http/middleware/authorize.rs
@@ -6,20 +6,15 @@ use axum_extra::{
    TypedHeader,
    headers::{Authorization, authorization::Bearer},
 };
+use compute_api::requests::ComputeClaims;
 use futures::future::BoxFuture;
 use http::{Request, Response, StatusCode};
 use jsonwebtoken::{Algorithm, DecodingKey, TokenData, Validation, jwk::JwkSet};
-use serde::Deserialize;
 use tower_http::auth::AsyncAuthorizeRequest;
-use tracing::warn;
+use tracing::{debug, warn};

 use crate::http::{JsonResponse, extract::RequestId};

-#[derive(Clone, Debug, Deserialize)]
-pub(in crate::http) struct Claims {
-    compute_id: String,
-}
-
 #[derive(Clone, Debug)]
 pub(in crate::http) struct Authorize {
    compute_id: String,
@@ -97,7 +92,7 @@ impl AsyncAuthorizeRequest<Body> for Authorize {
            if data.claims.compute_id != compute_id {
                return Err(JsonResponse::error(
                    StatusCode::UNAUTHORIZED,
-                    "invalid claims in authorization token",
+                    "invalid compute ID in authorization token claims",
                ));
            }

@@ -112,13 +107,19 @@ impl AsyncAuthorizeRequest<Body> for Authorize {

 impl Authorize {
    /// Verify the token using the JSON Web Key set and return the token data.
-    fn verify(jwks: &JwkSet, token: &str, validation: &Validation) -> Result<TokenData<Claims>> {
+    fn verify(
+        jwks: &JwkSet,
+        token: &str,
+        validation: &Validation,
+    ) -> Result<TokenData<ComputeClaims>> {
+        debug!("verifying token {}", token);
+
        for jwk in jwks.keys.iter() {
            let decoding_key = match DecodingKey::from_jwk(jwk) {
                Ok(key) => key,
                Err(e) => {
                    warn!(
-                        "Failed to construct decoding key from {}: {}",
+                        "failed to construct decoding key from {}: {}",
                        jwk.common.key_id.as_ref().unwrap(),
                        e
                    );
@@ -127,11 +128,11 @@ impl Authorize {
                }
            };

-            match jsonwebtoken::decode::<Claims>(token, &decoding_key, validation) {
+            match jsonwebtoken::decode::<ComputeClaims>(token, &decoding_key, validation) {
                Ok(data) => return Ok(data),
                Err(e) => {
                    warn!(
-                        "Failed to decode authorization token using {}: {}",
+                        "failed to decode authorization token using {}: {}",
                        jwk.common.key_id.as_ref().unwrap(),
                        e
                    );
@@ -141,6 +142,6 @@ impl Authorize {
            }
        }

-        Err(anyhow!("Failed to verify authorization token"))
+        Err(anyhow!("failed to verify authorization token"))
    }
 }
--- a/compute_tools/src/http/openapi_spec.yaml
+++ b/compute_tools/src/http/openapi_spec.yaml
@@ -306,36 +306,6 @@ paths:
              schema:
                $ref: "#/components/schemas/GenericError"

-  /configure_telemetry:
-    post:
-      tags:
-        - Configure
-      summary: Configure rsyslog
-      description: |
-        This API endpoint configures rsyslog to forward Postgres logs
-        to a specified otel collector.
-      operationId: configureTelemetry
-      requestBody:
-        required: true
-        content:
-          application/json:
-            schema:
-              type: object
-              properties:
-                logs_export_host:
-                  type: string
-                  description: |
-                    Hostname and the port of the otel collector. Leave empty to disable logs forwarding.
-                    Example: config-shy-breeze-123-collector-monitoring.neon-telemetry.svc.cluster.local:54526
-      responses:
-        204:
-          description: "Telemetry configured successfully"
-        500:
-          content:
-            application/json:
-              schema:
-                $ref: "#/components/schemas/GenericError"
-
 components:
  securitySchemes:
    JWT:
--- a/compute_tools/src/http/routes/configure.rs
+++ b/compute_tools/src/http/routes/configure.rs
@@ -1,11 +1,9 @@
 use std::sync::Arc;

-use axum::body::Body;
 use axum::extract::State;
 use axum::response::Response;
-use compute_api::requests::{ConfigurationRequest, ConfigureTelemetryRequest};
+use compute_api::requests::ConfigurationRequest;
 use compute_api::responses::{ComputeStatus, ComputeStatusResponse};
-use compute_api::spec::ComputeFeature;
 use http::StatusCode;
 use tokio::task;
 use tracing::info;
@@ -13,7 +11,6 @@ use tracing::info;
 use crate::compute::{ComputeNode, ParsedSpec};
 use crate::http::JsonResponse;
 use crate::http::extract::Json;
-use crate::rsyslog::{PostgresLogsRsyslogConfig, configure_postgres_logs_export};

 // Accept spec in JSON format and request compute configuration. If anything
 // goes wrong after we set the compute status to `ConfigurationPending` and
@@ -25,13 +22,6 @@ pub(in crate::http) async fn configure(
    State(compute): State<Arc<ComputeNode>>,
    request: Json<ConfigurationRequest>,
 ) -> Response {
-    if !compute.params.live_config_allowed {
-        return JsonResponse::error(
-            StatusCode::PRECONDITION_FAILED,
-            "live configuration is not allowed for this compute node".to_string(),
-        );
-    }
-
    let pspec = match ParsedSpec::try_from(request.spec.clone()) {
        Ok(p) => p,
        Err(e) => return JsonResponse::error(StatusCode::BAD_REQUEST, e),
@@ -95,25 +85,3 @@ pub(in crate::http) async fn configure(

    JsonResponse::success(StatusCode::OK, body)
 }
-
-pub(in crate::http) async fn configure_telemetry(
-    State(compute): State<Arc<ComputeNode>>,
-    request: Json<ConfigureTelemetryRequest>,
-) -> Response {
-    if !compute.has_feature(ComputeFeature::PostgresLogsExport) {
-        return JsonResponse::error(
-            StatusCode::PRECONDITION_FAILED,
-            "Postgres logs export feature is not enabled".to_string(),
-        );
-    }
-
-    let conf = PostgresLogsRsyslogConfig::new(request.logs_export_host.as_deref());
-    if let Err(err) = configure_postgres_logs_export(conf) {
-        return JsonResponse::error(StatusCode::INTERNAL_SERVER_ERROR, err.to_string());
-    }
-
-    Response::builder()
-        .status(StatusCode::NO_CONTENT)
-        .body(Body::from(""))
-        .unwrap()
-}
--- a/compute_tools/src/http/server.rs
+++ b/compute_tools/src/http/server.rs
@@ -87,7 +87,6 @@ impl From<&Server> for Router<Arc<ComputeNode>> {
                let authenticated_router = Router::<Arc<ComputeNode>>::new()
                    .route("/check_writability", post(check_writability::is_writable))
                    .route("/configure", post(configure::configure))
-                    .route("/configure_telemetry", post(configure::configure_telemetry))
                    .route("/database_schema", get(database_schema::get_schema_dump))
                    .route("/dbs_and_roles", get(dbs_and_roles::get_catalog_objects))
                    .route("/insights", get(insights::get_insights))
--- a/compute_tools/src/metrics.rs
+++ b/compute_tools/src/metrics.rs
@@ -19,13 +19,13 @@ pub(crate) static INSTALLED_EXTENSIONS: Lazy<UIntGaugeVec> = Lazy::new(|| {
 // but for all our APIs we defined a 'slug'/method/operationId in the OpenAPI spec.
 // And it's fair to call it a 'RPC' (Remote Procedure Call).
 pub enum CPlaneRequestRPC {
-    GetSpec,
+    GetConfig,
 }

 impl CPlaneRequestRPC {
    pub fn as_str(&self) -> &str {
        match self {
-            CPlaneRequestRPC::GetSpec => "GetSpec",
+            CPlaneRequestRPC::GetConfig => "GetConfig",
        }
    }
 }
--- a/compute_tools/src/rsyslog.rs
+++ b/compute_tools/src/rsyslog.rs
@@ -119,16 +119,9 @@ impl<'a> PostgresLogsRsyslogConfig<'a> {
        };
        Ok(config_content)
    }
-
-    /// Returns the default host for otel collector that receives Postgres logs
-    pub fn default_host(project_id: &str) -> String {
-        format!(
-            "config-{}-collector.neon-telemetry.svc.cluster.local:10514",
-            project_id
-        )
-    }
 }

+/// Writes rsyslogd configuration for Postgres logs export and restarts rsyslog.
 pub fn configure_postgres_logs_export(conf: PostgresLogsRsyslogConfig) -> Result<()> {
    let new_config = conf.build()?;
    let current_config = PostgresLogsRsyslogConfig::current_config()?;
@@ -261,16 +254,5 @@ mod tests {
            let res = conf.build();
            assert!(res.is_err());
        }
-
-        {
-            // Verify config with default host
-            let host = PostgresLogsRsyslogConfig::default_host("shy-breeze-123");
-            let conf = PostgresLogsRsyslogConfig::new(Some(&host));
-            let res = conf.build();
-            assert!(res.is_ok());
-            let conf_str = res.unwrap();
-            assert!(conf_str.contains(r#"shy-breeze-123"#));
-            assert!(conf_str.contains(r#"port="10514""#));
-        }
    }
 }
--- a/compute_tools/src/spec.rs
+++ b/compute_tools/src/spec.rs
@@ -3,9 +3,8 @@ use std::path::Path;

 use anyhow::{Result, anyhow, bail};
 use compute_api::responses::{
-    ComputeCtlConfig, ControlPlaneComputeStatus, ControlPlaneSpecResponse,
+    ComputeConfig, ControlPlaneComputeStatus, ControlPlaneConfigResponse,
 };
-use compute_api::spec::ComputeSpec;
 use reqwest::StatusCode;
 use tokio_postgres::Client;
 use tracing::{error, info, instrument};
@@ -21,7 +20,7 @@ use crate::params::PG_HBA_ALL_MD5;
 fn do_control_plane_request(
    uri: &str,
    jwt: &str,
-) -> Result<ControlPlaneSpecResponse, (bool, String, String)> {
+) -> Result<ControlPlaneConfigResponse, (bool, String, String)> {
    let resp = reqwest::blocking::Client::new()
        .get(uri)
        .header("Authorization", format!("Bearer {}", jwt))
@@ -29,14 +28,14 @@ fn do_control_plane_request(
        .map_err(|e| {
            (
                true,
-                format!("could not perform spec request to control plane: {:?}", e),
+                format!("could not perform request to control plane: {:?}", e),
                UNKNOWN_HTTP_STATUS.to_string(),
            )
        })?;

    let status = resp.status();
    match status {
-        StatusCode::OK => match resp.json::<ControlPlaneSpecResponse>() {
+        StatusCode::OK => match resp.json::<ControlPlaneConfigResponse>() {
            Ok(spec_resp) => Ok(spec_resp),
            Err(e) => Err((
                true,
@@ -69,40 +68,35 @@ fn do_control_plane_request(
    }
 }

-/// Request spec from the control-plane by compute_id. If `NEON_CONTROL_PLANE_TOKEN`
-/// env variable is set, it will be used for authorization.
-pub fn get_spec_from_control_plane(
-    base_uri: &str,
-    compute_id: &str,
-) -> Result<(Option<ComputeSpec>, ComputeCtlConfig)> {
+/// Request config from the control-plane by compute_id. If
+/// `NEON_CONTROL_PLANE_TOKEN` env variable is set, it will be used for
+/// authorization.
+pub fn get_config_from_control_plane(base_uri: &str, compute_id: &str) -> Result<ComputeConfig> {
    let cp_uri = format!("{base_uri}/compute/api/v2/computes/{compute_id}/spec");
-    let jwt: String = match std::env::var("NEON_CONTROL_PLANE_TOKEN") {
-        Ok(v) => v,
-        Err(_) => "".to_string(),
-    };
+    let jwt: String = std::env::var("NEON_CONTROL_PLANE_TOKEN").unwrap_or_default();
    let mut attempt = 1;

-    info!("getting spec from control plane: {}", cp_uri);
+    info!("getting config from control plane: {}", cp_uri);

    // Do 3 attempts to get spec from the control plane using the following logic:
    // - network error -> then retry
    // - compute id is unknown or any other error -> bail out
    // - no spec for compute yet (Empty state) -> return Ok(None)
-    // - got spec -> return Ok(Some(spec))
+    // - got config -> return Ok(Some(config))
    while attempt < 4 {
        let result = match do_control_plane_request(&cp_uri, &jwt) {
-            Ok(spec_resp) => {
+            Ok(config_resp) => {
                CPLANE_REQUESTS_TOTAL
                    .with_label_values(&[
-                        CPlaneRequestRPC::GetSpec.as_str(),
+                        CPlaneRequestRPC::GetConfig.as_str(),
                        &StatusCode::OK.to_string(),
                    ])
                    .inc();
-                match spec_resp.status {
-                    ControlPlaneComputeStatus::Empty => Ok((None, spec_resp.compute_ctl_config)),
+                match config_resp.status {
+                    ControlPlaneComputeStatus::Empty => Ok(config_resp.into()),
                    ControlPlaneComputeStatus::Attached => {
-                        if let Some(spec) = spec_resp.spec {
-                            Ok((Some(spec), spec_resp.compute_ctl_config))
+                        if config_resp.spec.is_some() {
+                            Ok(config_resp.into())
                        } else {
                            bail!("compute is attached, but spec is empty")
                        }
@@ -111,7 +105,7 @@ pub fn get_spec_from_control_plane(
            }
            Err((retry, msg, status)) => {
                CPLANE_REQUESTS_TOTAL
-                    .with_label_values(&[CPlaneRequestRPC::GetSpec.as_str(), &status])
+                    .with_label_values(&[CPlaneRequestRPC::GetConfig.as_str(), &status])
                    .inc();
                if retry {
                    Err(anyhow!(msg))
@@ -122,7 +116,7 @@ pub fn get_spec_from_control_plane(
        };

        if let Err(e) = &result {
-            error!("attempt {} to get spec failed with: {}", attempt, e);
+            error!("attempt {} to get config failed with: {}", attempt, e);
        } else {
            return result;
        }
@@ -133,13 +127,13 @@ pub fn get_spec_from_control_plane(

    // All attempts failed, return error.
    Err(anyhow::anyhow!(
-        "Exhausted all attempts to retrieve the spec from the control plane"
+        "Exhausted all attempts to retrieve the config from the control plane"
    ))
 }

 /// Check `pg_hba.conf` and update if needed to allow external connections.
 pub fn update_pg_hba(pgdata_path: &Path) -> Result<()> {
-    // XXX: consider making it a part of spec.json
+    // XXX: consider making it a part of config.json
    let pghba_path = pgdata_path.join("pg_hba.conf");

    if config::line_in_file(&pghba_path, PG_HBA_ALL_MD5)? {
@@ -153,7 +147,7 @@ pub fn update_pg_hba(pgdata_path: &Path) -> Result<()> {

 /// Create a standby.signal file
 pub fn add_standby_signal(pgdata_path: &Path) -> Result<()> {
-    // XXX: consider making it a part of spec.json
+    // XXX: consider making it a part of config.json
    let signalfile = pgdata_path.join("standby.signal");

    if !signalfile.exists() {
--- a/compute_tools/src/spec_apply.rs
+++ b/compute_tools/src/spec_apply.rs
@@ -278,12 +278,12 @@ impl ComputeNode {
            // so that all config operations are audit logged.
            match spec.audit_log_level
            {
-                ComputeAudit::Hipaa => {
+                ComputeAudit::Hipaa | ComputeAudit::Extended | ComputeAudit::Full => {
                    phases.push(CreatePgauditExtension);
                    phases.push(CreatePgauditlogtofileExtension);
                    phases.push(DisablePostgresDBPgAudit);
                }
-                ComputeAudit::Log => {
+                ComputeAudit::Log | ComputeAudit::Base => {
                    phases.push(CreatePgauditExtension);
                    phases.push(DisablePostgresDBPgAudit);
                }
--- a/control_plane/src/bin/neon_local.rs
+++ b/control_plane/src/bin/neon_local.rs
@@ -20,8 +20,10 @@ use compute_api::spec::ComputeMode;
 use control_plane::endpoint::ComputeControlPlane;
 use control_plane::local_env::{
    InitForceMode, LocalEnv, NeonBroker, NeonLocalInitConf, NeonLocalInitPageserverConf,
-    SafekeeperConf,
+    ObjectStorageConf, SafekeeperConf,
 };
+use control_plane::object_storage::OBJECT_STORAGE_DEFAULT_PORT;
+use control_plane::object_storage::ObjectStorage;
 use control_plane::pageserver::PageServerNode;
 use control_plane::safekeeper::SafekeeperNode;
 use control_plane::storage_controller::{
@@ -39,7 +41,7 @@ use pageserver_api::controller_api::{
 use pageserver_api::models::{
    ShardParameters, TenantConfigRequest, TimelineCreateRequest, TimelineInfo,
 };
-use pageserver_api::shard::{ShardCount, ShardStripeSize, TenantShardId};
+use pageserver_api::shard::{DEFAULT_STRIPE_SIZE, ShardCount, ShardStripeSize, TenantShardId};
 use postgres_backend::AuthType;
 use postgres_connection::parse_host_port;
 use safekeeper_api::membership::SafekeeperGeneration;
@@ -91,6 +93,8 @@ enum NeonLocalCmd {
    #[command(subcommand)]
    Safekeeper(SafekeeperCmd),
    #[command(subcommand)]
+    ObjectStorage(ObjectStorageCmd),
+    #[command(subcommand)]
    Endpoint(EndpointCmd),
    #[command(subcommand)]
    Mappings(MappingsCmd),
@@ -454,6 +458,32 @@ enum SafekeeperCmd {
    Restart(SafekeeperRestartCmdArgs),
 }

+#[derive(clap::Subcommand)]
+#[clap(about = "Manage object storage")]
+enum ObjectStorageCmd {
+    Start(ObjectStorageStartCmd),
+    Stop(ObjectStorageStopCmd),
+}
+
+#[derive(clap::Args)]
+#[clap(about = "Start object storage")]
+struct ObjectStorageStartCmd {
+    #[clap(short = 't', long, help = "timeout until we fail the command")]
+    #[arg(default_value = "10s")]
+    start_timeout: humantime::Duration,
+}
+
+#[derive(clap::Args)]
+#[clap(about = "Stop object storage")]
+struct ObjectStorageStopCmd {
+    #[arg(value_enum, default_value = "fast")]
+    #[clap(
+        short = 'm',
+        help = "If 'immediate', don't flush repository data at shutdown"
+    )]
+    stop_mode: StopMode,
+}
+
 #[derive(clap::Args)]
 #[clap(about = "Start local safekeeper")]
 struct SafekeeperStartCmdArgs {
@@ -759,6 +789,7 @@ fn main() -> Result<()> {
            }
            NeonLocalCmd::StorageBroker(subcmd) => rt.block_on(handle_storage_broker(&subcmd, env)),
            NeonLocalCmd::Safekeeper(subcmd) => rt.block_on(handle_safekeeper(&subcmd, env)),
+            NeonLocalCmd::ObjectStorage(subcmd) => rt.block_on(handle_object_storage(&subcmd, env)),
            NeonLocalCmd::Endpoint(subcmd) => rt.block_on(handle_endpoint(&subcmd, env)),
            NeonLocalCmd::Mappings(subcmd) => handle_mappings(&subcmd, env),
        };
@@ -975,6 +1006,9 @@ fn handle_init(args: &InitCmdArgs) -> anyhow::Result<LocalEnv> {
                    }
                })
                .collect(),
+            object_storage: ObjectStorageConf {
+                port: OBJECT_STORAGE_DEFAULT_PORT,
+            },
            pg_distrib_dir: None,
            neon_distrib_dir: None,
            default_tenant_id: TenantId::from_array(std::array::from_fn(|_| 0)),
@@ -1083,7 +1117,7 @@ async fn handle_tenant(subcmd: &TenantCmd, env: &mut local_env::LocalEnv) -> any
                        stripe_size: args
                            .shard_stripe_size
                            .map(ShardStripeSize)
-                            .unwrap_or(ShardParameters::DEFAULT_STRIPE_SIZE),
+                            .unwrap_or(DEFAULT_STRIPE_SIZE),
                    },
                    placement_policy: args.placement_policy.clone(),
                    config: tenant_conf,
@@ -1396,7 +1430,7 @@ async fn handle_endpoint(subcmd: &EndpointCmd, env: &local_env::LocalEnv) -> Res
                    vec![(parsed.0, parsed.1.unwrap_or(5432))],
                    // If caller is telling us what pageserver to use, this is not a tenant which is
                    // full managed by storage controller, therefore not sharded.
-                    ShardParameters::DEFAULT_STRIPE_SIZE,
+                    DEFAULT_STRIPE_SIZE,
                )
            } else {
                // Look up the currently attached location of the tenant, and its striping metadata,
@@ -1683,6 +1717,41 @@ async fn handle_safekeeper(subcmd: &SafekeeperCmd, env: &local_env::LocalEnv) ->
    Ok(())
 }

+async fn handle_object_storage(subcmd: &ObjectStorageCmd, env: &local_env::LocalEnv) -> Result<()> {
+    use ObjectStorageCmd::*;
+    let storage = ObjectStorage::from_env(env);
+
+    // In tests like test_forward_compatibility or test_graceful_cluster_restart
+    // old neon binaries (without object_storage) are present
+    if !storage.bin.exists() {
+        eprintln!(
+            "{} binary not found. Ignore if this is a compatibility test",
+            storage.bin
+        );
+        return Ok(());
+    }
+
+    match subcmd {
+        Start(ObjectStorageStartCmd { start_timeout }) => {
+            if let Err(e) = storage.start(start_timeout).await {
+                eprintln!("object_storage start failed: {e}");
+                exit(1);
+            }
+        }
+        Stop(ObjectStorageStopCmd { stop_mode }) => {
+            let immediate = match stop_mode {
+                StopMode::Fast => false,
+                StopMode::Immediate => true,
+            };
+            if let Err(e) = storage.stop(immediate) {
+                eprintln!("proxy stop failed: {e}");
+                exit(1);
+            }
+        }
+    };
+    Ok(())
+}
+
 async fn handle_storage_broker(subcmd: &StorageBrokerCmd, env: &local_env::LocalEnv) -> Result<()> {
    match subcmd {
        StorageBrokerCmd::Start(args) => {
@@ -1777,6 +1846,13 @@ async fn handle_start_all_impl(
                    .map_err(|e| e.context(format!("start safekeeper {}", safekeeper.id)))
            });
        }
+
+        js.spawn(async move {
+            ObjectStorage::from_env(env)
+                .start(&retry_timeout)
+                .await
+                .map_err(|e| e.context("start object_storage"))
+        });
    })();

    let mut errors = Vec::new();
@@ -1874,6 +1950,11 @@ async fn try_stop_all(env: &local_env::LocalEnv, immediate: bool) {
        }
    }

+    let storage = ObjectStorage::from_env(env);
+    if let Err(e) = storage.stop(immediate) {
+        eprintln!("object_storage stop failed: {:#}", e);
+    }
+
    for ps_conf in &env.pageservers {
        let pageserver = PageServerNode::from_env(env, ps_conf);
        if let Err(e) = pageserver.stop(immediate) {
--- a/control_plane/src/endpoint.rs
+++ b/control_plane/src/endpoint.rs
@@ -29,7 +29,7 @@
 //!     compute.log               - log output of `compute_ctl` and `postgres`
 //!     endpoint.json             - serialized `EndpointConf` struct
 //!     postgresql.conf           - postgresql settings
-//!     spec.json                 - passed to `compute_ctl`
+//!     config.json                 - passed to `compute_ctl`
 //!     pgdata/
 //!         postgresql.conf       - copy of postgresql.conf created by `compute_ctl`
 //!         zenith.signal
@@ -46,7 +46,9 @@ use std::time::{Duration, Instant, SystemTime, UNIX_EPOCH};

 use anyhow::{Context, Result, anyhow, bail};
 use compute_api::requests::ConfigurationRequest;
-use compute_api::responses::{ComputeCtlConfig, ComputeStatus, ComputeStatusResponse};
+use compute_api::responses::{
+    ComputeConfig, ComputeCtlConfig, ComputeStatus, ComputeStatusResponse,
+};
 use compute_api::spec::{
    Cluster, ComputeAudit, ComputeFeature, ComputeMode, ComputeSpec, Database, PgIdent,
    RemoteExtSpec, Role,
@@ -619,86 +621,101 @@ impl Endpoint {
            remote_extensions = None;
        };

-        // Create spec file
-        let mut spec = ComputeSpec {
-            skip_pg_catalog_updates: self.skip_pg_catalog_updates,
-            format_version: 1.0,
-            operation_uuid: None,
-            features: self.features.clone(),
-            swap_size_bytes: None,
-            disk_quota_bytes: None,
-            disable_lfc_resizing: None,
-            cluster: Cluster {
-                cluster_id: None, // project ID: not used
-                name: None,       // project name: not used
-                state: None,
-                roles: if create_test_user {
-                    vec![Role {
+        // Create config file
+        let config = {
+            let mut spec = ComputeSpec {
+                skip_pg_catalog_updates: self.skip_pg_catalog_updates,
+                format_version: 1.0,
+                operation_uuid: None,
+                features: self.features.clone(),
+                swap_size_bytes: None,
+                disk_quota_bytes: None,
+                disable_lfc_resizing: None,
+                cluster: Cluster {
+                    cluster_id: None, // project ID: not used
+                    name: None,       // project name: not used
+                    state: None,
+                    roles: if create_test_user {
+                        vec![Role {
+                            name: PgIdent::from_str("test").unwrap(),
+                            encrypted_password: None,
+                            options: None,
+                        }]
+                    } else {
+                        Vec::new()
+                    },
+                    databases: if create_test_user {
+                        vec![Database {
+                            name: PgIdent::from_str("neondb").unwrap(),
+                            owner: PgIdent::from_str("test").unwrap(),
+                            options: None,
+                            restrict_conn: false,
+                            invalid: false,
+                        }]
+                    } else {
+                        Vec::new()
+                    },
+                    settings: None,
+                    postgresql_conf: Some(postgresql_conf.clone()),
+                },
+                delta_operations: None,
+                tenant_id: Some(self.tenant_id),
+                timeline_id: Some(self.timeline_id),
+                project_id: None,
+                branch_id: None,
+                endpoint_id: Some(self.endpoint_id.clone()),
+                mode: self.mode,
+                pageserver_connstring: Some(pageserver_connstring),
+                safekeepers_generation: safekeepers_generation.map(|g| g.into_inner()),
+                safekeeper_connstrings,
+                storage_auth_token: auth_token.clone(),
+                remote_extensions,
+                pgbouncer_settings: None,
+                shard_stripe_size: Some(shard_stripe_size),
+                local_proxy_config: None,
+                reconfigure_concurrency: self.reconfigure_concurrency,
+                drop_subscriptions_before_start: self.drop_subscriptions_before_start,
+                audit_log_level: ComputeAudit::Disabled,
+                logs_export_host: None::<String>,
+            };
+
+            // this strange code is needed to support respec() in tests
+            if self.cluster.is_some() {
+                debug!("Cluster is already set in the endpoint spec, using it");
+                spec.cluster = self.cluster.clone().unwrap();
+
+                debug!("spec.cluster {:?}", spec.cluster);
+
+                // fill missing fields again
+                if create_test_user {
+                    spec.cluster.roles.push(Role {
                        name: PgIdent::from_str("test").unwrap(),
                        encrypted_password: None,
                        options: None,
-                    }]
-                } else {
-                    Vec::new()
-                },
-                databases: if create_test_user {
-                    vec![Database {
+                    });
+                    spec.cluster.databases.push(Database {
                        name: PgIdent::from_str("neondb").unwrap(),
                        owner: PgIdent::from_str("test").unwrap(),
                        options: None,
                        restrict_conn: false,
                        invalid: false,
-                    }]
-                } else {
-                    Vec::new()
-                },
-                settings: None,
-                postgresql_conf: Some(postgresql_conf.clone()),
-            },
-            delta_operations: None,
-            tenant_id: Some(self.tenant_id),
-            timeline_id: Some(self.timeline_id),
-            mode: self.mode,
-            pageserver_connstring: Some(pageserver_connstring),
-            safekeepers_generation: safekeepers_generation.map(|g| g.into_inner()),
-            safekeeper_connstrings,
-            storage_auth_token: auth_token.clone(),
-            remote_extensions,
-            pgbouncer_settings: None,
-            shard_stripe_size: Some(shard_stripe_size),
-            local_proxy_config: None,
-            reconfigure_concurrency: self.reconfigure_concurrency,
-            drop_subscriptions_before_start: self.drop_subscriptions_before_start,
-            audit_log_level: ComputeAudit::Disabled,
+                    });
+                }
+                spec.cluster.postgresql_conf = Some(postgresql_conf);
+            }
+
+            ComputeConfig {
+                spec: Some(spec),
+                compute_ctl_config: ComputeCtlConfig::default(),
+            }
        };

-        // this strange code is needed to support respec() in tests
-        if self.cluster.is_some() {
-            debug!("Cluster is already set in the endpoint spec, using it");
-            spec.cluster = self.cluster.clone().unwrap();
-
-            debug!("spec.cluster {:?}", spec.cluster);
-
-            // fill missing fields again
-            if create_test_user {
-                spec.cluster.roles.push(Role {
-                    name: PgIdent::from_str("test").unwrap(),
-                    encrypted_password: None,
-                    options: None,
-                });
-                spec.cluster.databases.push(Database {
-                    name: PgIdent::from_str("neondb").unwrap(),
-                    owner: PgIdent::from_str("test").unwrap(),
-                    options: None,
-                    restrict_conn: false,
-                    invalid: false,
-                });
-            }
-            spec.cluster.postgresql_conf = Some(postgresql_conf);
-        }
-
+        // TODO(tristan957): Remove the write to spec.json after compatibility
+        // tests work themselves out
        let spec_path = self.endpoint_path().join("spec.json");
-        std::fs::write(spec_path, serde_json::to_string_pretty(&spec)?)?;
+        std::fs::write(spec_path, serde_json::to_string_pretty(&config.spec)?)?;
+        let config_path = self.endpoint_path().join("config.json");
+        std::fs::write(config_path, serde_json::to_string_pretty(&config)?)?;

        // Open log file. We'll redirect the stdout and stderr of `compute_ctl` to it.
        let logfile = std::fs::OpenOptions::new()
@@ -706,6 +723,16 @@ impl Endpoint {
            .append(true)
            .open(self.endpoint_path().join("compute.log"))?;

+        // TODO(tristan957): Remove when compatibility tests are no longer an
+        // issue
+        let old_compute_ctl = {
+            let mut cmd = Command::new(self.env.neon_distrib_dir.join("compute_ctl"));
+            let help_output = cmd.arg("--help").output()?;
+            let help_output = String::from_utf8_lossy(&help_output.stdout);
+
+            !help_output.contains("--config")
+        };
+
        // Launch compute_ctl
        let conn_str = self.connstr("cloud_admin", "postgres");
        println!("Starting postgres node at '{}'", conn_str);
@@ -724,9 +751,18 @@ impl Endpoint {
        ])
        .args(["--pgdata", self.pgdata().to_str().unwrap()])
        .args(["--connstr", &conn_str])
+        // TODO(tristan957): Change this to --config when compatibility tests
+        // are no longer an issue
        .args([
            "--spec-path",
-            self.endpoint_path().join("spec.json").to_str().unwrap(),
+            self.endpoint_path()
+                .join(if old_compute_ctl {
+                    "spec.json"
+                } else {
+                    "config.json"
+                })
+                .to_str()
+                .unwrap(),
        ])
        .args([
            "--pgbin",
@@ -869,10 +905,12 @@ impl Endpoint {
        stripe_size: Option<ShardStripeSize>,
        safekeepers: Option<Vec<NodeId>>,
    ) -> Result<()> {
-        let mut spec: ComputeSpec = {
-            let spec_path = self.endpoint_path().join("spec.json");
-            let file = std::fs::File::open(spec_path)?;
-            serde_json::from_reader(file)?
+        let (mut spec, compute_ctl_config) = {
+            let config_path = self.endpoint_path().join("config.json");
+            let file = std::fs::File::open(config_path)?;
+            let config: ComputeConfig = serde_json::from_reader(file)?;
+
+            (config.spec.unwrap(), config.compute_ctl_config)
        };

        let postgresql_conf = self.read_postgresql_conf()?;
@@ -922,7 +960,7 @@ impl Endpoint {
            .body(
                serde_json::to_string(&ConfigurationRequest {
                    spec,
-                    compute_ctl_config: ComputeCtlConfig::default(),
+                    compute_ctl_config,
                })
                .unwrap(),
            )
--- a/control_plane/src/lib.rs
+++ b/control_plane/src/lib.rs
@@ -10,6 +10,7 @@ mod background_process;
 pub mod broker;
 pub mod endpoint;
 pub mod local_env;
+pub mod object_storage;
 pub mod pageserver;
 pub mod postgresql_conf;
 pub mod safekeeper;
--- a/control_plane/src/local_env.rs
+++ b/control_plane/src/local_env.rs
@@ -15,9 +15,10 @@ use clap::ValueEnum;
 use postgres_backend::AuthType;
 use reqwest::Url;
 use serde::{Deserialize, Serialize};
-use utils::auth::{Claims, encode_from_key_file};
+use utils::auth::encode_from_key_file;
 use utils::id::{NodeId, TenantId, TenantTimelineId, TimelineId};

+use crate::object_storage::{OBJECT_STORAGE_REMOTE_STORAGE_DIR, ObjectStorage};
 use crate::pageserver::{PAGESERVER_REMOTE_STORAGE_DIR, PageServerNode};
 use crate::safekeeper::SafekeeperNode;

@@ -55,6 +56,7 @@ pub struct LocalEnv {

    // used to issue tokens during e.g pg start
    pub private_key_path: PathBuf,
+    pub public_key_path: PathBuf,

    pub broker: NeonBroker,

@@ -68,6 +70,8 @@ pub struct LocalEnv {

    pub safekeepers: Vec<SafekeeperConf>,

+    pub object_storage: ObjectStorageConf,
+
    // Control plane upcall API for pageserver: if None, we will not run storage_controller  If set, this will
    // be propagated into each pageserver's configuration.
    pub control_plane_api: Url,
@@ -95,6 +99,7 @@ pub struct OnDiskConfig {
    pub neon_distrib_dir: PathBuf,
    pub default_tenant_id: Option<TenantId>,
    pub private_key_path: PathBuf,
+    pub public_key_path: PathBuf,
    pub broker: NeonBroker,
    pub storage_controller: NeonStorageControllerConf,
    #[serde(
@@ -103,6 +108,7 @@ pub struct OnDiskConfig {
    )]
    pub pageservers: Vec<PageServerConf>,
    pub safekeepers: Vec<SafekeeperConf>,
+    pub object_storage: ObjectStorageConf,
    pub control_plane_api: Option<Url>,
    pub control_plane_hooks_api: Option<Url>,
    pub control_plane_compute_hook_api: Option<Url>,
@@ -136,11 +142,18 @@ pub struct NeonLocalInitConf {
    pub storage_controller: Option<NeonStorageControllerConf>,
    pub pageservers: Vec<NeonLocalInitPageserverConf>,
    pub safekeepers: Vec<SafekeeperConf>,
+    pub object_storage: ObjectStorageConf,
    pub control_plane_api: Option<Url>,
    pub control_plane_hooks_api: Option<Url>,
    pub generate_local_ssl_certs: bool,
 }

+#[derive(Serialize, Default, Deserialize, PartialEq, Eq, Clone, Debug)]
+#[serde(default)]
+pub struct ObjectStorageConf {
+    pub port: u16,
+}
+
 /// Broker config for cluster internal communication.
 #[derive(Serialize, Deserialize, PartialEq, Eq, Clone, Debug)]
 #[serde(default)]
@@ -398,6 +411,10 @@ impl LocalEnv {
        self.pg_dir(pg_version, "lib")
    }

+    pub fn object_storage_bin(&self) -> PathBuf {
+        self.neon_distrib_dir.join("object_storage")
+    }
+
    pub fn pageserver_bin(&self) -> PathBuf {
        self.neon_distrib_dir.join("pageserver")
    }
@@ -431,6 +448,10 @@ impl LocalEnv {
        self.base_data_dir.join("safekeepers").join(data_dir_name)
    }

+    pub fn object_storage_data_dir(&self) -> PathBuf {
+        self.base_data_dir.join("object_storage")
+    }
+
    pub fn get_pageserver_conf(&self, id: NodeId) -> anyhow::Result<&PageServerConf> {
        if let Some(conf) = self.pageservers.iter().find(|node| node.id == id) {
            Ok(conf)
@@ -582,6 +603,7 @@ impl LocalEnv {
                neon_distrib_dir,
                default_tenant_id,
                private_key_path,
+                public_key_path,
                broker,
                storage_controller,
                pageservers,
@@ -591,6 +613,7 @@ impl LocalEnv {
                control_plane_compute_hook_api: _,
                branch_name_mappings,
                generate_local_ssl_certs,
+                object_storage,
            } = on_disk_config;
            LocalEnv {
                base_data_dir: repopath.to_owned(),
@@ -598,6 +621,7 @@ impl LocalEnv {
                neon_distrib_dir,
                default_tenant_id,
                private_key_path,
+                public_key_path,
                broker,
                storage_controller,
                pageservers,
@@ -606,6 +630,7 @@ impl LocalEnv {
                control_plane_hooks_api,
                branch_name_mappings,
                generate_local_ssl_certs,
+                object_storage,
            }
        };

@@ -705,6 +730,7 @@ impl LocalEnv {
                neon_distrib_dir: self.neon_distrib_dir.clone(),
                default_tenant_id: self.default_tenant_id,
                private_key_path: self.private_key_path.clone(),
+                public_key_path: self.public_key_path.clone(),
                broker: self.broker.clone(),
                storage_controller: self.storage_controller.clone(),
                pageservers: vec![], // it's skip_serializing anyway
@@ -714,6 +740,7 @@ impl LocalEnv {
                control_plane_compute_hook_api: None,
                branch_name_mappings: self.branch_name_mappings.clone(),
                generate_local_ssl_certs: self.generate_local_ssl_certs,
+                object_storage: self.object_storage.clone(),
            },
        )
    }
@@ -730,7 +757,7 @@ impl LocalEnv {
    }

    // this function is used only for testing purposes in CLI e g generate tokens during init
-    pub fn generate_auth_token(&self, claims: &Claims) -> anyhow::Result<String> {
+    pub fn generate_auth_token<S: Serialize>(&self, claims: &S) -> anyhow::Result<String> {
        let private_key_path = self.get_private_key_path();
        let key_data = fs::read(private_key_path)?;
        encode_from_key_file(claims, &key_data)
@@ -797,6 +824,7 @@ impl LocalEnv {
            control_plane_api,
            generate_local_ssl_certs,
            control_plane_hooks_api,
+            object_storage,
        } = conf;

        // Find postgres binaries.
@@ -828,6 +856,7 @@ impl LocalEnv {
        )
        .context("generate auth keys")?;
        let private_key_path = PathBuf::from("auth_private_key.pem");
+        let public_key_path = PathBuf::from("auth_public_key.pem");

        // create the runtime type because the remaining initialization code below needs
        // a LocalEnv instance op operation
@@ -838,6 +867,7 @@ impl LocalEnv {
            neon_distrib_dir,
            default_tenant_id: Some(default_tenant_id),
            private_key_path,
+            public_key_path,
            broker,
            storage_controller: storage_controller.unwrap_or_default(),
            pageservers: pageservers.iter().map(Into::into).collect(),
@@ -846,6 +876,7 @@ impl LocalEnv {
            control_plane_hooks_api,
            branch_name_mappings: Default::default(),
            generate_local_ssl_certs,
+            object_storage,
        };

        if generate_local_ssl_certs {
@@ -873,8 +904,13 @@ impl LocalEnv {
                .context("pageserver init failed")?;
        }

+        ObjectStorage::from_env(&env)
+            .init()
+            .context("object storage init failed")?;
+
        // setup remote remote location for default LocalFs remote storage
        std::fs::create_dir_all(env.base_data_dir.join(PAGESERVER_REMOTE_STORAGE_DIR))?;
+        std::fs::create_dir_all(env.base_data_dir.join(OBJECT_STORAGE_REMOTE_STORAGE_DIR))?;

        env.persist_config()
    }
@@ -944,7 +980,7 @@ fn generate_ssl_ca_cert(cert_path: &Path, key_path: &Path) -> anyhow::Result<()>
    // -out rootCA.crt -keyout rootCA.key
    let keygen_output = Command::new("openssl")
        .args([
-            "req", "-x509", "-newkey", "rsa:2048", "-nodes", "-days", "36500",
+            "req", "-x509", "-newkey", "ed25519", "-nodes", "-days", "36500",
        ])
        .args(["-subj", "/CN=Neon Local CA"])
        .args(["-out", cert_path.to_str().unwrap()])
@@ -974,7 +1010,7 @@ fn generate_ssl_cert(
    // -subj "/CN=localhost" -addext "subjectAltName=DNS:localhost,IP:127.0.0.1"
    let keygen_output = Command::new("openssl")
        .args(["req", "-new", "-nodes"])
-        .args(["-newkey", "rsa:2048"])
+        .args(["-newkey", "ed25519"])
        .args(["-subj", "/CN=localhost"])
        .args(["-addext", "subjectAltName=DNS:localhost,IP:127.0.0.1"])
        .args(["-keyout", key_path.to_str().unwrap()])
--- a/control_plane/src/object_storage.rs
+++ b/control_plane/src/object_storage.rs
@@ -0,0 +1,107 @@
+use crate::background_process::{self, start_process, stop_process};
+use crate::local_env::LocalEnv;
+use anyhow::anyhow;
+use anyhow::{Context, Result};
+use camino::Utf8PathBuf;
+use std::io::Write;
+use std::time::Duration;
+
+/// Directory within .neon which will be used by default for LocalFs remote storage.
+pub const OBJECT_STORAGE_REMOTE_STORAGE_DIR: &str = "local_fs_remote_storage/object_storage";
+pub const OBJECT_STORAGE_DEFAULT_PORT: u16 = 9993;
+
+pub struct ObjectStorage {
+    pub bin: Utf8PathBuf,
+    pub data_dir: Utf8PathBuf,
+    pub pemfile: Utf8PathBuf,
+    pub port: u16,
+}
+
+impl ObjectStorage {
+    pub fn from_env(env: &LocalEnv) -> ObjectStorage {
+        ObjectStorage {
+            bin: Utf8PathBuf::from_path_buf(env.object_storage_bin()).unwrap(),
+            data_dir: Utf8PathBuf::from_path_buf(env.object_storage_data_dir()).unwrap(),
+            pemfile: Utf8PathBuf::from_path_buf(env.public_key_path.clone()).unwrap(),
+            port: env.object_storage.port,
+        }
+    }
+
+    fn config_path(&self) -> Utf8PathBuf {
+        self.data_dir.join("object_storage.json")
+    }
+
+    fn listen_addr(&self) -> Utf8PathBuf {
+        format!("127.0.0.1:{}", self.port).into()
+    }
+
+    pub fn init(&self) -> Result<()> {
+        println!("Initializing object storage in {:?}", self.data_dir);
+        let parent = self.data_dir.parent().unwrap();
+
+        #[derive(serde::Serialize)]
+        struct Cfg {
+            listen: Utf8PathBuf,
+            pemfile: Utf8PathBuf,
+            local_path: Utf8PathBuf,
+            r#type: String,
+        }
+        let cfg = Cfg {
+            listen: self.listen_addr(),
+            pemfile: parent.join(self.pemfile.clone()),
+            local_path: parent.join(OBJECT_STORAGE_REMOTE_STORAGE_DIR),
+            r#type: "LocalFs".to_string(),
+        };
+        std::fs::create_dir_all(self.config_path().parent().unwrap())?;
+        std::fs::write(self.config_path(), serde_json::to_string(&cfg)?)
+            .context("write object storage config")?;
+        Ok(())
+    }
+
+    pub async fn start(&self, retry_timeout: &Duration) -> Result<()> {
+        println!("Starting s3 proxy at {}", self.listen_addr());
+        std::io::stdout().flush().context("flush stdout")?;
+
+        let process_status_check = || async {
+            tokio::time::sleep(Duration::from_millis(500)).await;
+            let res = reqwest::Client::new()
+                .get(format!("http://{}/metrics", self.listen_addr()))
+                .send()
+                .await;
+            match res {
+                Ok(response) if response.status().is_success() => Ok(true),
+                Ok(_) => Err(anyhow!("Failed to query /metrics")),
+                Err(e) => Err(anyhow!("Failed to check node status: {e}")),
+            }
+        };
+
+        let res = start_process(
+            "object_storage",
+            &self.data_dir.clone().into_std_path_buf(),
+            &self.bin.clone().into_std_path_buf(),
+            vec![self.config_path().to_string()],
+            vec![("RUST_LOG".into(), "debug".into())],
+            background_process::InitialPidFile::Create(self.pid_file()),
+            retry_timeout,
+            process_status_check,
+        )
+        .await;
+        if res.is_err() {
+            eprintln!("Logs:\n{}", std::fs::read_to_string(self.log_file())?);
+        }
+
+        res
+    }
+
+    pub fn stop(&self, immediate: bool) -> anyhow::Result<()> {
+        stop_process(immediate, "object_storage", &self.pid_file())
+    }
+
+    fn log_file(&self) -> Utf8PathBuf {
+        self.data_dir.join("object_storage.log")
+    }
+
+    fn pid_file(&self) -> Utf8PathBuf {
+        self.data_dir.join("object_storage.pid")
+    }
+}
--- a/control_plane/src/pageserver.rs
+++ b/control_plane/src/pageserver.rs
@@ -535,6 +535,11 @@ impl PageServerNode {
                .map(|x| x.parse::<bool>())
                .transpose()
                .context("Failed to parse 'gc_compaction_enabled' as bool")?,
+            gc_compaction_verification: settings
+                .remove("gc_compaction_verification")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_verification' as bool")?,
            gc_compaction_initial_threshold_kb: settings
                .remove("gc_compaction_initial_threshold_kb")
                .map(|x| x.parse::<u64>())
@@ -545,6 +550,11 @@ impl PageServerNode {
                .map(|x| x.parse::<u64>())
                .transpose()
                .context("Failed to parse 'gc_compaction_ratio_percent' as integer")?,
+            sampling_ratio: settings
+                .remove("sampling_ratio")
+                .map(serde_json::from_str)
+                .transpose()
+                .context("Falied to parse 'sampling_ratio'")?,
        };
        if !settings.is_empty() {
            bail!("Unrecognized tenant settings: {settings:?}")
--- a/control_plane/src/storage_controller.rs
+++ b/control_plane/src/storage_controller.rs
@@ -13,7 +13,9 @@ use pageserver_api::controller_api::{
    NodeConfigureRequest, NodeDescribeResponse, NodeRegisterRequest, TenantCreateRequest,
    TenantCreateResponse, TenantLocateResponse,
 };
-use pageserver_api::models::{TenantConfigRequest, TimelineCreateRequest, TimelineInfo};
+use pageserver_api::models::{
+    TenantConfig, TenantConfigRequest, TimelineCreateRequest, TimelineInfo,
+};
 use pageserver_api::shard::TenantShardId;
 use pageserver_client::mgmt_api::ResponseErrorMessageExt;
 use postgres_backend::AuthType;
@@ -82,7 +84,8 @@ impl NeonStorageControllerStopArgs {
 pub struct AttachHookRequest {
    pub tenant_shard_id: TenantShardId,
    pub node_id: Option<NodeId>,
-    pub generation_override: Option<i32>,
+    pub generation_override: Option<i32>, // only new tenants
+    pub config: Option<TenantConfig>,     // only new tenants
 }

 #[derive(Serialize, Deserialize)]
@@ -805,6 +808,7 @@ impl StorageController {
            tenant_shard_id,
            node_id: Some(pageserver_id),
            generation_override: None,
+            config: None,
        };

        let response = self
--- a/control_plane/storcon_cli/src/main.rs
+++ b/control_plane/storcon_cli/src/main.rs
@@ -941,7 +941,7 @@ async fn main() -> anyhow::Result<()> {
            let mut node_to_fill_descs = Vec::new();

            for desc in node_descs {
-                let to_drain = nodes.iter().any(|id| *id == desc.id);
+                let to_drain = nodes.contains(&desc.id);
                if to_drain {
                    node_to_drain_descs.push(desc);
                } else {
--- a/docker-compose/compute_wrapper/shell/compute.sh
+++ b/docker-compose/compute_wrapper/shell/compute.sh
@@ -11,8 +11,8 @@ generate_id() {

 PG_VERSION=${PG_VERSION:-14}

-SPEC_FILE_ORG=/var/db/postgres/specs/spec.json
-SPEC_FILE=/tmp/spec.json
+CONFIG_FILE_ORG=/var/db/postgres/configs/config.json
+CONFIG_FILE=/tmp/config.json

 echo "Waiting pageserver become ready."
 while ! nc -z pageserver 6400; do
@@ -20,7 +20,7 @@ while ! nc -z pageserver 6400; do
 done
 echo "Page server is ready."

-cp ${SPEC_FILE_ORG} ${SPEC_FILE}
+cp ${CONFIG_FILE_ORG} ${CONFIG_FILE}

 if [ -n "${TENANT_ID:-}" ] && [ -n "${TIMELINE_ID:-}" ]; then
   tenant_id=${TENANT_ID}
@@ -73,17 +73,27 @@ else
  ulid_extension=ulid
 fi
 echo "Adding pgx_ulid"
-shared_libraries=$(jq -r '.cluster.settings[] | select(.name=="shared_preload_libraries").value' ${SPEC_FILE})
-sed -i "s/${shared_libraries}/${shared_libraries},${ulid_extension}/" ${SPEC_FILE}
+shared_libraries=$(jq -r '.spec.cluster.settings[] | select(.name=="shared_preload_libraries").value' ${CONFIG_FILE})
+sed -i "s/${shared_libraries}/${shared_libraries},${ulid_extension}/" ${CONFIG_FILE}
 echo "Overwrite tenant id and timeline id in spec file"
-sed -i "s/TENANT_ID/${tenant_id}/" ${SPEC_FILE}
-sed -i "s/TIMELINE_ID/${timeline_id}/" ${SPEC_FILE}
+sed -i "s/TENANT_ID/${tenant_id}/" ${CONFIG_FILE}
+sed -i "s/TIMELINE_ID/${timeline_id}/" ${CONFIG_FILE}

-cat ${SPEC_FILE}
+cat ${CONFIG_FILE}
+
+# TODO(tristan957): Remove these workarounds for backwards compatibility after
+# the next compute release. That includes these next few lines and the
+# --spec-path in the compute_ctl invocation.
+if compute_ctl --help | grep --quiet -- '--config'; then
+  SPEC_PATH="$CONFIG_FILE"
+else
+  jq '.spec' < "$CONFIG_FILE" > /tmp/spec.json
+  SPEC_PATH=/tmp/spec.json
+fi

 echo "Start compute node"
 /usr/local/bin/compute_ctl --pgdata /var/db/postgres/compute \
     -C "postgresql://cloud_admin@localhost:55433/postgres"  \
     -b /usr/local/bin/postgres                              \
     --compute-id "compute-$RANDOM"                          \
-     -S ${SPEC_FILE}
+     --spec-path "$SPEC_PATH"
--- a/docker-compose/compute_wrapper/var/db/postgres/configs/config.json
+++ b/docker-compose/compute_wrapper/var/db/postgres/configs/config.json
@@ -0,0 +1,148 @@
+{
+    "spec": {
+        "format_version": 1.0,
+
+        "timestamp": "2022-10-12T18:00:00.000Z",
+        "operation_uuid": "0f657b36-4b0f-4a2d-9c2e-1dcd615e7d8c",
+
+        "cluster": {
+            "cluster_id": "docker_compose",
+            "name": "docker_compose_test",
+            "state": "restarted",
+            "roles": [
+                {
+                    "name": "cloud_admin",
+                    "encrypted_password": "b093c0d3b281ba6da1eacc608620abd8",
+                    "options": null
+                }
+            ],
+            "databases": [
+            ],
+            "settings": [
+                {
+                    "name": "fsync",
+                    "value": "off",
+                    "vartype": "bool"
+                },
+                {
+                    "name": "wal_level",
+                    "value": "logical",
+                    "vartype": "enum"
+                },
+                {
+                    "name": "wal_log_hints",
+                    "value": "on",
+                    "vartype": "bool"
+                },
+                {
+                    "name": "log_connections",
+                    "value": "on",
+                    "vartype": "bool"
+                },
+                {
+                    "name": "port",
+                    "value": "55433",
+                    "vartype": "integer"
+                },
+                {
+                    "name": "shared_buffers",
+                    "value": "1MB",
+                    "vartype": "string"
+                },
+                {
+                    "name": "max_connections",
+                    "value": "100",
+                    "vartype": "integer"
+                },
+                {
+                    "name": "listen_addresses",
+                    "value": "0.0.0.0",
+                    "vartype": "string"
+                },
+                {
+                    "name": "max_wal_senders",
+                    "value": "10",
+                    "vartype": "integer"
+                },
+                {
+                    "name": "max_replication_slots",
+                    "value": "10",
+                    "vartype": "integer"
+                },
+                {
+                    "name": "wal_sender_timeout",
+                    "value": "5s",
+                    "vartype": "string"
+                },
+                {
+                    "name": "wal_keep_size",
+                    "value": "0",
+                    "vartype": "integer"
+                },
+                {
+                    "name": "password_encryption",
+                    "value": "md5",
+                    "vartype": "enum"
+                },
+                {
+                    "name": "restart_after_crash",
+                    "value": "off",
+                    "vartype": "bool"
+                },
+                {
+                    "name": "synchronous_standby_names",
+                    "value": "walproposer",
+                    "vartype": "string"
+                },
+                {
+                    "name": "shared_preload_libraries",
+                    "value": "neon,pg_cron,timescaledb,pg_stat_statements",
+                    "vartype": "string"
+                },
+                {
+                    "name": "neon.safekeepers",
+                    "value": "safekeeper1:5454,safekeeper2:5454,safekeeper3:5454",
+                    "vartype": "string"
+                },
+                {
+                    "name": "neon.timeline_id",
+                    "value": "TIMELINE_ID",
+                    "vartype": "string"
+                },
+                {
+                    "name": "neon.tenant_id",
+                    "value": "TENANT_ID",
+                    "vartype": "string"
+                },
+                {
+                    "name": "neon.pageserver_connstring",
+                    "value": "host=pageserver port=6400",
+                    "vartype": "string"
+                },
+                {
+                    "name": "max_replication_write_lag",
+                    "value": "500MB",
+                    "vartype": "string"
+                },
+                {
+                    "name": "max_replication_flush_lag",
+                    "value": "10GB",
+                    "vartype": "string"
+                },
+                {
+                    "name": "cron.database",
+                    "value": "postgres",
+                    "vartype": "string"
+                }
+            ]
+        },
+
+        "delta_operations": [
+        ]
+    },
+    "compute_ctl_config": {
+        "jwks": {
+            "keys": []
+        }
+    }
+}
--- a/docker-compose/compute_wrapper/var/db/postgres/specs/spec.json
+++ b/docker-compose/compute_wrapper/var/db/postgres/specs/spec.json
@@ -1,141 +0,0 @@
-{
-    "format_version": 1.0,
-
-    "timestamp": "2022-10-12T18:00:00.000Z",
-    "operation_uuid": "0f657b36-4b0f-4a2d-9c2e-1dcd615e7d8c",
-
-    "cluster": {
-        "cluster_id": "docker_compose",
-        "name": "docker_compose_test",
-        "state": "restarted",
-        "roles": [
-            {
-                "name": "cloud_admin",
-                "encrypted_password": "b093c0d3b281ba6da1eacc608620abd8",
-                "options": null
-            }
-        ],
-        "databases": [
-        ],
-        "settings": [
-            {
-                "name": "fsync",
-                "value": "off",
-                "vartype": "bool"
-            },
-            {
-                "name": "wal_level",
-                "value": "logical",
-                "vartype": "enum"
-            },
-            {
-                "name": "wal_log_hints",
-                "value": "on",
-                "vartype": "bool"
-            },
-            {
-                "name": "log_connections",
-                "value": "on",
-                "vartype": "bool"
-            },
-            {
-                "name": "port",
-                "value": "55433",
-                "vartype": "integer"
-            },
-            {
-                "name": "shared_buffers",
-                "value": "1MB",
-                "vartype": "string"
-            },
-            {
-                "name": "max_connections",
-                "value": "100",
-                "vartype": "integer"
-            },
-            {
-                "name": "listen_addresses",
-                "value": "0.0.0.0",
-                "vartype": "string"
-            },
-            {
-                "name": "max_wal_senders",
-                "value": "10",
-                "vartype": "integer"
-            },
-            {
-                "name": "max_replication_slots",
-                "value": "10",
-                "vartype": "integer"
-            },
-            {
-                "name": "wal_sender_timeout",
-                "value": "5s",
-                "vartype": "string"
-            },
-            {
-                "name": "wal_keep_size",
-                "value": "0",
-                "vartype": "integer"
-            },
-            {
-                "name": "password_encryption",
-                "value": "md5",
-                "vartype": "enum"
-            },
-            {
-                "name": "restart_after_crash",
-                "value": "off",
-                "vartype": "bool"
-            },
-            {
-                "name": "synchronous_standby_names",
-                "value": "walproposer",
-                "vartype": "string"
-            },
-            {
-                "name": "shared_preload_libraries",
-                "value": "neon,pg_cron,timescaledb,pg_stat_statements",
-                "vartype": "string"
-            },
-            {
-                "name": "neon.safekeepers",
-                "value": "safekeeper1:5454,safekeeper2:5454,safekeeper3:5454",
-                "vartype": "string"
-            },
-            {
-                "name": "neon.timeline_id",
-                "value": "TIMELINE_ID",
-                "vartype": "string"
-            },
-            {
-                "name": "neon.tenant_id",
-                "value": "TENANT_ID",
-                "vartype": "string"
-            },
-            {
-                "name": "neon.pageserver_connstring",
-                "value": "host=pageserver port=6400",
-                "vartype": "string"
-            },
-            {
-                "name": "max_replication_write_lag",
-                "value": "500MB",
-                "vartype": "string"
-            },
-            {
-                "name": "max_replication_flush_lag",
-                "value": "10GB",
-                "vartype": "string"
-            },
-            {
-                "name": "cron.database",
-                "value": "postgres",
-                "vartype": "string"
-            }
-        ]
-    },
-
-    "delta_operations": [
-    ]
-}
--- a/docker-compose/docker-compose.yml
+++ b/docker-compose/docker-compose.yml
@@ -159,7 +159,7 @@ services:
      #- RUST_BACKTRACE=1
    # Mount the test files directly, for faster editing cycle.
    volumes:
-      - ./compute_wrapper/var/db/postgres/specs/:/var/db/postgres/specs/
+      - ./compute_wrapper/var/db/postgres/configs/:/var/db/postgres/configs/
      - ./compute_wrapper/shell/:/shell/
    ports:
      - 55433:55433 # pg protocol handler
--- a/docker-compose/ext-src/pg_jsonschema-src/Makefile
+++ b/docker-compose/ext-src/pg_jsonschema-src/Makefile
@@ -0,0 +1,8 @@
+EXTENSION = pg_jsonschema
+DATA = pg_jsonschema--1.0.sql
+REGRESS = jsonschema_valid_api  jsonschema_edge_cases
+REGRESS_OPTS = --load-extension=pg_jsonschema
+
+PG_CONFIG ?= pg_config
+PGXS := $(shell $(PG_CONFIG) --pgxs)
+include $(PGXS)
--- a/docker-compose/ext-src/pg_jsonschema-src/expected/jsonschema_edge_cases.out
+++ b/docker-compose/ext-src/pg_jsonschema-src/expected/jsonschema_edge_cases.out
@@ -0,0 +1,87 @@
+-- Schema with enums, nulls, extra properties disallowed
+SELECT jsonschema_is_valid('{
+  "type": "object",
+  "properties": {
+    "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+    "email": { "type": ["string", "null"], "format": "email" }
+  },
+  "required": ["status"],
+  "additionalProperties": false
+}'::json);
+ jsonschema_is_valid 
+---------------------
+ t
+(1 row)
+
+-- Valid enum and null email
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+      "email": { "type": ["string", "null"], "format": "email" }
+    },
+    "required": ["status"],
+    "additionalProperties": false
+  }'::json,
+  '{"status": "active", "email": null}'::json
+);
+ jsonschema_validation_errors 
+------------------------------
+ {}
+(1 row)
+
+-- Invalid enum value
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+      "email": { "type": ["string", "null"], "format": "email" }
+    },
+    "required": ["status"],
+    "additionalProperties": false
+  }'::json,
+  '{"status": "disabled", "email": null}'::json
+);
+                     jsonschema_validation_errors                     
+----------------------------------------------------------------------
+ {"\"disabled\" is not one of [\"active\",\"inactive\",\"pending\"]"}
+(1 row)
+
+-- Invalid email format (assuming format is validated)
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+      "email": { "type": ["string", "null"], "format": "email" }
+    },
+    "required": ["status"],
+    "additionalProperties": false
+  }'::json,
+  '{"status": "active", "email": "not-an-email"}'::json
+);
+      jsonschema_validation_errors       
+-----------------------------------------
+ {"\"not-an-email\" is not a \"email\""}
+(1 row)
+
+-- Extra property not allowed
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+      "email": { "type": ["string", "null"], "format": "email" }
+    },
+    "required": ["status"],
+    "additionalProperties": false
+  }'::json,
+  '{"status": "active", "extra": "should not be here"}'::json
+);
+                    jsonschema_validation_errors                    
+--------------------------------------------------------------------
+ {"Additional properties are not allowed ('extra' was unexpected)"}
+(1 row)
+
--- a/docker-compose/ext-src/pg_jsonschema-src/expected/jsonschema_valid_api.out
+++ b/docker-compose/ext-src/pg_jsonschema-src/expected/jsonschema_valid_api.out
@@ -0,0 +1,65 @@
+-- Define schema
+SELECT jsonschema_is_valid('{
+  "type": "object",
+  "properties": {
+    "username": { "type": "string" },
+    "age": { "type": "integer" }
+  },
+  "required": ["username"]
+}'::json);
+ jsonschema_is_valid 
+---------------------
+ t
+(1 row)
+
+-- Valid instance
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "username": { "type": "string" },
+      "age": { "type": "integer" }
+    },
+    "required": ["username"]
+  }'::json,
+  '{"username": "alice", "age": 25}'::json
+);
+ jsonschema_validation_errors 
+------------------------------
+ {}
+(1 row)
+
+-- Invalid instance: missing required "username"
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "username": { "type": "string" },
+      "age": { "type": "integer" }
+    },
+    "required": ["username"]
+  }'::json,
+  '{"age": 25}'::json
+);
+      jsonschema_validation_errors       
+-----------------------------------------
+ {"\"username\" is a required property"}
+(1 row)
+
+-- Invalid instance: wrong type for "age"
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "username": { "type": "string" },
+      "age": { "type": "integer" }
+    },
+    "required": ["username"]
+  }'::json,
+  '{"username": "bob", "age": "twenty"}'::json
+);
+       jsonschema_validation_errors        
+-------------------------------------------
+ {"\"twenty\" is not of type \"integer\""}
+(1 row)
+
--- a/docker-compose/ext-src/pg_jsonschema-src/sql/jsonschema_edge_cases.sql
+++ b/docker-compose/ext-src/pg_jsonschema-src/sql/jsonschema_edge_cases.sql
@@ -0,0 +1,66 @@
+-- Schema with enums, nulls, extra properties disallowed
+SELECT jsonschema_is_valid('{
+  "type": "object",
+  "properties": {
+    "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+    "email": { "type": ["string", "null"], "format": "email" }
+  },
+  "required": ["status"],
+  "additionalProperties": false
+}'::json);
+
+-- Valid enum and null email
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+      "email": { "type": ["string", "null"], "format": "email" }
+    },
+    "required": ["status"],
+    "additionalProperties": false
+  }'::json,
+  '{"status": "active", "email": null}'::json
+);
+
+-- Invalid enum value
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+      "email": { "type": ["string", "null"], "format": "email" }
+    },
+    "required": ["status"],
+    "additionalProperties": false
+  }'::json,
+  '{"status": "disabled", "email": null}'::json
+);
+
+-- Invalid email format (assuming format is validated)
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+      "email": { "type": ["string", "null"], "format": "email" }
+    },
+    "required": ["status"],
+    "additionalProperties": false
+  }'::json,
+  '{"status": "active", "email": "not-an-email"}'::json
+);
+
+-- Extra property not allowed
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "status": { "type": "string", "enum": ["active", "inactive", "pending"] },
+      "email": { "type": ["string", "null"], "format": "email" }
+    },
+    "required": ["status"],
+    "additionalProperties": false
+  }'::json,
+  '{"status": "active", "extra": "should not be here"}'::json
+);
--- a/docker-compose/ext-src/pg_jsonschema-src/sql/jsonschema_valid_api.sql
+++ b/docker-compose/ext-src/pg_jsonschema-src/sql/jsonschema_valid_api.sql
@@ -0,0 +1,48 @@
+-- Define schema
+SELECT jsonschema_is_valid('{
+  "type": "object",
+  "properties": {
+    "username": { "type": "string" },
+    "age": { "type": "integer" }
+  },
+  "required": ["username"]
+}'::json);
+
+-- Valid instance
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "username": { "type": "string" },
+      "age": { "type": "integer" }
+    },
+    "required": ["username"]
+  }'::json,
+  '{"username": "alice", "age": 25}'::json
+);
+
+-- Invalid instance: missing required "username"
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "username": { "type": "string" },
+      "age": { "type": "integer" }
+    },
+    "required": ["username"]
+  }'::json,
+  '{"age": 25}'::json
+);
+
+-- Invalid instance: wrong type for "age"
+SELECT jsonschema_validation_errors(
+  '{
+    "type": "object",
+    "properties": {
+      "username": { "type": "string" },
+      "age": { "type": "integer" }
+    },
+    "required": ["username"]
+  }'::json,
+  '{"username": "bob", "age": "twenty"}'::json
+);
--- a/docker-compose/ext-src/pg_session_jwt-src/Makefile
+++ b/docker-compose/ext-src/pg_session_jwt-src/Makefile
@@ -0,0 +1,9 @@
+EXTENSION = pg_session_jwt
+
+REGRESS = basic_functions
+REGRESS_OPTS = --load-extension=$(EXTENSION)
+export PGOPTIONS = -c pg_session_jwt.jwk={"crv":"Ed25519","kty":"OKP","x":"R_Abz-63zJ00l-IraL5fQhwkhGVZCSooQFV5ntC3C7M"}
+
+PG_CONFIG ?= pg_config
+PGXS := $(shell $(PG_CONFIG) --pgxs)
+include $(PGXS)
--- a/docker-compose/ext-src/pg_session_jwt-src/expected/basic_functions.out
+++ b/docker-compose/ext-src/pg_session_jwt-src/expected/basic_functions.out
@@ -0,0 +1,35 @@
+-- Basic functionality tests for pg_session_jwt
+-- Test auth.init() function
+SELECT auth.init();
+ init 
+------
+ 
+(1 row)
+
+-- Test an invalid JWT
+SELECT auth.jwt_session_init('INVALID-JWT');
+ERROR:  invalid JWT encoding
+-- Test creating a session with an expired JWT
+SELECT auth.jwt_session_init('eyJhbGciOiJFZERTQSJ9.eyJleHAiOjE3NDI1NjQ0MzIsImlhdCI6MTc0MjU2NDI1MiwianRpIjo0MjQyNDIsInN1YiI6InVzZXIxMjMifQ.A6FwKuaSduHB9O7Gz37g0uoD_U9qVS0JNtT7YABGVgB7HUD1AMFc9DeyhNntWBqncg8k5brv-hrNTuUh5JYMAw');
+ERROR:  Token used after it has expired
+-- Test creating a session with a valid JWT
+SELECT auth.jwt_session_init('eyJhbGciOiJFZERTQSJ9.eyJleHAiOjQ4OTYxNjQyNTIsImlhdCI6MTc0MjU2NDI1MiwianRpIjo0MzQzNDMsInN1YiI6InVzZXIxMjMifQ.2TXVgjb6JSUq6_adlvp-m_SdOxZSyGS30RS9TLB0xu2N83dMSs2NybwE1NMU8Fb0tcAZR_ET7M2rSxbTrphfCg');
+ jwt_session_init 
+------------------
+ 
+(1 row)
+
+-- Test auth.session() function
+SELECT auth.session();
+                                 session                                 
+-------------------------------------------------------------------------
+ {"exp": 4896164252, "iat": 1742564252, "jti": 434343, "sub": "user123"}
+(1 row)
+
+-- Test auth.user_id() function
+SELECT auth.user_id() AS user_id;
+ user_id 
+---------
+ user123
+(1 row)
+
--- a/docker-compose/ext-src/pg_session_jwt-src/sql/basic_functions.sql
+++ b/docker-compose/ext-src/pg_session_jwt-src/sql/basic_functions.sql
@@ -0,0 +1,19 @@
+-- Basic functionality tests for pg_session_jwt
+
+-- Test auth.init() function
+SELECT auth.init();
+
+-- Test an invalid JWT
+SELECT auth.jwt_session_init('INVALID-JWT');
+
+-- Test creating a session with an expired JWT
+SELECT auth.jwt_session_init('eyJhbGciOiJFZERTQSJ9.eyJleHAiOjE3NDI1NjQ0MzIsImlhdCI6MTc0MjU2NDI1MiwianRpIjo0MjQyNDIsInN1YiI6InVzZXIxMjMifQ.A6FwKuaSduHB9O7Gz37g0uoD_U9qVS0JNtT7YABGVgB7HUD1AMFc9DeyhNntWBqncg8k5brv-hrNTuUh5JYMAw');
+
+-- Test creating a session with a valid JWT
+SELECT auth.jwt_session_init('eyJhbGciOiJFZERTQSJ9.eyJleHAiOjQ4OTYxNjQyNTIsImlhdCI6MTc0MjU2NDI1MiwianRpIjo0MzQzNDMsInN1YiI6InVzZXIxMjMifQ.2TXVgjb6JSUq6_adlvp-m_SdOxZSyGS30RS9TLB0xu2N83dMSs2NybwE1NMU8Fb0tcAZR_ET7M2rSxbTrphfCg');
+
+-- Test auth.session() function
+SELECT auth.session();
+
+-- Test auth.user_id() function
+SELECT auth.user_id() AS user_id;
--- a/docs/storage_controller.md
+++ b/docs/storage_controller.md
@@ -151,7 +151,7 @@ Example body:
 ```
 {
  "tenant_id": "1f359dd625e519a1a4e8d7509690f6fc",
-  "stripe_size": 32768,
+  "stripe_size": 2048,
  "shards": [
      {"node_id": 344, "shard_number": 0},
      {"node_id": 722, "shard_number": 1},
--- a/libs/compute_api/src/requests.rs
+++ b/libs/compute_api/src/requests.rs
@@ -5,6 +5,14 @@ use crate::privilege::Privilege;
 use crate::responses::ComputeCtlConfig;
 use crate::spec::{ComputeSpec, ExtVersion, PgIdent};

+/// When making requests to the `compute_ctl` external HTTP server, the client
+/// must specify a set of claims in `Authorization` header JWTs such that
+/// `compute_ctl` can authorize the request.
+#[derive(Clone, Debug, Deserialize, Serialize)]
+pub struct ComputeClaims {
+    pub compute_id: String,
+}
+
 /// Request of the /configure API
 ///
 /// We now pass only `spec` in the configuration request, but later we can
@@ -30,9 +38,3 @@ pub struct SetRoleGrantsRequest {
    pub privileges: Vec<Privilege>,
    pub role: PgIdent,
 }
-
-/// Request of the /configure_telemetry API
-#[derive(Debug, Deserialize, Serialize)]
-pub struct ConfigureTelemetryRequest {
-    pub logs_export_host: Option<String>,
-}
--- a/libs/compute_api/src/responses.rs
+++ b/libs/compute_api/src/responses.rs
@@ -14,6 +14,32 @@ pub struct GenericAPIError {
    pub error: String,
 }

+/// All configuration parameters necessary for a compute. When
+/// [`ComputeConfig::spec`] is provided, it means that the compute is attached
+/// to a tenant. [`ComputeConfig::compute_ctl_config`] will always be provided
+/// and contains parameters necessary for operating `compute_ctl` independently
+/// of whether a tenant is attached to the compute or not.
+///
+/// This also happens to be the body of `compute_ctl`'s /configure request.
+#[derive(Debug, Deserialize, Serialize)]
+pub struct ComputeConfig {
+    /// The compute spec
+    pub spec: Option<ComputeSpec>,
+
+    /// The compute_ctl configuration
+    #[allow(dead_code)]
+    pub compute_ctl_config: ComputeCtlConfig,
+}
+
+impl From<ControlPlaneConfigResponse> for ComputeConfig {
+    fn from(value: ControlPlaneConfigResponse) -> Self {
+        Self {
+            spec: value.spec,
+            compute_ctl_config: value.compute_ctl_config,
+        }
+    }
+}
+
 #[derive(Debug, Clone, Serialize)]
 pub struct ExtensionInstallResponse {
    pub extension: PgIdent,
@@ -161,7 +187,7 @@ pub struct TlsConfig {

 /// Response of the `/computes/{compute_id}/spec` control-plane API.
 #[derive(Deserialize, Debug)]
-pub struct ControlPlaneSpecResponse {
+pub struct ControlPlaneConfigResponse {
    pub spec: Option<ComputeSpec>,
    pub status: ControlPlaneComputeStatus,
    pub compute_ctl_config: ComputeCtlConfig,
--- a/libs/compute_api/src/spec.rs
+++ b/libs/compute_api/src/spec.rs
@@ -1,8 +1,8 @@
-//! `ComputeSpec` represents the contents of the spec.json file.
-//!
-//! The spec.json file is used to pass information to 'compute_ctl'. It contains
-//! all the information needed to start up the right version of PostgreSQL,
-//! and connect it to the storage nodes.
+//! The ComputeSpec contains all the information needed to start up
+//! the right version of PostgreSQL, and connect it to the storage nodes.
+//! It can be passed as part of the `config.json`, or the control plane can
+//! provide it by calling the compute_ctl's `/compute_ctl` endpoint, or
+//! compute_ctl can fetch it by calling the control plane's API.
 use std::collections::HashMap;

 use indexmap::IndexMap;
@@ -104,6 +104,12 @@ pub struct ComputeSpec {
    pub timeline_id: Option<TimelineId>,
    pub pageserver_connstring: Option<String>,

+    // More neon ids that we expose to the compute_ctl
+    // and to postgres as neon extension GUCs.
+    pub project_id: Option<String>,
+    pub branch_id: Option<String>,
+    pub endpoint_id: Option<String>,
+
    /// Safekeeper membership config generation. It is put in
    /// neon.safekeepers GUC and serves two purposes:
    /// 1) Non zero value forces walproposer to use membership configurations.
@@ -159,15 +165,13 @@ pub struct ComputeSpec {
    #[serde(default)] // Default false
    pub drop_subscriptions_before_start: bool,

-    /// Log level for audit logging:
-    ///
-    /// Disabled - no audit logging. This is the default.
-    /// log - log masked statements to the postgres log using pgaudit extension
-    /// hipaa - log unmasked statements to the file using pgaudit and pgauditlogtofile extension
-    ///
-    /// Extensions should be present in shared_preload_libraries
+    /// Log level for compute audit logging
    #[serde(default)]
    pub audit_log_level: ComputeAudit,
+
+    /// Hostname and the port of the otel collector. Leave empty to disable Postgres logs forwarding.
+    /// Example: config-shy-breeze-123-collector-monitoring.neon-telemetry.svc.cluster.local:10514
+    pub logs_export_host: Option<String>,
 }

 /// Feature flag to signal `compute_ctl` to enable certain experimental functionality.
@@ -179,9 +183,6 @@ pub enum ComputeFeature {
    /// track short-lived connections as user activity.
    ActivityMonitorExperimental,

-    /// Allow to configure rsyslog for Postgres logs export
-    PostgresLogsExport,
-
    /// This is a special feature flag that is used to represent unknown feature flags.
    /// Basically all unknown to enum flags are represented as this one. See unit test
    /// `parse_unknown_features()` for more details.
@@ -288,14 +289,25 @@ impl ComputeMode {
 }

 /// Log level for audit logging
-/// Disabled, log, hipaa
-/// Default is Disabled
 #[derive(Clone, Debug, Default, Eq, PartialEq, Deserialize, Serialize)]
 pub enum ComputeAudit {
    #[default]
    Disabled,
+    // Deprecated, use Base instead
    Log,
+    // (pgaudit.log = 'ddl', pgaudit.log_parameter='off')
+    // logged to the standard postgresql log stream
+    Base,
+    // Deprecated, use Full or Extended instead
    Hipaa,
+    // (pgaudit.log = 'all, -misc', pgaudit.log_parameter='off')
+    // logged to separate files collected by rsyslog
+    // into dedicated log storage with strict access
+    Extended,
+    // (pgaudit.log='all', pgaudit.log_parameter='on'),
+    // logged to separate files collected by rsyslog
+    // into dedicated log storage with strict access.
+    Full,
 }

 #[derive(Clone, Debug, Default, Deserialize, Serialize, PartialEq, Eq)]
--- a/libs/http-utils/Cargo.toml
+++ b/libs/http-utils/Cargo.toml
@@ -30,6 +30,7 @@ tokio.workspace = true
 tracing.workspace = true
 url.workspace = true
 uuid.workspace = true
+x509-cert.workspace = true

 # to use tokio channels as streams, this is faster to compile than async_stream
 # why is it only here? no other crate should use it, streams are rarely needed.
--- a/libs/http-utils/src/server.rs
+++ b/libs/http-utils/src/server.rs
@@ -4,6 +4,8 @@ use futures::StreamExt;
 use futures::stream::FuturesUnordered;
 use hyper0::Body;
 use hyper0::server::conn::Http;
+use metrics::{IntCounterVec, register_int_counter_vec};
+use once_cell::sync::Lazy;
 use routerify::{RequestService, RequestServiceBuilder};
 use tokio::io::{AsyncRead, AsyncWrite};
 use tokio_rustls::TlsAcceptor;
@@ -26,6 +28,24 @@ pub struct Server {
    tls_acceptor: Option<TlsAcceptor>,
 }

+static CONNECTION_STARTED_COUNT: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "http_server_connection_started_total",
+        "Number of established http/https connections",
+        &["scheme"]
+    )
+    .expect("failed to define a metric")
+});
+
+static CONNECTION_ERROR_COUNT: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "http_server_connection_errors_total",
+        "Number of occured connection errors by type",
+        &["type"]
+    )
+    .expect("failed to define a metric")
+});
+
 impl Server {
    pub fn new(
        request_service: Arc<RequestServiceBuilder<Body, ApiError>>,
@@ -60,6 +80,15 @@ impl Server {
            false
        }

+        let tcp_error_cnt = CONNECTION_ERROR_COUNT.with_label_values(&["tcp"]);
+        let tls_error_cnt = CONNECTION_ERROR_COUNT.with_label_values(&["tls"]);
+        let http_error_cnt = CONNECTION_ERROR_COUNT.with_label_values(&["http"]);
+        let https_error_cnt = CONNECTION_ERROR_COUNT.with_label_values(&["https"]);
+        let panic_error_cnt = CONNECTION_ERROR_COUNT.with_label_values(&["panic"]);
+
+        let http_connection_cnt = CONNECTION_STARTED_COUNT.with_label_values(&["http"]);
+        let https_connection_cnt = CONNECTION_STARTED_COUNT.with_label_values(&["https"]);
+
        let mut connections = FuturesUnordered::new();
        loop {
            tokio::select! {
@@ -67,6 +96,7 @@ impl Server {
                    let (tcp_stream, remote_addr) = match stream {
                        Ok(stream) => stream,
                        Err(err) => {
+                            tcp_error_cnt.inc();
                            if !suppress_io_error(&err) {
                                info!("Failed to accept TCP connection: {err:#}");
                            }
@@ -78,11 +108,18 @@ impl Server {
                    let tls_acceptor = self.tls_acceptor.clone();
                    let cancel = cancel.clone();

+                    let tls_error_cnt = tls_error_cnt.clone();
+                    let http_error_cnt = http_error_cnt.clone();
+                    let https_error_cnt = https_error_cnt.clone();
+                    let http_connection_cnt = http_connection_cnt.clone();
+                    let https_connection_cnt = https_connection_cnt.clone();
+
                    connections.push(tokio::spawn(
                        async move {
                            match tls_acceptor {
                                Some(tls_acceptor) => {
                                    // Handle HTTPS connection.
+                                    https_connection_cnt.inc();
                                    let tls_stream = tokio::select! {
                                        tls_stream = tls_acceptor.accept(tcp_stream) => tls_stream,
                                        _ = cancel.cancelled() => return,
@@ -90,6 +127,7 @@ impl Server {
                                    let tls_stream = match tls_stream {
                                        Ok(tls_stream) => tls_stream,
                                        Err(err) => {
+                                            tls_error_cnt.inc();
                                            if !suppress_io_error(&err) {
                                                info!(%remote_addr, "Failed to accept TLS connection: {err:#}");
                                            }
@@ -97,6 +135,7 @@ impl Server {
                                        }
                                    };
                                    if let Err(err) = Self::serve_connection(tls_stream, service, cancel).await {
+                                        https_error_cnt.inc();
                                        if !suppress_hyper_error(&err) {
                                            info!(%remote_addr, "Failed to serve HTTPS connection: {err:#}");
                                        }
@@ -104,7 +143,9 @@ impl Server {
                                }
                                None => {
                                    // Handle HTTP connection.
+                                    http_connection_cnt.inc();
                                    if let Err(err) = Self::serve_connection(tcp_stream, service, cancel).await {
+                                        http_error_cnt.inc();
                                        if !suppress_hyper_error(&err) {
                                            info!(%remote_addr, "Failed to serve HTTP connection: {err:#}");
                                        }
@@ -115,6 +156,7 @@ impl Server {
                 }
                Some(conn) = connections.next() => {
                    if let Err(err) = conn {
+                        panic_error_cnt.inc();
                        error!("Connection panicked: {err:#}");
                    }
                }
@@ -122,6 +164,7 @@ impl Server {
                    // Wait for graceful shutdown of all connections.
                    while let Some(conn) = connections.next().await {
                        if let Err(err) = conn {
+                            panic_error_cnt.inc();
                            error!("Connection panicked: {err:#}");
                        }
                    }
--- a/libs/http-utils/src/tls_certs.rs
+++ b/libs/http-utils/src/tls_certs.rs
@@ -3,11 +3,14 @@ use std::{sync::Arc, time::Duration};
 use anyhow::Context;
 use arc_swap::ArcSwap;
 use camino::Utf8Path;
+use metrics::{IntCounterVec, UIntGaugeVec, register_int_counter_vec, register_uint_gauge_vec};
+use once_cell::sync::Lazy;
 use rustls::{
-    pki_types::{CertificateDer, PrivateKeyDer},
+    pki_types::{CertificateDer, PrivateKeyDer, UnixTime},
    server::{ClientHello, ResolvesServerCert},
    sign::CertifiedKey,
 };
+use x509_cert::der::Reader;

 pub async fn load_cert_chain(filename: &Utf8Path) -> anyhow::Result<Vec<CertificateDer<'static>>> {
    let cert_data = tokio::fs::read(filename)
@@ -53,6 +56,76 @@ pub async fn load_certified_key(
    Ok(certified_key)
 }

+/// rustls's CertifiedKey with extra parsed fields used for metrics.
+struct ParsedCertifiedKey {
+    certified_key: CertifiedKey,
+    expiration_time: UnixTime,
+}
+
+/// Parse expiration time from an X509 certificate.
+fn parse_expiration_time(cert: &CertificateDer<'_>) -> anyhow::Result<UnixTime> {
+    let parsed_cert = x509_cert::der::SliceReader::new(cert)
+        .context("Failed to parse cerficiate")?
+        .decode::<x509_cert::Certificate>()
+        .context("Failed to parse cerficiate")?;
+
+    Ok(UnixTime::since_unix_epoch(
+        parsed_cert
+            .tbs_certificate
+            .validity
+            .not_after
+            .to_unix_duration(),
+    ))
+}
+
+async fn load_and_parse_certified_key(
+    key_filename: &Utf8Path,
+    cert_filename: &Utf8Path,
+) -> anyhow::Result<ParsedCertifiedKey> {
+    let certified_key = load_certified_key(key_filename, cert_filename).await?;
+    let expiration_time = parse_expiration_time(certified_key.end_entity_cert()?)?;
+    Ok(ParsedCertifiedKey {
+        certified_key,
+        expiration_time,
+    })
+}
+
+static CERT_EXPIRATION_TIME: Lazy<UIntGaugeVec> = Lazy::new(|| {
+    register_uint_gauge_vec!(
+        "tls_certs_expiration_time_seconds",
+        "Expiration time of the loaded certificate since unix epoch in seconds",
+        &["resolver_name"]
+    )
+    .expect("failed to define a metric")
+});
+
+static CERT_RELOAD_STARTED_COUNTER: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "tls_certs_reload_started_total",
+        "Number of certificate reload loop iterations started",
+        &["resolver_name"]
+    )
+    .expect("failed to define a metric")
+});
+
+static CERT_RELOAD_UPDATED_COUNTER: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "tls_certs_reload_updated_total",
+        "Number of times the certificate was updated to the new one",
+        &["resolver_name"]
+    )
+    .expect("failed to define a metric")
+});
+
+static CERT_RELOAD_FAILED_COUNTER: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "tls_certs_reload_failed_total",
+        "Number of times the certificate reload failed",
+        &["resolver_name"]
+    )
+    .expect("failed to define a metric")
+});
+
 /// Implementation of [`rustls::server::ResolvesServerCert`] which reloads certificates from
 /// the disk periodically.
 #[derive(Debug)]
@@ -63,16 +136,28 @@ pub struct ReloadingCertificateResolver {
 impl ReloadingCertificateResolver {
    /// Creates a new Resolver by loading certificate and private key from FS and
    /// creating tokio::task to reload them with provided reload_period.
+    /// resolver_name is used as metric's label.
    pub async fn new(
+        resolver_name: &str,
        key_filename: &Utf8Path,
        cert_filename: &Utf8Path,
        reload_period: Duration,
    ) -> anyhow::Result<Arc<Self>> {
+        // Create metrics for current resolver.
+        let cert_expiration_time = CERT_EXPIRATION_TIME.with_label_values(&[resolver_name]);
+        let cert_reload_started_counter =
+            CERT_RELOAD_STARTED_COUNTER.with_label_values(&[resolver_name]);
+        let cert_reload_updated_counter =
+            CERT_RELOAD_UPDATED_COUNTER.with_label_values(&[resolver_name]);
+        let cert_reload_failed_counter =
+            CERT_RELOAD_FAILED_COUNTER.with_label_values(&[resolver_name]);
+
+        let parsed_key = load_and_parse_certified_key(key_filename, cert_filename).await?;
+
        let this = Arc::new(Self {
-            certified_key: ArcSwap::from_pointee(
-                load_certified_key(key_filename, cert_filename).await?,
-            ),
+            certified_key: ArcSwap::from_pointee(parsed_key.certified_key),
        });
+        cert_expiration_time.set(parsed_key.expiration_time.as_secs());

        tokio::spawn({
            let weak_this = Arc::downgrade(&this);
@@ -88,17 +173,22 @@ impl ReloadingCertificateResolver {
                        Some(this) => this,
                        None => break, // Resolver has been destroyed, exit.
                    };
-                    match load_certified_key(&key_filename, &cert_filename).await {
-                        Ok(new_certified_key) => {
-                            if new_certified_key.cert == this.certified_key.load().cert {
+                    cert_reload_started_counter.inc();
+
+                    match load_and_parse_certified_key(&key_filename, &cert_filename).await {
+                        Ok(parsed_key) => {
+                            if parsed_key.certified_key.cert == this.certified_key.load().cert {
                                tracing::debug!("Certificate has not changed since last reloading");
                            } else {
                                tracing::info!("Certificate has been reloaded");
-                                this.certified_key.store(Arc::new(new_certified_key));
+                                this.certified_key.store(Arc::new(parsed_key.certified_key));
+                                cert_expiration_time.set(parsed_key.expiration_time.as_secs());
+                                cert_reload_updated_counter.inc();
                            }
                            last_reload_failed = false;
                        }
                        Err(err) => {
+                            cert_reload_failed_counter.inc();
                            // Note: Reloading certs may fail if it conflicts with the script updating
                            // the files at the same time. Warn only if the error is persistent.
                            if last_reload_failed {
--- a/libs/pageserver_api/src/config.rs
+++ b/libs/pageserver_api/src/config.rs
@@ -51,9 +51,54 @@ pub struct NodeMetadata {
 /// If there cannot be a static default value because we need to make runtime
 /// checks to determine the default, make it an `Option` (which defaults to None).
 /// The runtime check should be done in the consuming crate, i.e., `pageserver`.
+///
+/// Unknown fields are silently ignored during deserialization.
+/// The alternative, which we used in the past, was to set `deny_unknown_fields`,
+/// which fails deserialization, and hence pageserver startup, if there is an unknown field.
+/// The reason we don't do that anymore is that it complicates
+/// usage of config fields for feature flagging, which we commonly do for
+/// region-by-region rollouts.
+/// The complications mainly arise because the `pageserver.toml` contents on a
+/// prod server have a separate lifecycle from the pageserver binary.
+/// For instance, `pageserver.toml` contents today are defined in the internal
+/// infra repo, and thus introducing a new config field to pageserver and
+/// rolling it out to prod servers are separate commits in separate repos
+/// that can't be made or rolled back atomically.
+/// Rollbacks in particular pose a risk with deny_unknown_fields because
+/// the old pageserver binary may reject a new config field, resulting in
+/// an outage unless the person doing the pageserver rollback remembers
+/// to also revert the commit that added the config field in to the
+/// `pageserver.toml` templates in the internal infra repo.
+/// (A pre-deploy config check would eliminate this risk during rollbacks,
+///  cf [here](https://github.com/neondatabase/cloud/issues/24349).)
+/// In addition to this compatibility problem during emergency rollbacks,
+/// deny_unknown_fields adds further complications when decomissioning a feature
+/// flag: with deny_unknown_fields, we can't remove a flag from the [`ConfigToml`]
+/// until all prod servers' `pageserver.toml` files have been updated to a version
+/// that doesn't specify the flag. Otherwise new software would fail to start up.
+/// This adds the requirement for an intermediate step where the new config field
+/// is accepted but ignored, prolonging the decomissioning process by an entire
+/// release cycle.
+/// By contrast  with unknown fields silently ignored, decomissioning a feature
+/// flag is a one-step process: we can skip the intermediate step and straight
+/// remove the field from the [`ConfigToml`]. We leave the field in the
+/// `pageserver.toml` files on prod servers until we reach certainty that we
+/// will not roll back to old software whose behavior was dependent on config.
+/// Then we can remove the field from the templates in the internal infra repo.
+/// This process is [documented internally](
+/// https://docs.neon.build/storage/pageserver_configuration.html).
+///
+/// Note that above relaxed compatbility for the config format does NOT APPLY
+/// TO THE STORAGE FORMAT. As general guidance, when introducing storage format
+/// changes, ensure that the potential rollback target version will be compatible
+/// with the new format. This must hold regardless of what flags are set in in the `pageserver.toml`:
+/// any format version that exists in an environment must be compatible with the software that runs there.
+/// Use a pageserver.toml flag only to gate whether software _writes_ the new format.
+/// For more compatibility considerations, refer to [internal docs](
+/// https://docs.neon.build/storage/compat.html?highlight=compat#format-versions--compatibility)
 #[serde_as]
 #[derive(Clone, Debug, serde::Deserialize, serde::Serialize)]
-#[serde(default, deny_unknown_fields)]
+#[serde(default)]
 pub struct ConfigToml {
    // types mapped 1:1 into the runtime PageServerConfig type
    pub listen_pg_addr: String,
@@ -135,10 +180,10 @@ pub struct ConfigToml {
    #[serde(skip_serializing_if = "Option::is_none")]
    pub generate_unarchival_heatmap: Option<bool>,
    pub tracing: Option<Tracing>,
+    pub enable_tls_page_service_api: bool,
 }

 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
-#[serde(deny_unknown_fields)]
 pub struct DiskUsageEvictionTaskConfig {
    pub max_usage_pct: utils::serde_percent::Percent,
    pub min_avail_bytes: u64,
@@ -153,17 +198,19 @@ pub struct DiskUsageEvictionTaskConfig {

 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
 #[serde(tag = "mode", rename_all = "kebab-case")]
-#[serde(deny_unknown_fields)]
 pub enum PageServicePipeliningConfig {
    Serial,
    Pipelined(PageServicePipeliningConfigPipelined),
 }
 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
-#[serde(deny_unknown_fields)]
 pub struct PageServicePipeliningConfigPipelined {
    /// Causes runtime errors if larger than max get_vectored batch size.
    pub max_batch_size: NonZeroUsize,
    pub execution: PageServiceProtocolPipelinedExecutionStrategy,
+    // The default below is such that new versions of the software can start
+    // with the old configuration.
+    #[serde(default)]
+    pub batching: PageServiceProtocolPipelinedBatchingStrategy,
 }

 #[derive(Debug, Clone, Copy, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -173,9 +220,21 @@ pub enum PageServiceProtocolPipelinedExecutionStrategy {
    Tasks,
 }

+#[derive(Default, Debug, Clone, Copy, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
+#[serde(rename_all = "kebab-case")]
+pub enum PageServiceProtocolPipelinedBatchingStrategy {
+    /// All get page requests in a batch will be at the same LSN
+    #[default]
+    UniformLsn,
+    /// Get page requests in a batch may be at different LSN
+    ///
+    /// One key cannot be present more than once at different LSNs in
+    /// the same batch.
+    ScatteredLsn,
+}
+
 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
 #[serde(tag = "mode", rename_all = "kebab-case")]
-#[serde(deny_unknown_fields)]
 pub enum GetVectoredConcurrentIo {
    /// The read path is fully sequential: layers are visited
    /// one after the other and IOs are issued and waited upon
@@ -192,7 +251,7 @@ pub enum GetVectoredConcurrentIo {
    SidecarTask,
 }

-#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
+#[derive(Debug, Copy, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
 pub struct Ratio {
    pub numerator: usize,
    pub denominator: usize,
@@ -294,7 +353,7 @@ pub struct MaxVectoredReadBytes(pub NonZeroUsize);

 /// Tenant-level configuration values, used for various purposes.
 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
-#[serde(deny_unknown_fields, default)]
+#[serde(default)]
 pub struct TenantConfigToml {
    // Flush out an inmemory layer, if it's holding WAL older than this
    // This puts a backstop on how much WAL needs to be re-digested if the
@@ -410,12 +469,17 @@ pub struct TenantConfigToml {
    // gc-compaction related configs
    /// Enable automatic gc-compaction trigger on this tenant.
    pub gc_compaction_enabled: bool,
+    /// Enable verification of gc-compaction results.
+    pub gc_compaction_verification: bool,
    /// The initial threshold for gc-compaction in KB. Once the total size of layers below the gc-horizon is above this threshold,
    /// gc-compaction will be triggered.
    pub gc_compaction_initial_threshold_kb: u64,
    /// The ratio that triggers the auto gc-compaction. If (the total size of layers between L2 LSN and gc-horizon) / (size below the L2 LSN)
    /// is above this ratio, gc-compaction will be triggered.
    pub gc_compaction_ratio_percent: u64,
+    /// Tenant level performance sampling ratio override. Controls the ratio of get page requests
+    /// that will get perf sampling for the tenant.
+    pub sampling_ratio: Option<Ratio>,
 }

 pub mod defaults {
@@ -568,9 +632,12 @@ impl Default for ConfigToml {
            page_service_pipelining: if !cfg!(test) {
                PageServicePipeliningConfig::Serial
            } else {
+                // Do not turn this into the default until scattered reads have been
+                // validated and rolled-out fully.
                PageServicePipeliningConfig::Pipelined(PageServicePipeliningConfigPipelined {
                    max_batch_size: NonZeroUsize::new(32).unwrap(),
                    execution: PageServiceProtocolPipelinedExecutionStrategy::ConcurrentFutures,
+                    batching: PageServiceProtocolPipelinedBatchingStrategy::ScatteredLsn,
                })
            },
            get_vectored_concurrent_io: if !cfg!(test) {
@@ -587,6 +654,7 @@ impl Default for ConfigToml {
            load_previous_heatmap: None,
            generate_unarchival_heatmap: None,
            tracing: None,
+            enable_tls_page_service_api: false,
        }
    }
 }
@@ -646,6 +714,7 @@ pub mod tenant_conf_defaults {
    // image layers should be created.
    pub const DEFAULT_IMAGE_LAYER_CREATION_CHECK_THRESHOLD: u8 = 2;
    pub const DEFAULT_GC_COMPACTION_ENABLED: bool = false;
+    pub const DEFAULT_GC_COMPACTION_VERIFICATION: bool = true;
    pub const DEFAULT_GC_COMPACTION_INITIAL_THRESHOLD_KB: u64 = 5 * 1024 * 1024; // 5GB
    pub const DEFAULT_GC_COMPACTION_RATIO_PERCENT: u64 = 100;
 }
@@ -700,8 +769,10 @@ impl Default for TenantConfigToml {
            wal_receiver_protocol_override: None,
            rel_size_v2_enabled: false,
            gc_compaction_enabled: DEFAULT_GC_COMPACTION_ENABLED,
+            gc_compaction_verification: DEFAULT_GC_COMPACTION_VERIFICATION,
            gc_compaction_initial_threshold_kb: DEFAULT_GC_COMPACTION_INITIAL_THRESHOLD_KB,
            gc_compaction_ratio_percent: DEFAULT_GC_COMPACTION_RATIO_PERCENT,
+            sampling_ratio: None,
        }
    }
 }
--- a/libs/pageserver_api/src/controller_api.rs
+++ b/libs/pageserver_api/src/controller_api.rs
@@ -7,7 +7,8 @@ use std::time::{Duration, Instant};
 /// API (`/control/v1` prefix).  Implemented by the server
 /// in [`storage_controller::http`]
 use serde::{Deserialize, Serialize};
-use utils::id::{NodeId, TenantId};
+use utils::id::{NodeId, TenantId, TimelineId};
+use utils::lsn::Lsn;

 use crate::models::{PageserverUtilization, ShardParameters, TenantConfig};
 use crate::shard::{ShardStripeSize, TenantShardId};
@@ -499,6 +500,15 @@ pub struct SafekeeperSchedulingPolicyRequest {
    pub scheduling_policy: SkSchedulingPolicy,
 }

+/// Import request for safekeeper timelines.
+#[derive(Serialize, Deserialize, Clone)]
+pub struct TimelineImportRequest {
+    pub tenant_id: TenantId,
+    pub timeline_id: TimelineId,
+    pub start_lsn: Lsn,
+    pub sk_set: Vec<NodeId>,
+}
+
 #[cfg(test)]
 mod test {
    use serde_json;
--- a/libs/pageserver_api/src/key.rs
+++ b/libs/pageserver_api/src/key.rs
@@ -927,7 +927,7 @@ impl Key {

    /// Guaranteed to return `Ok()` if [`Self::is_rel_block_key`] returns `true` for `key`.
    #[inline(always)]
-    pub fn to_rel_block(self) -> anyhow::Result<(RelTag, BlockNumber)> {
+    pub fn to_rel_block(self) -> Result<(RelTag, BlockNumber), ToRelBlockError> {
        Ok(match self.field1 {
            0x00 => (
                RelTag {
@@ -938,7 +938,7 @@ impl Key {
                },
                self.field6,
            ),
-            _ => anyhow::bail!("unexpected value kind 0x{:02x}", self.field1),
+            _ => return Err(ToRelBlockError(self.field1)),
        })
    }
 }
@@ -951,6 +951,17 @@ impl std::str::FromStr for Key {
    }
 }

+#[derive(Debug)]
+pub struct ToRelBlockError(u8);
+
+impl fmt::Display for ToRelBlockError {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        write!(f, "unexpected value kind 0x{:02x}", self.0)
+    }
+}
+
+impl std::error::Error for ToRelBlockError {}
+
 #[cfg(test)]
 mod tests {
    use std::str::FromStr;
--- a/libs/pageserver_api/src/keyspace.rs
+++ b/libs/pageserver_api/src/keyspace.rs
@@ -613,8 +613,7 @@ mod tests {
    use rand::{RngCore, SeedableRng};

    use super::*;
-    use crate::models::ShardParameters;
-    use crate::shard::{ShardCount, ShardNumber};
+    use crate::shard::{DEFAULT_STRIPE_SIZE, ShardCount, ShardNumber, ShardStripeSize};

    // Helper function to create a key range.
    //
@@ -964,12 +963,8 @@ mod tests {
    }
    #[test]
    fn sharded_range_relation_gap() {
-        let shard_identity = ShardIdentity::new(
-            ShardNumber(0),
-            ShardCount::new(4),
-            ShardParameters::DEFAULT_STRIPE_SIZE,
-        )
-        .unwrap();
+        let shard_identity =
+            ShardIdentity::new(ShardNumber(0), ShardCount::new(4), DEFAULT_STRIPE_SIZE).unwrap();

        let range = ShardedRange::new(
            Range {
@@ -985,12 +980,8 @@ mod tests {

    #[test]
    fn shard_identity_keyspaces_single_key() {
-        let shard_identity = ShardIdentity::new(
-            ShardNumber(1),
-            ShardCount::new(4),
-            ShardParameters::DEFAULT_STRIPE_SIZE,
-        )
-        .unwrap();
+        let shard_identity =
+            ShardIdentity::new(ShardNumber(1), ShardCount::new(4), DEFAULT_STRIPE_SIZE).unwrap();

        let range = ShardedRange::new(
            Range {
@@ -1034,12 +1025,8 @@ mod tests {

    #[test]
    fn shard_identity_keyspaces_forkno_gap() {
-        let shard_identity = ShardIdentity::new(
-            ShardNumber(1),
-            ShardCount::new(4),
-            ShardParameters::DEFAULT_STRIPE_SIZE,
-        )
-        .unwrap();
+        let shard_identity =
+            ShardIdentity::new(ShardNumber(1), ShardCount::new(4), DEFAULT_STRIPE_SIZE).unwrap();

        let range = ShardedRange::new(
            Range {
@@ -1061,7 +1048,7 @@ mod tests {
            let shard_identity = ShardIdentity::new(
                ShardNumber(shard_number),
                ShardCount::new(4),
-                ShardParameters::DEFAULT_STRIPE_SIZE,
+                DEFAULT_STRIPE_SIZE,
            )
            .unwrap();

@@ -1144,37 +1131,44 @@ mod tests {
    /// for a single tenant.
    #[test]
    fn sharded_range_fragment_simple() {
+        const SHARD_COUNT: u8 = 4;
+        const STRIPE_SIZE: u32 = DEFAULT_STRIPE_SIZE.0;
+
        let shard_identity = ShardIdentity::new(
            ShardNumber(0),
-            ShardCount::new(4),
-            ShardParameters::DEFAULT_STRIPE_SIZE,
+            ShardCount::new(SHARD_COUNT),
+            ShardStripeSize(STRIPE_SIZE),
        )
        .unwrap();

        // A range which we happen to know covers exactly one stripe which belongs to this shard
        let input_start = Key::from_hex("000000067f00000001000000ae0000000000").unwrap();
-        let input_end = Key::from_hex("000000067f00000001000000ae0000008000").unwrap();
+        let mut input_end = input_start;
+        input_end.field6 += STRIPE_SIZE; // field6 is block number

        // Ask for stripe_size blocks, we get the whole stripe
        assert_eq!(
-            do_fragment(input_start, input_end, &shard_identity, 32768),
-            (32768, vec![(32768, input_start..input_end)])
+            do_fragment(input_start, input_end, &shard_identity, STRIPE_SIZE),
+            (STRIPE_SIZE, vec![(STRIPE_SIZE, input_start..input_end)])
        );

        // Ask for more, we still get the whole stripe
        assert_eq!(
-            do_fragment(input_start, input_end, &shard_identity, 10000000),
-            (32768, vec![(32768, input_start..input_end)])
+            do_fragment(input_start, input_end, &shard_identity, 10 * STRIPE_SIZE),
+            (STRIPE_SIZE, vec![(STRIPE_SIZE, input_start..input_end)])
        );

        // Ask for target_nblocks of half the stripe size, we get two halves
        assert_eq!(
-            do_fragment(input_start, input_end, &shard_identity, 16384),
+            do_fragment(input_start, input_end, &shard_identity, STRIPE_SIZE / 2),
            (
-                32768,
+                STRIPE_SIZE,
                vec![
-                    (16384, input_start..input_start.add(16384)),
-                    (16384, input_start.add(16384)..input_end)
+                    (
+                        STRIPE_SIZE / 2,
+                        input_start..input_start.add(STRIPE_SIZE / 2)
+                    ),
+                    (STRIPE_SIZE / 2, input_start.add(STRIPE_SIZE / 2)..input_end)
                ]
            )
        );
@@ -1182,40 +1176,53 @@ mod tests {

    #[test]
    fn sharded_range_fragment_multi_stripe() {
+        const SHARD_COUNT: u8 = 4;
+        const STRIPE_SIZE: u32 = DEFAULT_STRIPE_SIZE.0;
+        const RANGE_SIZE: u32 = SHARD_COUNT as u32 * STRIPE_SIZE;
+
        let shard_identity = ShardIdentity::new(
            ShardNumber(0),
-            ShardCount::new(4),
-            ShardParameters::DEFAULT_STRIPE_SIZE,
+            ShardCount::new(SHARD_COUNT),
+            ShardStripeSize(STRIPE_SIZE),
        )
        .unwrap();

        // A range which covers multiple stripes, exactly one of which belongs to the current shard.
        let input_start = Key::from_hex("000000067f00000001000000ae0000000000").unwrap();
-        let input_end = Key::from_hex("000000067f00000001000000ae0000020000").unwrap();
+        let mut input_end = input_start;
+        input_end.field6 += RANGE_SIZE; // field6 is block number
+
        // Ask for all the blocks, get a fragment that covers the whole range but reports
        // its size to be just the blocks belonging to our shard.
        assert_eq!(
-            do_fragment(input_start, input_end, &shard_identity, 131072),
-            (32768, vec![(32768, input_start..input_end)])
+            do_fragment(input_start, input_end, &shard_identity, RANGE_SIZE),
+            (STRIPE_SIZE, vec![(STRIPE_SIZE, input_start..input_end)])
        );

-        // Ask for a sub-stripe quantity
+        // Ask for a sub-stripe quantity that results in 3 fragments.
+        let limit = STRIPE_SIZE / 3 + 1;
        assert_eq!(
-            do_fragment(input_start, input_end, &shard_identity, 16000),
+            do_fragment(input_start, input_end, &shard_identity, limit),
            (
-                32768,
+                STRIPE_SIZE,
                vec![
-                    (16000, input_start..input_start.add(16000)),
-                    (16000, input_start.add(16000)..input_start.add(32000)),
-                    (768, input_start.add(32000)..input_end),
+                    (limit, input_start..input_start.add(limit)),
+                    (limit, input_start.add(limit)..input_start.add(2 * limit)),
+                    (
+                        STRIPE_SIZE - 2 * limit,
+                        input_start.add(2 * limit)..input_end
+                    ),
                ]
            )
        );

        // Try on a range that starts slightly after our owned stripe
        assert_eq!(
-            do_fragment(input_start.add(1), input_end, &shard_identity, 131072),
-            (32767, vec![(32767, input_start.add(1)..input_end)])
+            do_fragment(input_start.add(1), input_end, &shard_identity, RANGE_SIZE),
+            (
+                STRIPE_SIZE - 1,
+                vec![(STRIPE_SIZE - 1, input_start.add(1)..input_end)]
+            )
        );
    }

@@ -1223,32 +1230,40 @@ mod tests {
    /// a previous relation.
    #[test]
    fn sharded_range_fragment_starting_from_logical_size() {
+        const SHARD_COUNT: u8 = 4;
+        const STRIPE_SIZE: u32 = DEFAULT_STRIPE_SIZE.0;
+        const RANGE_SIZE: u32 = SHARD_COUNT as u32 * STRIPE_SIZE;
+
        let input_start = Key::from_hex("000000067f00000001000000ae00ffffffff").unwrap();
-        let input_end = Key::from_hex("000000067f00000001000000ae0100008000").unwrap();
+        let mut input_end = Key::from_hex("000000067f00000001000000ae0100000000").unwrap();
+        input_end.field6 += RANGE_SIZE; // field6 is block number

        // Shard 0 owns the first stripe in the relation, and the preceding logical size is shard local too
        let shard_identity = ShardIdentity::new(
            ShardNumber(0),
-            ShardCount::new(4),
-            ShardParameters::DEFAULT_STRIPE_SIZE,
+            ShardCount::new(SHARD_COUNT),
+            ShardStripeSize(STRIPE_SIZE),
        )
        .unwrap();
        assert_eq!(
-            do_fragment(input_start, input_end, &shard_identity, 0x10000),
-            (0x8001, vec![(0x8001, input_start..input_end)])
+            do_fragment(input_start, input_end, &shard_identity, 2 * STRIPE_SIZE),
+            (
+                STRIPE_SIZE + 1,
+                vec![(STRIPE_SIZE + 1, input_start..input_end)]
+            )
        );

        // Shard 1 does not own the first stripe in the relation, but it does own the logical size (all shards
        // store all logical sizes)
        let shard_identity = ShardIdentity::new(
            ShardNumber(1),
-            ShardCount::new(4),
-            ShardParameters::DEFAULT_STRIPE_SIZE,
+            ShardCount::new(SHARD_COUNT),
+            ShardStripeSize(STRIPE_SIZE),
        )
        .unwrap();
        assert_eq!(
-            do_fragment(input_start, input_end, &shard_identity, 0x10000),
-            (0x1, vec![(0x1, input_start..input_end)])
+            do_fragment(input_start, input_end, &shard_identity, 2 * STRIPE_SIZE),
+            (1, vec![(1, input_start..input_end)])
        );
    }

@@ -1284,12 +1299,8 @@ mod tests {
        );

        // Same, but using a sharded identity
-        let shard_identity = ShardIdentity::new(
-            ShardNumber(0),
-            ShardCount::new(4),
-            ShardParameters::DEFAULT_STRIPE_SIZE,
-        )
-        .unwrap();
+        let shard_identity =
+            ShardIdentity::new(ShardNumber(0), ShardCount::new(4), DEFAULT_STRIPE_SIZE).unwrap();
        assert_eq!(
            do_fragment(input_start, input_end, &shard_identity, 0x8000),
            (u32::MAX, vec![(u32::MAX, input_start..input_end),])
@@ -1331,7 +1342,7 @@ mod tests {
                ShardIdentity::new(
                    ShardNumber((prng.next_u32() % shard_count) as u8),
                    ShardCount::new(shard_count as u8),
-                    ShardParameters::DEFAULT_STRIPE_SIZE,
+                    DEFAULT_STRIPE_SIZE,
                )
                .unwrap()
            };
--- a/libs/pageserver_api/src/models.rs
+++ b/libs/pageserver_api/src/models.rs
@@ -23,9 +23,10 @@ use utils::lsn::Lsn;
 use utils::postgres_client::PostgresClientProtocol;
 use utils::{completion, serde_system_time};

+use crate::config::Ratio;
 use crate::key::{CompactKey, Key};
 use crate::reltag::RelTag;
-use crate::shard::{ShardCount, ShardStripeSize, TenantShardId};
+use crate::shard::{DEFAULT_STRIPE_SIZE, ShardCount, ShardStripeSize, TenantShardId};

 /// The state of a tenant in this pageserver.
 ///
@@ -79,10 +80,22 @@ pub enum TenantState {
    ///
    /// Transitions out of this state are possible through `set_broken()`.
    Stopping {
+        /// The barrier can be used to wait for shutdown to complete. The first caller to set
+        /// Some(Barrier) is responsible for driving shutdown to completion. Subsequent callers
+        /// will wait for the first caller's existing barrier.
+        ///
+        /// None is set when an attach is cancelled, to signal to shutdown that the attach has in
+        /// fact cancelled:
+        ///
+        /// 1. `shutdown` sees `TenantState::Attaching`, and cancels the tenant.
+        /// 2. `attach` sets `TenantState::Stopping(None)` and exits.
+        /// 3. `set_stopping` waits for `TenantState::Stopping(None)` and sets
+        ///    `TenantState::Stopping(Some)` to claim the barrier as the shutdown owner.
+        //
        // Because of https://github.com/serde-rs/serde/issues/2105 this has to be a named field,
        // otherwise it will not be skipped during deserialization
        #[serde(skip)]
-        progress: completion::Barrier,
+        progress: Option<completion::Barrier>,
    },
    /// The tenant is recognized by the pageserver, but can no longer be used for
    /// any operations.
@@ -425,8 +438,6 @@ pub struct ShardParameters {
 }

 impl ShardParameters {
-    pub const DEFAULT_STRIPE_SIZE: ShardStripeSize = ShardStripeSize(256 * 1024 / 8);
-
    pub fn is_unsharded(&self) -> bool {
        self.count.is_unsharded()
    }
@@ -436,7 +447,7 @@ impl Default for ShardParameters {
    fn default() -> Self {
        Self {
            count: ShardCount::new(0),
-            stripe_size: Self::DEFAULT_STRIPE_SIZE,
+            stripe_size: DEFAULT_STRIPE_SIZE,
        }
    }
 }
@@ -565,9 +576,13 @@ pub struct TenantConfigPatch {
    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
    pub gc_compaction_enabled: FieldPatch<bool>,
    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_verification: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
    pub gc_compaction_initial_threshold_kb: FieldPatch<u64>,
    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
    pub gc_compaction_ratio_percent: FieldPatch<u64>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub sampling_ratio: FieldPatch<Option<Ratio>>,
 }

 /// Like [`crate::config::TenantConfigToml`], but preserves the information
@@ -683,11 +698,17 @@ pub struct TenantConfig {
    #[serde(skip_serializing_if = "Option::is_none")]
    pub gc_compaction_enabled: Option<bool>,

+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub gc_compaction_verification: Option<bool>,
+
    #[serde(skip_serializing_if = "Option::is_none")]
    pub gc_compaction_initial_threshold_kb: Option<u64>,

    #[serde(skip_serializing_if = "Option::is_none")]
    pub gc_compaction_ratio_percent: Option<u64>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub sampling_ratio: Option<Option<Ratio>>,
 }

 impl TenantConfig {
@@ -728,8 +749,10 @@ impl TenantConfig {
            mut wal_receiver_protocol_override,
            mut rel_size_v2_enabled,
            mut gc_compaction_enabled,
+            mut gc_compaction_verification,
            mut gc_compaction_initial_threshold_kb,
            mut gc_compaction_ratio_percent,
+            mut sampling_ratio,
        } = self;

        patch.checkpoint_distance.apply(&mut checkpoint_distance);
@@ -818,12 +841,16 @@ impl TenantConfig {
        patch
            .gc_compaction_enabled
            .apply(&mut gc_compaction_enabled);
+        patch
+            .gc_compaction_verification
+            .apply(&mut gc_compaction_verification);
        patch
            .gc_compaction_initial_threshold_kb
            .apply(&mut gc_compaction_initial_threshold_kb);
        patch
            .gc_compaction_ratio_percent
            .apply(&mut gc_compaction_ratio_percent);
+        patch.sampling_ratio.apply(&mut sampling_ratio);

        Ok(Self {
            checkpoint_distance,
@@ -858,8 +885,10 @@ impl TenantConfig {
            wal_receiver_protocol_override,
            rel_size_v2_enabled,
            gc_compaction_enabled,
+            gc_compaction_verification,
            gc_compaction_initial_threshold_kb,
            gc_compaction_ratio_percent,
+            sampling_ratio,
        })
    }

@@ -955,12 +984,16 @@ impl TenantConfig {
            gc_compaction_enabled: self
                .gc_compaction_enabled
                .unwrap_or(global_conf.gc_compaction_enabled),
+            gc_compaction_verification: self
+                .gc_compaction_verification
+                .unwrap_or(global_conf.gc_compaction_verification),
            gc_compaction_initial_threshold_kb: self
                .gc_compaction_initial_threshold_kb
                .unwrap_or(global_conf.gc_compaction_initial_threshold_kb),
            gc_compaction_ratio_percent: self
                .gc_compaction_ratio_percent
                .unwrap_or(global_conf.gc_compaction_ratio_percent),
+            sampling_ratio: self.sampling_ratio.unwrap_or(global_conf.sampling_ratio),
        }
    }
 }
@@ -1094,7 +1127,7 @@ pub struct CompactionAlgorithmSettings {
 }

 #[derive(Debug, PartialEq, Eq, Clone, Deserialize, Serialize)]
-#[serde(tag = "mode", rename_all = "kebab-case", deny_unknown_fields)]
+#[serde(tag = "mode", rename_all = "kebab-case")]
 pub enum L0FlushConfig {
    #[serde(rename_all = "snake_case")]
    Direct { max_concurrency: NonZeroUsize },
@@ -1658,6 +1691,7 @@ pub struct SecondaryProgress {
 pub struct TenantScanRemoteStorageShard {
    pub tenant_shard_id: TenantShardId,
    pub generation: Option<u32>,
+    pub stripe_size: Option<ShardStripeSize>,
 }

 #[derive(Serialize, Deserialize, Debug, Default)]
@@ -2709,10 +2743,15 @@ mod tests {
                "Activating",
            ),
            (line!(), TenantState::Active, "Active"),
+            (
+                line!(),
+                TenantState::Stopping { progress: None },
+                "Stopping",
+            ),
            (
                line!(),
                TenantState::Stopping {
-                    progress: utils::completion::Barrier::default(),
+                    progress: Some(completion::Barrier::default()),
                },
                "Stopping",
            ),
--- a/libs/pageserver_api/src/record.rs
+++ b/libs/pageserver_api/src/record.rs
@@ -58,6 +58,8 @@ pub enum NeonWalRecord {
        /// to true. This record does not need the history WALs to reconstruct. See [`NeonWalRecord::will_init`] and
        /// its references in `timeline.rs`.
        will_init: bool,
+        /// Only append the record if the current image is the same as the one specified in this field.
+        only_if: Option<String>,
    },
 }

@@ -81,6 +83,17 @@ impl NeonWalRecord {
            append: s.as_ref().to_string(),
            clear: false,
            will_init: false,
+            only_if: None,
+        }
+    }
+
+    #[cfg(feature = "testing")]
+    pub fn wal_append_conditional(s: impl AsRef<str>, only_if: impl AsRef<str>) -> Self {
+        Self::Test {
+            append: s.as_ref().to_string(),
+            clear: false,
+            will_init: false,
+            only_if: Some(only_if.as_ref().to_string()),
        }
    }

@@ -90,6 +103,7 @@ impl NeonWalRecord {
            append: s.as_ref().to_string(),
            clear: true,
            will_init: false,
+            only_if: None,
        }
    }

@@ -99,6 +113,7 @@ impl NeonWalRecord {
            append: s.as_ref().to_string(),
            clear: true,
            will_init: true,
+            only_if: None,
        }
    }
 }
--- a/libs/pageserver_api/src/shard.rs
+++ b/libs/pageserver_api/src/shard.rs
@@ -78,6 +78,12 @@ impl Default for ShardStripeSize {
    }
 }

+impl std::fmt::Display for ShardStripeSize {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        self.0.fmt(f)
+    }
+}
+
 /// Layout version: for future upgrades where we might change how the key->shard mapping works
 #[derive(Clone, Copy, Serialize, Deserialize, Eq, PartialEq, Hash, Debug)]
 pub struct ShardLayout(u8);
@@ -86,8 +92,11 @@ const LAYOUT_V1: ShardLayout = ShardLayout(1);
 /// ShardIdentity uses a magic layout value to indicate if it is unusable
 const LAYOUT_BROKEN: ShardLayout = ShardLayout(255);

-/// Default stripe size in pages: 256MiB divided by 8kiB page size.
-const DEFAULT_STRIPE_SIZE: ShardStripeSize = ShardStripeSize(256 * 1024 / 8);
+/// The default stripe size in pages. 16 MiB divided by 8 kiB page size.
+///
+/// A lower stripe size distributes ingest load better across shards, but reduces IO amortization.
+/// 16 MiB appears to be a reasonable balance: <https://github.com/neondatabase/neon/pull/10510>.
+pub const DEFAULT_STRIPE_SIZE: ShardStripeSize = ShardStripeSize(16 * 1024 / 8);

 #[derive(thiserror::Error, Debug, PartialEq, Eq)]
 pub enum ShardConfigError {
@@ -537,7 +546,7 @@ mod tests {
            field6: 0x7d06,
        };

-        let shard = key_to_shard_number(ShardCount(10), DEFAULT_STRIPE_SIZE, &key);
+        let shard = key_to_shard_number(ShardCount(10), ShardStripeSize(32768), &key);
        assert_eq!(shard, ShardNumber(8));
    }

--- a/libs/postgres_backend/src/lib.rs
+++ b/libs/postgres_backend/src/lib.rs
@@ -5,7 +5,6 @@
 #![deny(unsafe_code)]
 #![deny(clippy::undocumented_unsafe_blocks)]
 use std::future::Future;
-use std::io::ErrorKind;
 use std::net::SocketAddr;
 use std::os::fd::{AsRawFd, RawFd};
 use std::pin::Pin;
@@ -227,7 +226,7 @@ impl<IO: AsyncRead + AsyncWrite + Unpin> MaybeWriteOnly<IO> {
        match self {
            MaybeWriteOnly::Full(framed) => framed.read_startup_message().await,
            MaybeWriteOnly::WriteOnly(_) => {
-                Err(io::Error::new(ErrorKind::Other, "reading from write only half").into())
+                Err(io::Error::other("reading from write only half").into())
            }
            MaybeWriteOnly::Broken => panic!("IO on invalid MaybeWriteOnly"),
        }
@@ -237,7 +236,7 @@ impl<IO: AsyncRead + AsyncWrite + Unpin> MaybeWriteOnly<IO> {
        match self {
            MaybeWriteOnly::Full(framed) => framed.read_message().await,
            MaybeWriteOnly::WriteOnly(_) => {
-                Err(io::Error::new(ErrorKind::Other, "reading from write only half").into())
+                Err(io::Error::other("reading from write only half").into())
            }
            MaybeWriteOnly::Broken => panic!("IO on invalid MaybeWriteOnly"),
        }
@@ -975,7 +974,7 @@ impl<IO: AsyncRead + AsyncWrite + Unpin> AsyncWrite for CopyDataWriter<'_, IO> {
            .write_message_noflush(&BeMessage::CopyData(buf))
            // write_message only writes to the buffer, so it can fail iff the
            // message is invaid, but CopyData can't be invalid.
-            .map_err(|_| io::Error::new(ErrorKind::Other, "failed to serialize CopyData"))?;
+            .map_err(|_| io::Error::other("failed to serialize CopyData"))?;

        Poll::Ready(Ok(buf.len()))
    }
--- a/libs/postgres_backend/tests/simple_select.rs
+++ b/libs/postgres_backend/tests/simple_select.rs
@@ -85,8 +85,8 @@ static KEY: Lazy<rustls::pki_types::PrivateKeyDer<'static>> = Lazy::new(|| {

 static CERT: Lazy<rustls::pki_types::CertificateDer<'static>> = Lazy::new(|| {
    let mut cursor = Cursor::new(include_bytes!("cert.pem"));
-    let cert = rustls_pemfile::certs(&mut cursor).next().unwrap().unwrap();
-    cert
+
+    rustls_pemfile::certs(&mut cursor).next().unwrap().unwrap()
 });

 // test that basic select with ssl works
--- a/libs/pq_proto/src/framed.rs
+++ b/libs/pq_proto/src/framed.rs
@@ -35,7 +35,7 @@ impl ConnectionError {
    pub fn into_io_error(self) -> io::Error {
        match self {
            ConnectionError::Io(io) => io,
-            ConnectionError::Protocol(pe) => io::Error::new(io::ErrorKind::Other, pe.to_string()),
+            ConnectionError::Protocol(pe) => io::Error::other(pe.to_string()),
        }
    }
 }
--- a/libs/pq_proto/src/lib.rs
+++ b/libs/pq_proto/src/lib.rs
@@ -257,7 +257,7 @@ pub enum ProtocolError {
 impl ProtocolError {
    /// Proxy stream.rs uses only io::Error; provide it.
    pub fn into_io_error(self) -> io::Error {
-        io::Error::new(io::ErrorKind::Other, self.to_string())
+        io::Error::other(self.to_string())
    }
 }

--- a/libs/proxy/postgres-protocol2/src/authentication/sasl.rs
+++ b/libs/proxy/postgres-protocol2/src/authentication/sasl.rs
@@ -212,7 +212,7 @@ impl ScramSha256 {
                    password,
                    channel_binding,
                } => (nonce, password, channel_binding),
-                _ => return Err(io::Error::new(io::ErrorKind::Other, "invalid SCRAM state")),
+                _ => return Err(io::Error::other("invalid SCRAM state")),
            };

        let message =
@@ -291,7 +291,7 @@ impl ScramSha256 {
                server_key,
                auth_message,
            } => (server_key, auth_message),
-            _ => return Err(io::Error::new(io::ErrorKind::Other, "invalid SCRAM state")),
+            _ => return Err(io::Error::other("invalid SCRAM state")),
        };

        let message =
@@ -301,10 +301,7 @@ impl ScramSha256 {

        let verifier = match parsed {
            ServerFinalMessage::Error(e) => {
-                return Err(io::Error::new(
-                    io::ErrorKind::Other,
-                    format!("SCRAM error: {}", e),
-                ));
+                return Err(io::Error::other(format!("SCRAM error: {}", e)));
            }
            ServerFinalMessage::Verifier(verifier) => verifier,
        };
--- a/libs/remote_keys/Cargo.toml
+++ b/libs/remote_keys/Cargo.toml
@@ -0,0 +1,13 @@
+[package]
+name = "remote_keys"
+version = "0.1.0"
+edition = "2024"
+license.workspace = true
+
+[dependencies]
+anyhow.workspace = true
+utils.workspace = true
+workspace_hack.workspace = true
+
+[dev-dependencies]
+rand.workspace = true
--- a/libs/remote_keys/src/lib.rs
+++ b/libs/remote_keys/src/lib.rs
@@ -0,0 +1,42 @@
+//! A module that provides a KMS implementation that generates and unwraps the keys.
+//!
+
+/// A KMS implementation that does static wrapping and unwrapping of the keys.
+pub struct NaiveKms {
+    account_id: String,
+}
+
+impl NaiveKms {
+    pub fn new(account_id: String) -> Self {
+        Self { account_id }
+    }
+
+    pub fn encrypt(&self, plain: &[u8]) -> anyhow::Result<Vec<u8>> {
+        let wrapped = [self.account_id.as_bytes(), "-wrapped-".as_bytes(), plain].concat();
+        Ok(wrapped)
+    }
+
+    pub fn decrypt(&self, wrapped: &[u8]) -> anyhow::Result<Vec<u8>> {
+        let Some(wrapped) = wrapped.strip_prefix(self.account_id.as_bytes()) else {
+            return Err(anyhow::anyhow!("invalid key"));
+        };
+        let Some(plain) = wrapped.strip_prefix(b"-wrapped-") else {
+            return Err(anyhow::anyhow!("invalid key"));
+        };
+        Ok(plain.to_vec())
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_generate_key() {
+        let kms = NaiveKms::new("test-tenant".to_string());
+        let data = rand::random::<[u8; 32]>().to_vec();
+        let encrypted = kms.encrypt(&data).unwrap();
+        let decrypted = kms.decrypt(&encrypted).unwrap();
+        assert_eq!(data, decrypted);
+    }
+}
--- a/libs/remote_storage/Cargo.toml
+++ b/libs/remote_storage/Cargo.toml
@@ -13,6 +13,7 @@ aws-smithy-async.workspace = true
 aws-smithy-types.workspace = true
 aws-config.workspace = true
 aws-sdk-s3.workspace = true
+base64.workspace = true
 bytes.workspace = true
 camino = { workspace = true, features = ["serde1"] }
 humantime-serde.workspace = true
@@ -27,8 +28,9 @@ tokio-util = { workspace = true, features = ["compat"] }
 toml_edit.workspace = true
 tracing.workspace = true
 scopeguard.workspace = true
+md5.workspace = true
 metrics.workspace = true
-utils.workspace = true
+utils = { path = "../utils", default-features = false }
 pin-project-lite.workspace = true

 azure_core.workspace = true
--- a/libs/remote_storage/src/azure_blob.rs
+++ b/libs/remote_storage/src/azure_blob.rs
@@ -550,6 +550,19 @@ impl RemoteStorage for AzureBlobStorage {
        self.download_for_builder(builder, timeout, cancel).await
    }

+    #[allow(unused_variables)]
+    async fn upload_with_encryption(
+        &self,
+        from: impl Stream<Item = std::io::Result<Bytes>> + Send + Sync + 'static,
+        data_size_bytes: usize,
+        to: &RemotePath,
+        metadata: Option<StorageMetadata>,
+        encryption_key: Option<&[u8]>,
+        cancel: &CancellationToken,
+    ) -> anyhow::Result<()> {
+        unimplemented!()
+    }
+
    async fn delete(&self, path: &RemotePath, cancel: &CancellationToken) -> anyhow::Result<()> {
        self.delete_objects(std::array::from_ref(path), cancel)
            .await
@@ -801,8 +814,7 @@ where
            // that support needs to be hacked in.
            //
            // including {self:?} into the message would be useful, but unsure how to unproject.
-            _ => std::task::Poll::Ready(Err(std::io::Error::new(
-                std::io::ErrorKind::Other,
+            _ => std::task::Poll::Ready(Err(std::io::Error::other(
                "cloned or initial values cannot be read",
            ))),
        }
@@ -855,7 +867,7 @@ where
        };
        Err(azure_core::error::Error::new(
            azure_core::error::ErrorKind::Io,
-            std::io::Error::new(std::io::ErrorKind::Other, msg),
+            std::io::Error::other(msg),
        ))
    }

--- a/libs/remote_storage/src/lib.rs
+++ b/libs/remote_storage/src/lib.rs
@@ -190,6 +190,8 @@ pub struct DownloadOpts {
    /// timeouts: for something like an index/manifest/heatmap, we should time out faster than
    /// for layer files
    pub kind: DownloadKind,
+    /// The encryption key to use for the download.
+    pub encryption_key: Option<Vec<u8>>,
 }

 pub enum DownloadKind {
@@ -204,6 +206,7 @@ impl Default for DownloadOpts {
            byte_start: Bound::Unbounded,
            byte_end: Bound::Unbounded,
            kind: DownloadKind::Large,
+            encryption_key: None,
        }
    }
 }
@@ -241,6 +244,15 @@ impl DownloadOpts {
                None => format!("bytes={start}-"),
            })
    }
+
+    pub fn with_encryption_key(mut self, encryption_key: Option<impl AsRef<[u8]>>) -> Self {
+        self.encryption_key = encryption_key.map(|k| k.as_ref().to_vec());
+        self
+    }
+
+    pub fn encryption_key(&self) -> Option<&[u8]> {
+        self.encryption_key.as_deref()
+    }
 }

 /// Storage (potentially remote) API to manage its state.
@@ -331,6 +343,19 @@ pub trait RemoteStorage: Send + Sync + 'static {
        cancel: &CancellationToken,
    ) -> Result<Download, DownloadError>;

+    /// Same as upload, but with remote encryption if the backend supports it (e.g. SSE-C on AWS).
+    async fn upload_with_encryption(
+        &self,
+        from: impl Stream<Item = std::io::Result<Bytes>> + Send + Sync + 'static,
+        // S3 PUT request requires the content length to be specified,
+        // otherwise it starts to fail with the concurrent connection count increasing.
+        data_size_bytes: usize,
+        to: &RemotePath,
+        metadata: Option<StorageMetadata>,
+        encryption_key: Option<&[u8]>,
+        cancel: &CancellationToken,
+    ) -> anyhow::Result<()>;
+
    /// Delete a single path from remote storage.
    ///
    /// If the operation fails because of timeout or cancellation, the root cause of the error will be
@@ -615,6 +640,63 @@ impl<Other: RemoteStorage> GenericRemoteStorage<Arc<Other>> {
            }
        }
    }
+
+    pub async fn upload_with_encryption(
+        &self,
+        from: impl Stream<Item = std::io::Result<Bytes>> + Send + Sync + 'static,
+        data_size_bytes: usize,
+        to: &RemotePath,
+        metadata: Option<StorageMetadata>,
+        encryption_key: Option<&[u8]>,
+        cancel: &CancellationToken,
+    ) -> anyhow::Result<()> {
+        match self {
+            Self::LocalFs(s) => {
+                s.upload_with_encryption(
+                    from,
+                    data_size_bytes,
+                    to,
+                    metadata,
+                    encryption_key,
+                    cancel,
+                )
+                .await
+            }
+            Self::AwsS3(s) => {
+                s.upload_with_encryption(
+                    from,
+                    data_size_bytes,
+                    to,
+                    metadata,
+                    encryption_key,
+                    cancel,
+                )
+                .await
+            }
+            Self::AzureBlob(s) => {
+                s.upload_with_encryption(
+                    from,
+                    data_size_bytes,
+                    to,
+                    metadata,
+                    encryption_key,
+                    cancel,
+                )
+                .await
+            }
+            Self::Unreliable(s) => {
+                s.upload_with_encryption(
+                    from,
+                    data_size_bytes,
+                    to,
+                    metadata,
+                    encryption_key,
+                    cancel,
+                )
+                .await
+            }
+        }
+    }
 }

 impl GenericRemoteStorage {
--- a/libs/remote_storage/src/local_fs.rs
+++ b/libs/remote_storage/src/local_fs.rs
@@ -560,6 +560,19 @@ impl RemoteStorage for LocalFs {
        }
    }

+    #[allow(unused_variables)]
+    async fn upload_with_encryption(
+        &self,
+        from: impl Stream<Item = std::io::Result<Bytes>> + Send + Sync + 'static,
+        data_size_bytes: usize,
+        to: &RemotePath,
+        metadata: Option<StorageMetadata>,
+        encryption_key: Option<&[u8]>,
+        cancel: &CancellationToken,
+    ) -> anyhow::Result<()> {
+        unimplemented!()
+    }
+
    async fn delete_objects(
        &self,
        paths: &[RemotePath],
--- a/libs/remote_storage/src/s3_bucket.rs
+++ b/libs/remote_storage/src/s3_bucket.rs
@@ -66,7 +66,10 @@ struct GetObjectRequest {
    key: String,
    etag: Option<String>,
    range: Option<String>,
+    /// Base64 encoded SSE-C key for server-side encryption.
+    sse_c_key: Option<Vec<u8>>,
 }
+
 impl S3Bucket {
    /// Creates the S3 storage, errors if incorrect AWS S3 configuration provided.
    pub async fn new(remote_storage_config: &S3Config, timeout: Duration) -> anyhow::Result<Self> {
@@ -257,6 +260,13 @@ impl S3Bucket {
            builder = builder.if_none_match(etag);
        }

+        if let Some(encryption_key) = request.sse_c_key {
+            builder = builder.sse_customer_algorithm("AES256");
+            builder = builder.sse_customer_key(base64::encode(&encryption_key));
+            builder = builder
+                .sse_customer_key_md5(base64::encode(md5::compute(&encryption_key).as_slice()));
+        }
+
        let get_object = builder.send();

        let get_object = tokio::select! {
@@ -693,12 +703,13 @@ impl RemoteStorage for S3Bucket {
        })
    }

-    async fn upload(
+    async fn upload_with_encryption(
        &self,
        from: impl Stream<Item = std::io::Result<Bytes>> + Send + Sync + 'static,
        from_size_bytes: usize,
        to: &RemotePath,
        metadata: Option<StorageMetadata>,
+        encryption_key: Option<&[u8]>,
        cancel: &CancellationToken,
    ) -> anyhow::Result<()> {
        let kind = RequestKind::Put;
@@ -709,7 +720,7 @@ impl RemoteStorage for S3Bucket {
        let body = StreamBody::new(from.map(|x| x.map(Frame::data)));
        let bytes_stream = ByteStream::new(SdkBody::from_body_1_x(body));

-        let upload = self
+        let mut upload = self
            .client
            .put_object()
            .bucket(self.bucket_name.clone())
@@ -717,8 +728,17 @@ impl RemoteStorage for S3Bucket {
            .set_metadata(metadata.map(|m| m.0))
            .set_storage_class(self.upload_storage_class.clone())
            .content_length(from_size_bytes.try_into()?)
-            .body(bytes_stream)
-            .send();
+            .body(bytes_stream);
+
+        if let Some(encryption_key) = encryption_key {
+            upload = upload.sse_customer_algorithm("AES256");
+            let base64_key = base64::encode(encryption_key);
+            upload = upload.sse_customer_key(&base64_key);
+            upload = upload
+                .sse_customer_key_md5(base64::encode(md5::compute(encryption_key).as_slice()));
+        }
+
+        let upload = upload.send();

        let upload = tokio::time::timeout(self.timeout, upload);

@@ -742,6 +762,18 @@ impl RemoteStorage for S3Bucket {
        }
    }

+    async fn upload(
+        &self,
+        from: impl Stream<Item = std::io::Result<Bytes>> + Send + Sync + 'static,
+        data_size_bytes: usize,
+        to: &RemotePath,
+        metadata: Option<StorageMetadata>,
+        cancel: &CancellationToken,
+    ) -> anyhow::Result<()> {
+        self.upload_with_encryption(from, data_size_bytes, to, metadata, None, cancel)
+            .await
+    }
+
    async fn copy(
        &self,
        from: &RemotePath,
@@ -801,6 +833,7 @@ impl RemoteStorage for S3Bucket {
                key: self.relative_path_to_s3_object(from),
                etag: opts.etag.as_ref().map(|e| e.to_string()),
                range: opts.byte_range_header(),
+                sse_c_key: opts.encryption_key.clone(),
            },
            cancel,
        )
--- a/libs/remote_storage/src/simulate_failures.rs
+++ b/libs/remote_storage/src/simulate_failures.rs
@@ -178,6 +178,19 @@ impl RemoteStorage for UnreliableWrapper {
        self.inner.download(from, opts, cancel).await
    }

+    #[allow(unused_variables)]
+    async fn upload_with_encryption(
+        &self,
+        from: impl Stream<Item = std::io::Result<Bytes>> + Send + Sync + 'static,
+        data_size_bytes: usize,
+        to: &RemotePath,
+        metadata: Option<StorageMetadata>,
+        encryption_key: Option<&[u8]>,
+        cancel: &CancellationToken,
+    ) -> anyhow::Result<()> {
+        unimplemented!()
+    }
+
    async fn delete(&self, path: &RemotePath, cancel: &CancellationToken) -> anyhow::Result<()> {
        self.delete_inner(path, true, cancel).await
    }
--- a/libs/remote_storage/tests/test_real_s3.rs
+++ b/libs/remote_storage/tests/test_real_s3.rs
@@ -421,7 +421,7 @@ async fn download_is_timeouted(ctx: &mut MaybeEnabledStorage) {
    ))
    .unwrap();

-    let len = upload_large_enough_file(&ctx.client, &path, &cancel).await;
+    let len = upload_large_enough_file(&ctx.client, &path, &cancel, None).await;

    let timeout = std::time::Duration::from_secs(5);

@@ -500,7 +500,7 @@ async fn download_is_cancelled(ctx: &mut MaybeEnabledStorage) {
    ))
    .unwrap();

-    let file_len = upload_large_enough_file(&ctx.client, &path, &cancel).await;
+    let file_len = upload_large_enough_file(&ctx.client, &path, &cancel, None).await;

    {
        let stream = ctx
@@ -555,6 +555,7 @@ async fn upload_large_enough_file(
    client: &GenericRemoteStorage,
    path: &RemotePath,
    cancel: &CancellationToken,
+    encryption_key: Option<&[u8]>,
 ) -> usize {
    let header = bytes::Bytes::from_static("remote blob data content".as_bytes());
    let body = bytes::Bytes::from(vec![0u8; 1024]);
@@ -565,9 +566,54 @@ async fn upload_large_enough_file(
    let contents = futures::stream::iter(contents.map(std::io::Result::Ok));

    client
-        .upload(contents, len, path, None, cancel)
+        .upload_with_encryption(contents, len, path, None, encryption_key, cancel)
        .await
        .expect("upload succeeds");

    len
 }
+
+#[test_context(MaybeEnabledStorage)]
+#[tokio::test]
+async fn encryption_works(ctx: &mut MaybeEnabledStorage) {
+    let MaybeEnabledStorage::Enabled(ctx) = ctx else {
+        return;
+    };
+
+    let cancel = CancellationToken::new();
+
+    let path = RemotePath::new(Utf8Path::new(
+        format!("{}/file_to_copy", ctx.base_prefix).as_str(),
+    ))
+    .unwrap();
+
+    let key = rand::random::<[u8; 32]>();
+    let file_len = upload_large_enough_file(&ctx.client, &path, &cancel, Some(&key)).await;
+
+    {
+        let download = ctx
+            .client
+            .download(
+                &path,
+                &DownloadOpts::default().with_encryption_key(Some(&key)),
+                &cancel,
+            )
+            .await
+            .expect("should succeed");
+        let vec = download_to_vec(download).await.expect("should succeed");
+        assert_eq!(vec.len(), file_len);
+    }
+
+    {
+        // Download without encryption key should fail
+        let download = ctx
+            .client
+            .download(&path, &DownloadOpts::default(), &cancel)
+            .await;
+        assert!(download.is_err());
+    }
+
+    let cancel = CancellationToken::new();
+
+    ctx.client.delete_objects(&[path], &cancel).await.unwrap();
+}
--- a/libs/tracing-utils/src/perf_span.rs
+++ b/libs/tracing-utils/src/perf_span.rs
@@ -28,7 +28,7 @@ use core::{
    task::{Context, Poll},
 };
 use pin_project_lite::pin_project;
-use tracing::{Dispatch, field, span::Span};
+use tracing::{Dispatch, span::Span};

 #[derive(Debug, Clone)]
 pub struct PerfSpan {
@@ -49,15 +49,6 @@ impl PerfSpan {
        }
    }

-    pub fn record<Q: field::AsField + ?Sized, V: field::Value>(
-        &self,
-        field: &Q,
-        value: V,
-    ) -> &Self {
-        self.inner.record(field, value);
-        self
-    }
-
    pub fn enter(&self) -> PerfSpanEntered {
        if let Some(ref id) = self.inner.id() {
            self.dispatch.enter(id);
--- a/libs/utils/Cargo.toml
+++ b/libs/utils/Cargo.toml
@@ -5,7 +5,8 @@ edition.workspace = true
 license.workspace = true

 [features]
-default = []
+default = ["rename_noreplace"]
+rename_noreplace = []
 # Enables test-only APIs, incuding failpoints. In particular, enables the `fail_point!` macro,
 # which adds some runtime cost to run tests on outage conditions
 testing = ["fail/failpoints"]
@@ -35,7 +36,7 @@ serde_with.workspace = true
 serde_json.workspace = true
 signal-hook.workspace = true
 thiserror.workspace = true
-tokio.workspace = true
+tokio = { workspace = true, features = ["signal"] }
 tokio-tar.workspace = true
 tokio-util.workspace = true
 toml_edit = { workspace = true, features = ["serde"] }
--- a/libs/utils/src/auth.rs
+++ b/libs/utils/src/auth.rs
@@ -173,7 +173,7 @@ impl std::fmt::Debug for JwtAuth {
 }

 // this function is used only for testing purposes in CLI e g generate tokens during init
-pub fn encode_from_key_file(claims: &Claims, key_data: &[u8]) -> Result<String> {
+pub fn encode_from_key_file<S: Serialize>(claims: &S, key_data: &[u8]) -> Result<String> {
    let key = EncodingKey::from_ed_pem(key_data)?;
    Ok(encode(&Header::new(STORAGE_TOKEN_ALGORITHM), claims, &key)?)
 }
--- a/libs/utils/src/crashsafe.rs
+++ b/libs/utils/src/crashsafe.rs
@@ -81,12 +81,9 @@ pub fn path_with_suffix_extension(
 }

 pub fn fsync_file_and_parent(file_path: &Utf8Path) -> io::Result<()> {
-    let parent = file_path.parent().ok_or_else(|| {
-        io::Error::new(
-            io::ErrorKind::Other,
-            format!("File {file_path:?} has no parent"),
-        )
-    })?;
+    let parent = file_path
+        .parent()
+        .ok_or_else(|| io::Error::other(format!("File {file_path:?} has no parent")))?;

    fsync(file_path)?;
    fsync(parent)?;
--- a/libs/utils/src/elapsed_accum.rs
+++ b/libs/utils/src/elapsed_accum.rs
@@ -0,0 +1,26 @@
+use std::time::{Duration, Instant};
+
+#[derive(Default)]
+pub struct ElapsedAccum {
+    accum: Duration,
+}
+
+impl ElapsedAccum {
+    pub fn get(&self) -> Duration {
+        self.accum
+    }
+    pub fn guard(&mut self) -> impl Drop + '_ {
+        let start = Instant::now();
+        scopeguard::guard(start, |last_wait_at| {
+            self.accum += Instant::now() - last_wait_at;
+        })
+    }
+
+    pub async fn measure<Fut, O>(&mut self, fut: Fut) -> O
+    where
+        Fut: Future<Output = O>,
+    {
+        let _guard = self.guard();
+        fut.await
+    }
+}
--- a/libs/utils/src/fs_ext.rs
+++ b/libs/utils/src/fs_ext.rs
@@ -3,7 +3,9 @@ use std::{fs, io, path::Path};

 use anyhow::Context;

+#[cfg(feature = "rename_noreplace")]
 mod rename_noreplace;
+#[cfg(feature = "rename_noreplace")]
 pub use rename_noreplace::rename_noreplace;

 pub trait PathExt {
--- a/libs/utils/src/fs_ext/rename_noreplace.rs
+++ b/libs/utils/src/fs_ext/rename_noreplace.rs
@@ -8,7 +8,7 @@ pub fn rename_noreplace<P1: ?Sized + NixPath, P2: ?Sized + NixPath>(
    dst: &P2,
 ) -> nix::Result<()> {
    {
-        #[cfg(target_os = "linux")]
+        #[cfg(all(target_os = "linux", target_env = "gnu"))]
        {
            nix::fcntl::renameat2(
                None,
@@ -29,7 +29,7 @@ pub fn rename_noreplace<P1: ?Sized + NixPath, P2: ?Sized + NixPath>(
            })??;
            nix::errno::Errno::result(res).map(drop)
        }
-        #[cfg(not(any(target_os = "linux", target_os = "macos")))]
+        #[cfg(not(any(all(target_os = "linux", target_env = "gnu"), target_os = "macos")))]
        {
            std::compile_error!("OS does not support no-replace renames");
        }
--- a/libs/utils/src/lib.rs
+++ b/libs/utils/src/lib.rs
@@ -93,6 +93,8 @@ pub mod try_rcu;

 pub mod guard_arc_swap;

+pub mod elapsed_accum;
+
 #[cfg(target_os = "linux")]
 pub mod linux_socket_ioctl;

--- a/libs/utils/src/signals.rs
+++ b/libs/utils/src/signals.rs
@@ -1,6 +1,8 @@
 pub use signal_hook::consts::TERM_SIGNALS;
 pub use signal_hook::consts::signal::*;
 use signal_hook::iterator::Signals;
+use tokio::signal::unix::{SignalKind, signal};
+use tracing::info;

 pub enum Signal {
    Quit,
@@ -36,3 +38,30 @@ impl ShutdownSignals {
        Ok(())
    }
 }
+
+/// Runs in a loop since we want to be responsive to multiple signals
+/// even after triggering shutdown (e.g. a SIGQUIT after a slow SIGTERM shutdown)
+/// <https://github.com/neondatabase/neon/issues/9740>
+pub async fn signal_handler(token: tokio_util::sync::CancellationToken) {
+    let mut sigint = signal(SignalKind::interrupt()).unwrap();
+    let mut sigterm = signal(SignalKind::terminate()).unwrap();
+    let mut sigquit = signal(SignalKind::quit()).unwrap();
+
+    loop {
+        let signal = tokio::select! {
+            _ = sigquit.recv() => {
+                info!("Got signal SIGQUIT. Terminating in immediate shutdown mode.");
+                std::process::exit(111);
+            }
+            _ = sigint.recv() => "SIGINT",
+            _ = sigterm.recv() => "SIGTERM",
+        };
+
+        if !token.is_cancelled() {
+            info!("Got signal {signal}. Terminating gracefully in fast shutdown mode.");
+            token.cancel();
+        } else {
+            info!("Got signal {signal}. Already shutting down.");
+        }
+    }
+}
--- a/libs/utils/src/sync/heavier_once_cell.rs
+++ b/libs/utils/src/sync/heavier_once_cell.rs
@@ -111,9 +111,17 @@ impl<T> OnceCell<T> {
        }
    }

+    /// Like [`Self::get_or_init_detached_measured`], but without out parameter for time spent waiting.
+    pub async fn get_or_init_detached(&self) -> Result<Guard<'_, T>, InitPermit> {
+        self.get_or_init_detached_measured(None).await
+    }
+
    /// Returns a guard to an existing initialized value, or returns an unique initialization
    /// permit which can be used to initialize this `OnceCell` using `OnceCell::set`.
-    pub async fn get_or_init_detached(&self) -> Result<Guard<'_, T>, InitPermit> {
+    pub async fn get_or_init_detached_measured(
+        &self,
+        mut wait_time: Option<&mut crate::elapsed_accum::ElapsedAccum>,
+    ) -> Result<Guard<'_, T>, InitPermit> {
        // It looks like OnceCell::get_or_init could be implemented using this method instead of
        // duplication. However, that makes the future be !Send due to possibly holding on to the
        // MutexGuard over an await point.
@@ -125,12 +133,16 @@ impl<T> OnceCell<T> {
                }
                guard.init_semaphore.clone()
            };
-
            {
                let permit = {
                    // increment the count for the duration of queued
                    let _guard = CountWaitingInitializers::start(self);
-                    sem.acquire().await
+                    let fut = sem.acquire();
+                    if let Some(wait_time) = wait_time.as_mut() {
+                        wait_time.measure(fut).await
+                    } else {
+                        fut.await
+                    }
                };

                let Ok(permit) = permit else {
--- a/object_storage/Cargo.toml
+++ b/object_storage/Cargo.toml
@@ -0,0 +1,28 @@
+[package]
+name = "object_storage"
+version = "0.0.1"
+edition.workspace = true
+license.workspace = true
+[dependencies]
+anyhow.workspace = true
+axum-extra.workspace = true
+axum.workspace = true
+camino.workspace = true
+futures.workspace = true
+jsonwebtoken.workspace = true
+prometheus.workspace = true
+remote_storage.workspace = true
+serde.workspace = true
+serde_json.workspace = true
+tokio-util.workspace = true
+tokio.workspace = true
+tracing.workspace = true
+utils = { path = "../libs/utils", default-features = false }
+workspace_hack.workspace = true
+[dev-dependencies]
+camino-tempfile.workspace = true
+http-body-util.workspace = true
+itertools.workspace = true
+rand.workspace = true
+test-log.workspace = true
+tower.workspace = true
--- a/object_storage/src/app.rs
+++ b/object_storage/src/app.rs
@@ -0,0 +1,561 @@
+use anyhow::anyhow;
+use axum::body::{Body, Bytes};
+use axum::response::{IntoResponse, Response};
+use axum::{Router, http::StatusCode};
+use object_storage::{PrefixS3Path, S3Path, Storage, bad_request, internal_error, not_found, ok};
+use remote_storage::TimeoutOrCancel;
+use remote_storage::{DownloadError, DownloadOpts, GenericRemoteStorage, RemotePath};
+use std::{sync::Arc, time::SystemTime, time::UNIX_EPOCH};
+use tokio_util::sync::CancellationToken;
+use tracing::{error, info};
+use utils::backoff::retry;
+
+pub fn app(state: Arc<Storage>) -> Router<()> {
+    use axum::routing::{delete as _delete, get as _get};
+    let delete_prefix = _delete(delete_prefix);
+    Router::new()
+        .route(
+            "/{tenant_id}/{timeline_id}/{endpoint_id}/{*path}",
+            _get(get).put(set).delete(delete),
+        )
+        .route(
+            "/{tenant_id}/{timeline_id}/{endpoint_id}",
+            delete_prefix.clone(),
+        )
+        .route("/{tenant_id}/{timeline_id}", delete_prefix.clone())
+        .route("/{tenant_id}", delete_prefix)
+        .route("/metrics", _get(metrics))
+        .route("/status", _get(async || StatusCode::OK.into_response()))
+        .with_state(state)
+}
+
+type Result = anyhow::Result<Response, Response>;
+type State = axum::extract::State<Arc<Storage>>;
+
+const CONTENT_TYPE: &str = "content-type";
+const APPLICATION_OCTET_STREAM: &str = "application/octet-stream";
+const WARN_THRESHOLD: u32 = 3;
+const MAX_RETRIES: u32 = 10;
+
+async fn metrics() -> Result {
+    prometheus::TextEncoder::new()
+        .encode_to_string(&prometheus::gather())
+        .map(|s| s.into_response())
+        .map_err(|e| internal_error(e, "/metrics", "collecting metrics"))
+}
+
+async fn get(S3Path { path }: S3Path, state: State) -> Result {
+    info!(%path, "downloading");
+    let download_err = |e| {
+        if let DownloadError::NotFound = e {
+            info!(%path, %e, "downloading"); // 404 is not an issue of _this_ service
+            return not_found(&path);
+        }
+        internal_error(e, &path, "downloading")
+    };
+    let cancel = state.cancel.clone();
+    let opts = &DownloadOpts::default();
+
+    let stream = retry(
+        async || state.storage.download(&path, opts, &cancel).await,
+        DownloadError::is_permanent,
+        WARN_THRESHOLD,
+        MAX_RETRIES,
+        "downloading",
+        &cancel,
+    )
+    .await
+    .unwrap_or(Err(DownloadError::Cancelled))
+    .map_err(download_err)?
+    .download_stream;
+
+    Response::builder()
+        .status(StatusCode::OK)
+        .header(CONTENT_TYPE, APPLICATION_OCTET_STREAM)
+        .body(Body::from_stream(stream))
+        .map_err(|e| internal_error(e, path, "reading response"))
+}
+
+// Best solution for files is multipart upload, but remote_storage doesn't support it,
+// so we can either read Bytes in memory and push at once or forward BodyDataStream to
+// remote_storage. The latter may seem more peformant, but BodyDataStream doesn't have a
+// guaranteed size() which may produce issues while uploading to s3.
+// So, currently we're going with an in-memory copy plus a boundary to prevent uploading
+// very large files.
+async fn set(S3Path { path }: S3Path, state: State, bytes: Bytes) -> Result {
+    info!(%path, "uploading");
+    let request_len = bytes.len();
+    let max_len = state.max_upload_file_limit;
+    if request_len > max_len {
+        return Err(bad_request(
+            anyhow!("File size {request_len} exceeds max {max_len}"),
+            "uploading",
+        ));
+    }
+
+    let cancel = state.cancel.clone();
+    let fun = async || {
+        let stream = bytes_to_stream(bytes.clone());
+        state
+            .storage
+            .upload(stream, request_len, &path, None, &cancel)
+            .await
+    };
+    retry(
+        fun,
+        TimeoutOrCancel::caused_by_cancel,
+        WARN_THRESHOLD,
+        MAX_RETRIES,
+        "uploading",
+        &cancel,
+    )
+    .await
+    .unwrap_or(Err(anyhow!("uploading cancelled")))
+    .map_err(|e| internal_error(e, path, "reading response"))?;
+    Ok(ok())
+}
+
+async fn delete(S3Path { path }: S3Path, state: State) -> Result {
+    info!(%path, "deleting");
+    let cancel = state.cancel.clone();
+    retry(
+        async || state.storage.delete(&path, &cancel).await,
+        TimeoutOrCancel::caused_by_cancel,
+        WARN_THRESHOLD,
+        MAX_RETRIES,
+        "deleting",
+        &cancel,
+    )
+    .await
+    .unwrap_or(Err(anyhow!("deleting cancelled")))
+    .map_err(|e| internal_error(e, path, "deleting"))?;
+    Ok(ok())
+}
+
+async fn delete_prefix(PrefixS3Path { path }: PrefixS3Path, state: State) -> Result {
+    info!(%path, "deleting prefix");
+    let cancel = state.cancel.clone();
+    retry(
+        async || state.storage.delete_prefix(&path, &cancel).await,
+        TimeoutOrCancel::caused_by_cancel,
+        WARN_THRESHOLD,
+        MAX_RETRIES,
+        "deleting prefix",
+        &cancel,
+    )
+    .await
+    .unwrap_or(Err(anyhow!("deleting prefix cancelled")))
+    .map_err(|e| internal_error(e, path, "deleting prefix"))?;
+    Ok(ok())
+}
+
+pub async fn check_storage_permissions(
+    client: &GenericRemoteStorage,
+    cancel: CancellationToken,
+) -> anyhow::Result<()> {
+    info!("storage permissions check");
+
+    // as_nanos() as multiple instances proxying same bucket may be started at once
+    let now = SystemTime::now()
+        .duration_since(UNIX_EPOCH)?
+        .as_nanos()
+        .to_string();
+
+    let path = RemotePath::from_string(&format!("write_access_{now}"))?;
+    info!(%path, "uploading");
+
+    let body = now.to_string();
+    let stream = bytes_to_stream(Bytes::from(body.clone()));
+    client
+        .upload(stream, body.len(), &path, None, &cancel)
+        .await?;
+
+    use tokio::io::AsyncReadExt;
+    info!(%path, "downloading");
+    let download_opts = DownloadOpts {
+        kind: remote_storage::DownloadKind::Small,
+        ..Default::default()
+    };
+    let mut body_read_buf = Vec::new();
+    let stream = client
+        .download(&path, &download_opts, &cancel)
+        .await?
+        .download_stream;
+    tokio_util::io::StreamReader::new(stream)
+        .read_to_end(&mut body_read_buf)
+        .await?;
+    let body_read = String::from_utf8(body_read_buf)?;
+    if body != body_read {
+        error!(%body, %body_read, "File contents do not match");
+        anyhow::bail!("Read back file doesn't match original")
+    }
+
+    info!(%path, "removing");
+    client.delete(&path, &cancel).await
+}
+
+fn bytes_to_stream(bytes: Bytes) -> impl futures::Stream<Item = std::io::Result<Bytes>> {
+    futures::stream::once(futures::future::ready(Ok(bytes)))
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use axum::{body::Body, extract::Request, response::Response};
+    use http_body_util::BodyExt;
+    use itertools::iproduct;
+    use std::env::var;
+    use std::sync::Arc;
+    use std::time::Duration;
+    use test_log::test as testlog;
+    use tower::{Service, util::ServiceExt};
+    use utils::id::{TenantId, TimelineId};
+
+    // see libs/remote_storage/tests/test_real_s3.rs
+    const REAL_S3_ENV: &str = "ENABLE_REAL_S3_REMOTE_STORAGE";
+    const REAL_S3_BUCKET: &str = "REMOTE_STORAGE_S3_BUCKET";
+    const REAL_S3_REGION: &str = "REMOTE_STORAGE_S3_REGION";
+
+    async fn proxy() -> (Storage, Option<camino_tempfile::Utf8TempDir>) {
+        let cancel = CancellationToken::new();
+        let (dir, storage) = if var(REAL_S3_ENV).is_err() {
+            // tests execute in parallel and we need a new directory for each of them
+            let dir = camino_tempfile::tempdir().unwrap();
+            let fs =
+                remote_storage::LocalFs::new(dir.path().into(), Duration::from_secs(5)).unwrap();
+            (Some(dir), GenericRemoteStorage::LocalFs(fs))
+        } else {
+            // test_real_s3::create_s3_client is hard to reference, reimplementing here
+            let millis = SystemTime::now()
+                .duration_since(UNIX_EPOCH)
+                .unwrap()
+                .as_millis();
+            use rand::Rng;
+            let random = rand::thread_rng().r#gen::<u32>();
+
+            let s3_config = remote_storage::S3Config {
+                bucket_name: var(REAL_S3_BUCKET).unwrap(),
+                bucket_region: var(REAL_S3_REGION).unwrap(),
+                prefix_in_bucket: Some(format!("test_{millis}_{random:08x}/")),
+                endpoint: None,
+                concurrency_limit: std::num::NonZeroUsize::new(100).unwrap(),
+                max_keys_per_list_response: None,
+                upload_storage_class: None,
+            };
+            let bucket = remote_storage::S3Bucket::new(&s3_config, Duration::from_secs(1))
+                .await
+                .unwrap();
+            (None, GenericRemoteStorage::AwsS3(Arc::new(bucket)))
+        };
+
+        let proxy = Storage {
+            auth: object_storage::JwtAuth::new(TEST_PUB_KEY_ED25519).unwrap(),
+            storage,
+            cancel: cancel.clone(),
+            max_upload_file_limit: usize::MAX,
+        };
+        check_storage_permissions(&proxy.storage, cancel)
+            .await
+            .unwrap();
+        (proxy, dir)
+    }
+
+    // see libs/utils/src/auth.rs
+    const TEST_PUB_KEY_ED25519: &[u8] = b"
+-----BEGIN PUBLIC KEY-----
+MCowBQYDK2VwAyEARYwaNBayR+eGI0iXB4s3QxE3Nl2g1iWbr6KtLWeVD/w=
+-----END PUBLIC KEY-----
+";
+
+    const TEST_PRIV_KEY_ED25519: &[u8] = br#"
+-----BEGIN PRIVATE KEY-----
+MC4CAQAwBQYDK2VwBCIEID/Drmc1AA6U/znNRWpF3zEGegOATQxfkdWxitcOMsIH
+-----END PRIVATE KEY-----
+"#;
+
+    async fn request(req: Request<Body>) -> Response<Body> {
+        let (proxy, _) = proxy().await;
+        app(Arc::new(proxy))
+            .into_service()
+            .oneshot(req)
+            .await
+            .unwrap()
+    }
+
+    #[testlog(tokio::test)]
+    async fn status() {
+        let res = Request::builder()
+            .uri("/status")
+            .body(Body::empty())
+            .map(request)
+            .unwrap()
+            .await;
+        assert_eq!(res.status(), StatusCode::OK);
+    }
+
+    fn routes() -> impl Iterator<Item = (&'static str, &'static str)> {
+        iproduct!(
+            vec!["/1", "/1/2", "/1/2/3", "/1/2/3/4"],
+            vec!["GET", "PUT", "DELETE"]
+        )
+    }
+
+    #[testlog(tokio::test)]
+    async fn no_token() {
+        for (uri, method) in routes() {
+            info!(%uri, %method);
+            let res = Request::builder()
+                .uri(uri)
+                .method(method)
+                .body(Body::empty())
+                .map(request)
+                .unwrap()
+                .await;
+            assert!(matches!(
+                res.status(),
+                StatusCode::METHOD_NOT_ALLOWED | StatusCode::BAD_REQUEST
+            ));
+        }
+    }
+
+    #[testlog(tokio::test)]
+    async fn invalid_token() {
+        for (uri, method) in routes() {
+            info!(%uri, %method);
+            let status = Request::builder()
+                .uri(uri)
+                .header("Authorization", "Bearer 123")
+                .method(method)
+                .body(Body::empty())
+                .map(request)
+                .unwrap()
+                .await;
+            assert!(matches!(
+                status.status(),
+                StatusCode::METHOD_NOT_ALLOWED | StatusCode::BAD_REQUEST
+            ));
+        }
+    }
+
+    const TENANT_ID: TenantId =
+        TenantId::from_array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4, 5, 6]);
+    const TIMELINE_ID: TimelineId =
+        TimelineId::from_array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4, 5, 7]);
+    const ENDPOINT_ID: &str = "ep-winter-frost-a662z3vg";
+    fn token() -> String {
+        let claims = object_storage::Claims {
+            tenant_id: TENANT_ID,
+            timeline_id: TIMELINE_ID,
+            endpoint_id: ENDPOINT_ID.into(),
+            exp: u64::MAX,
+        };
+        let key = jsonwebtoken::EncodingKey::from_ed_pem(TEST_PRIV_KEY_ED25519).unwrap();
+        let header = jsonwebtoken::Header::new(object_storage::VALIDATION_ALGO);
+        jsonwebtoken::encode(&header, &claims, &key).unwrap()
+    }
+
+    #[testlog(tokio::test)]
+    async fn unauthorized() {
+        let (proxy, _) = proxy().await;
+        let mut app = app(Arc::new(proxy)).into_service();
+        let token = token();
+        let args = itertools::iproduct!(
+            vec![TENANT_ID.to_string(), TenantId::generate().to_string()],
+            vec![TIMELINE_ID.to_string(), TimelineId::generate().to_string()],
+            vec![ENDPOINT_ID, "ep-ololo"]
+        )
+        .skip(1);
+
+        for ((uri, method), (tenant, timeline, endpoint)) in iproduct!(routes(), args) {
+            info!(%uri, %method, %tenant, %timeline, %endpoint);
+            let request = Request::builder()
+                .uri(format!("/{tenant}/{timeline}/{endpoint}/sub/path/key"))
+                .method(method)
+                .header("Authorization", format!("Bearer {}", token))
+                .body(Body::empty())
+                .unwrap();
+            let status = ServiceExt::ready(&mut app)
+                .await
+                .unwrap()
+                .call(request)
+                .await
+                .unwrap()
+                .status();
+            assert_eq!(status, StatusCode::UNAUTHORIZED);
+        }
+    }
+
+    #[testlog(tokio::test)]
+    async fn method_not_allowed() {
+        let token = token();
+        let iter = iproduct!(vec!["", "/.."], vec!["GET", "PUT"]);
+        for (key, method) in iter {
+            let status = Request::builder()
+                .uri(format!("/{TENANT_ID}/{TIMELINE_ID}/{ENDPOINT_ID}{key}"))
+                .method(method)
+                .header("Authorization", format!("Bearer {token}"))
+                .body(Body::empty())
+                .map(request)
+                .unwrap()
+                .await
+                .status();
+            assert!(matches!(
+                status,
+                StatusCode::BAD_REQUEST | StatusCode::METHOD_NOT_ALLOWED
+            ));
+        }
+    }
+
+    async fn requests_chain(
+        chain: impl Iterator<Item = (String, &str, &'static str, StatusCode, bool)>,
+        token: impl Fn(&str) -> String,
+    ) {
+        let (proxy, _) = proxy().await;
+        let mut app = app(Arc::new(proxy)).into_service();
+        for (uri, method, body, expected_status, compare_body) in chain {
+            info!(%uri, %method, %body, %expected_status);
+            let bearer = format!("Bearer {}", token(&uri));
+            let request = Request::builder()
+                .uri(uri)
+                .method(method)
+                .header("Authorization", &bearer)
+                .body(Body::from(body))
+                .unwrap();
+            let response = ServiceExt::ready(&mut app)
+                .await
+                .unwrap()
+                .call(request)
+                .await
+                .unwrap();
+            assert_eq!(response.status(), expected_status);
+            if !compare_body {
+                continue;
+            }
+            let read_body = response.into_body().collect().await.unwrap().to_bytes();
+            assert_eq!(body, read_body);
+        }
+    }
+
+    #[testlog(tokio::test)]
+    async fn metrics() {
+        let uri = format!("/{TENANT_ID}/{TIMELINE_ID}/{ENDPOINT_ID}/key");
+        let req = vec![
+            (uri.clone(), "PUT", "body", StatusCode::OK, false),
+            (uri.clone(), "DELETE", "", StatusCode::OK, false),
+        ];
+        requests_chain(req.into_iter(), |_| token()).await;
+
+        let res = Request::builder()
+            .uri("/metrics")
+            .body(Body::empty())
+            .map(request)
+            .unwrap()
+            .await;
+        assert_eq!(res.status(), StatusCode::OK);
+        let body = res.into_body().collect().await.unwrap().to_bytes();
+        let body = String::from_utf8_lossy(&body);
+        tracing::debug!(%body);
+        // Storage metrics are not gathered for LocalFs
+        if var(REAL_S3_ENV).is_ok() {
+            assert!(body.contains("remote_storage_s3_deleted_objects_total"));
+        }
+        assert!(body.contains("process_threads"));
+    }
+
+    #[testlog(tokio::test)]
+    async fn insert_retrieve_remove() {
+        let uri = format!("/{TENANT_ID}/{TIMELINE_ID}/{ENDPOINT_ID}/key");
+        let chain = vec![
+            (uri.clone(), "GET", "", StatusCode::NOT_FOUND, false),
+            (uri.clone(), "PUT", "пыщьпыщь", StatusCode::OK, false),
+            (uri.clone(), "GET", "пыщьпыщь", StatusCode::OK, true),
+            (uri.clone(), "DELETE", "", StatusCode::OK, false),
+            (uri, "GET", "", StatusCode::NOT_FOUND, false),
+        ];
+        requests_chain(chain.into_iter(), |_| token()).await;
+    }
+
+    fn delete_prefix_token(uri: &str) -> String {
+        use serde::Serialize;
+        let parts = uri.split("/").collect::<Vec<&str>>();
+        #[derive(Serialize)]
+        struct PrefixClaims {
+            tenant_id: TenantId,
+            timeline_id: Option<TimelineId>,
+            endpoint_id: Option<object_storage::EndpointId>,
+            exp: u64,
+        }
+        let claims = PrefixClaims {
+            tenant_id: parts.get(1).map(|c| c.parse().unwrap()).unwrap(),
+            timeline_id: parts.get(2).map(|c| c.parse().unwrap()),
+            endpoint_id: parts.get(3).map(ToString::to_string),
+            exp: u64::MAX,
+        };
+        let key = jsonwebtoken::EncodingKey::from_ed_pem(TEST_PRIV_KEY_ED25519).unwrap();
+        let header = jsonwebtoken::Header::new(object_storage::VALIDATION_ALGO);
+        jsonwebtoken::encode(&header, &claims, &key).unwrap()
+    }
+
+    // Can't use single digit numbers as they won't be validated as TimelineId and EndpointId
+    #[testlog(tokio::test)]
+    async fn delete_prefix() {
+        let tenant_id =
+            TenantId::from_array([1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]).to_string();
+        let t2 = TimelineId::from_array([2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]);
+        let t3 = TimelineId::from_array([3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]);
+        let t4 = TimelineId::from_array([4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]);
+        let f = |timeline, path| format!("/{tenant_id}/{timeline}{path}");
+        // Why extra slash in string literals? Axum is weird with URIs:
+        // /1/2 and 1/2/ match different routes, thus first yields OK and second NOT_FOUND
+        //  as it matches /tenant/timeline/endpoint, see https://stackoverflow.com/a/75355932
+        // The cost of removing trailing slash is suprisingly hard:
+        // * Add tower dependency with NormalizePath layer
+        // * wrap Router<()> in this layer https://github.com/tokio-rs/axum/discussions/2377
+        // * Rewrite make_service() -> into_make_service()
+        // * Rewrite oneshot() (not available for NormalizePath)
+        // I didn't manage to get it working correctly
+        let chain = vec![
+            // create 1/2/3/4, 1/2/3/5, delete prefix 1/2/3 -> empty
+            (f(t2, "/3/4"), "PUT", "", StatusCode::OK, false),
+            (f(t2, "/3/4"), "PUT", "", StatusCode::OK, false), // we can override file contents
+            (f(t2, "/3/5"), "PUT", "", StatusCode::OK, false),
+            (f(t2, "/3"), "DELETE", "", StatusCode::OK, false),
+            (f(t2, "/3/4"), "GET", "", StatusCode::NOT_FOUND, false),
+            (f(t2, "/3/5"), "GET", "", StatusCode::NOT_FOUND, false),
+            // create 1/2/3/4, 1/2/5/6, delete prefix 1/2/3 -> 1/2/5/6
+            (f(t2, "/3/4"), "PUT", "", StatusCode::OK, false),
+            (f(t2, "/5/6"), "PUT", "", StatusCode::OK, false),
+            (f(t2, "/3"), "DELETE", "", StatusCode::OK, false),
+            (f(t2, "/3/4"), "GET", "", StatusCode::NOT_FOUND, false),
+            (f(t2, "/5/6"), "GET", "", StatusCode::OK, false),
+            // create 1/2/3/4, 1/2/7/8, delete prefix 1/2 -> empty
+            (f(t2, "/3/4"), "PUT", "", StatusCode::OK, false),
+            (f(t2, "/7/8"), "PUT", "", StatusCode::OK, false),
+            (f(t2, ""), "DELETE", "", StatusCode::OK, false),
+            (f(t2, "/3/4"), "GET", "", StatusCode::NOT_FOUND, false),
+            (f(t2, "/7/8"), "GET", "", StatusCode::NOT_FOUND, false),
+            // create 1/2/3/4, 1/2/5/6, 1/3/8/9, delete prefix 1/2/3 -> 1/2/5/6, 1/3/8/9
+            (f(t2, "/3/4"), "PUT", "", StatusCode::OK, false),
+            (f(t2, "/5/6"), "PUT", "", StatusCode::OK, false),
+            (f(t3, "/8/9"), "PUT", "", StatusCode::OK, false),
+            (f(t2, "/3"), "DELETE", "", StatusCode::OK, false),
+            (f(t2, "/3/4"), "GET", "", StatusCode::NOT_FOUND, false),
+            (f(t2, "/5/6"), "GET", "", StatusCode::OK, false),
+            (f(t3, "/8/9"), "GET", "", StatusCode::OK, false),
+            // create 1/4/5/6, delete prefix 1/2 -> 1/3/8/9, 1/4/5/6
+            (f(t4, "/5/6"), "PUT", "", StatusCode::OK, false),
+            (f(t2, ""), "DELETE", "", StatusCode::OK, false),
+            (f(t2, "/3/4"), "GET", "", StatusCode::NOT_FOUND, false),
+            (f(t2, "/5/6"), "GET", "", StatusCode::NOT_FOUND, false),
+            (f(t3, "/8/9"), "GET", "", StatusCode::OK, false),
+            (f(t4, "/5/6"), "GET", "", StatusCode::OK, false),
+            // delete prefix 1 -> empty
+            (format!("/{tenant_id}"), "DELETE", "", StatusCode::OK, false),
+            (f(t2, "/3/4"), "GET", "", StatusCode::NOT_FOUND, false),
+            (f(t2, "/5/6"), "GET", "", StatusCode::NOT_FOUND, false),
+            (f(t3, "/8/9"), "GET", "", StatusCode::NOT_FOUND, false),
+            (f(t4, "/5/6"), "GET", "", StatusCode::NOT_FOUND, false),
+        ];
+        requests_chain(chain.into_iter(), delete_prefix_token).await;
+    }
+}
--- a/object_storage/src/lib.rs
+++ b/object_storage/src/lib.rs
@@ -0,0 +1,344 @@
+use anyhow::Result;
+use axum::extract::{FromRequestParts, Path};
+use axum::response::{IntoResponse, Response};
+use axum::{RequestPartsExt, http::StatusCode, http::request::Parts};
+use axum_extra::TypedHeader;
+use axum_extra::headers::{Authorization, authorization::Bearer};
+use camino::Utf8PathBuf;
+use jsonwebtoken::{DecodingKey, Validation};
+use remote_storage::{GenericRemoteStorage, RemotePath};
+use serde::{Deserialize, Serialize};
+use std::fmt::Display;
+use std::result::Result as StdResult;
+use std::sync::Arc;
+use tokio_util::sync::CancellationToken;
+use tracing::{debug, error};
+use utils::id::{TenantId, TimelineId};
+
+// simplified version of utils::auth::JwtAuth
+pub struct JwtAuth {
+    decoding_key: DecodingKey,
+    validation: Validation,
+}
+
+pub const VALIDATION_ALGO: jsonwebtoken::Algorithm = jsonwebtoken::Algorithm::EdDSA;
+impl JwtAuth {
+    pub fn new(key: &[u8]) -> Result<Self> {
+        Ok(Self {
+            decoding_key: DecodingKey::from_ed_pem(key)?,
+            validation: Validation::new(VALIDATION_ALGO),
+        })
+    }
+
+    pub fn decode<T: serde::de::DeserializeOwned>(&self, token: &str) -> Result<T> {
+        Ok(jsonwebtoken::decode(token, &self.decoding_key, &self.validation).map(|t| t.claims)?)
+    }
+}
+
+fn normalize_key(key: &str) -> StdResult<Utf8PathBuf, String> {
+    let key = clean_utf8(&Utf8PathBuf::from(key));
+    if key.starts_with("..") || key == "." || key == "/" {
+        return Err(format!("invalid key {key}"));
+    }
+    match key.strip_prefix("/").map(Utf8PathBuf::from) {
+        Ok(p) => Ok(p),
+        _ => Ok(key),
+    }
+}
+
+// Copied from path_clean crate with PathBuf->Utf8PathBuf
+fn clean_utf8(path: &camino::Utf8Path) -> Utf8PathBuf {
+    use camino::Utf8Component as Comp;
+    let mut out = Vec::new();
+    for comp in path.components() {
+        match comp {
+            Comp::CurDir => (),
+            Comp::ParentDir => match out.last() {
+                Some(Comp::RootDir) => (),
+                Some(Comp::Normal(_)) => {
+                    out.pop();
+                }
+                None | Some(Comp::CurDir) | Some(Comp::ParentDir) | Some(Comp::Prefix(_)) => {
+                    out.push(comp)
+                }
+            },
+            comp => out.push(comp),
+        }
+    }
+    if !out.is_empty() {
+        out.iter().collect()
+    } else {
+        Utf8PathBuf::from(".")
+    }
+}
+
+pub struct Storage {
+    pub auth: JwtAuth,
+    pub storage: GenericRemoteStorage,
+    pub cancel: CancellationToken,
+    pub max_upload_file_limit: usize,
+}
+
+pub type EndpointId = String; // If needed, reuse small string from proxy/src/types.rc
+
+#[derive(Deserialize, Serialize, PartialEq)]
+pub struct Claims {
+    pub tenant_id: TenantId,
+    pub timeline_id: TimelineId,
+    pub endpoint_id: EndpointId,
+    pub exp: u64,
+}
+
+impl Display for Claims {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        write!(
+            f,
+            "Claims(tenant_id {} timeline_id {} endpoint_id {} exp {})",
+            self.tenant_id, self.timeline_id, self.endpoint_id, self.exp
+        )
+    }
+}
+
+#[derive(Deserialize, Serialize)]
+struct KeyRequest {
+    tenant_id: TenantId,
+    timeline_id: TimelineId,
+    endpoint_id: EndpointId,
+    path: String,
+}
+
+#[derive(Debug, PartialEq)]
+pub struct S3Path {
+    pub path: RemotePath,
+}
+
+impl TryFrom<&KeyRequest> for S3Path {
+    type Error = String;
+    fn try_from(req: &KeyRequest) -> StdResult<Self, Self::Error> {
+        let KeyRequest {
+            tenant_id,
+            timeline_id,
+            endpoint_id,
+            path,
+        } = &req;
+        let prefix = format!("{tenant_id}/{timeline_id}/{endpoint_id}",);
+        let path = Utf8PathBuf::from(prefix).join(normalize_key(path)?);
+        let path = RemotePath::new(&path).unwrap(); // unwrap() because the path is already relative
+        Ok(S3Path { path })
+    }
+}
+
+fn unauthorized(route: impl Display, claims: impl Display) -> Response {
+    debug!(%route, %claims, "route doesn't match claims");
+    StatusCode::UNAUTHORIZED.into_response()
+}
+
+pub fn bad_request(err: impl Display, desc: &'static str) -> Response {
+    debug!(%err, desc);
+    (StatusCode::BAD_REQUEST, err.to_string()).into_response()
+}
+
+pub fn ok() -> Response {
+    StatusCode::OK.into_response()
+}
+
+pub fn internal_error(err: impl Display, path: impl Display, desc: &'static str) -> Response {
+    error!(%err, %path, desc);
+    StatusCode::INTERNAL_SERVER_ERROR.into_response()
+}
+
+pub fn not_found(key: impl ToString) -> Response {
+    (StatusCode::NOT_FOUND, key.to_string()).into_response()
+}
+
+impl FromRequestParts<Arc<Storage>> for S3Path {
+    type Rejection = Response;
+    async fn from_request_parts(
+        parts: &mut Parts,
+        state: &Arc<Storage>,
+    ) -> Result<Self, Self::Rejection> {
+        let Path(path): Path<KeyRequest> = parts
+            .extract()
+            .await
+            .map_err(|e| bad_request(e, "invalid route"))?;
+        let TypedHeader(Authorization(bearer)) = parts
+            .extract::<TypedHeader<Authorization<Bearer>>>()
+            .await
+            .map_err(|e| bad_request(e, "invalid token"))?;
+        let claims: Claims = state
+            .auth
+            .decode(bearer.token())
+            .map_err(|e| bad_request(e, "decoding token"))?;
+        let route = Claims {
+            tenant_id: path.tenant_id,
+            timeline_id: path.timeline_id,
+            endpoint_id: path.endpoint_id.clone(),
+            exp: claims.exp,
+        };
+        if route != claims {
+            return Err(unauthorized(route, claims));
+        }
+        (&path)
+            .try_into()
+            .map_err(|e| bad_request(e, "invalid route"))
+    }
+}
+
+#[derive(Deserialize, Serialize, PartialEq)]
+pub struct PrefixKeyPath {
+    pub tenant_id: TenantId,
+    pub timeline_id: Option<TimelineId>,
+    pub endpoint_id: Option<EndpointId>,
+}
+
+impl Display for PrefixKeyPath {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        write!(
+            f,
+            "PrefixKeyPath(tenant_id {} timeline_id {} endpoint_id {})",
+            self.tenant_id,
+            self.timeline_id
+                .as_ref()
+                .map(ToString::to_string)
+                .unwrap_or("".to_string()),
+            self.endpoint_id
+                .as_ref()
+                .map(ToString::to_string)
+                .unwrap_or("".to_string())
+        )
+    }
+}
+
+#[derive(Debug, PartialEq)]
+pub struct PrefixS3Path {
+    pub path: RemotePath,
+}
+
+impl From<&PrefixKeyPath> for PrefixS3Path {
+    fn from(path: &PrefixKeyPath) -> Self {
+        let timeline_id = path
+            .timeline_id
+            .as_ref()
+            .map(ToString::to_string)
+            .unwrap_or("".to_string());
+        let endpoint_id = path
+            .endpoint_id
+            .as_ref()
+            .map(ToString::to_string)
+            .unwrap_or("".to_string());
+        let path = Utf8PathBuf::from(path.tenant_id.to_string())
+            .join(timeline_id)
+            .join(endpoint_id);
+        let path = RemotePath::new(&path).unwrap(); // unwrap() because the path is already relative
+        PrefixS3Path { path }
+    }
+}
+
+impl FromRequestParts<Arc<Storage>> for PrefixS3Path {
+    type Rejection = Response;
+    async fn from_request_parts(
+        parts: &mut Parts,
+        state: &Arc<Storage>,
+    ) -> Result<Self, Self::Rejection> {
+        let Path(path) = parts
+            .extract::<Path<PrefixKeyPath>>()
+            .await
+            .map_err(|e| bad_request(e, "invalid route"))?;
+        let TypedHeader(Authorization(bearer)) = parts
+            .extract::<TypedHeader<Authorization<Bearer>>>()
+            .await
+            .map_err(|e| bad_request(e, "invalid token"))?;
+        let claims: PrefixKeyPath = state
+            .auth
+            .decode(bearer.token())
+            .map_err(|e| bad_request(e, "invalid token"))?;
+        if path != claims {
+            return Err(unauthorized(path, claims));
+        }
+        Ok((&path).into())
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn normalize_key() {
+        let f = super::normalize_key;
+        assert_eq!(f("hello/world/..").unwrap(), Utf8PathBuf::from("hello"));
+        assert_eq!(
+            f("ololo/1/../../not_ololo").unwrap(),
+            Utf8PathBuf::from("not_ololo")
+        );
+        assert!(f("ololo/1/../../../").is_err());
+        assert!(f(".").is_err());
+        assert!(f("../").is_err());
+        assert!(f("").is_err());
+        assert_eq!(f("/1/2/3").unwrap(), Utf8PathBuf::from("1/2/3"));
+        assert!(f("/1/2/3/../../../").is_err());
+        assert!(f("/1/2/3/../../../../").is_err());
+    }
+
+    const TENANT_ID: TenantId =
+        TenantId::from_array([1, 1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4, 5, 6]);
+    const TIMELINE_ID: TimelineId =
+        TimelineId::from_array([1, 1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4, 5, 7]);
+    const ENDPOINT_ID: &str = "ep-winter-frost-a662z3vg";
+
+    #[test]
+    fn s3_path() {
+        let auth = Claims {
+            tenant_id: TENANT_ID,
+            timeline_id: TIMELINE_ID,
+            endpoint_id: ENDPOINT_ID.into(),
+            exp: u64::MAX,
+        };
+        let s3_path = |key| {
+            let path = &format!("{TENANT_ID}/{TIMELINE_ID}/{ENDPOINT_ID}/{key}");
+            let path = RemotePath::from_string(path).unwrap();
+            S3Path { path }
+        };
+
+        let path = "cache_key".to_string();
+        let mut key_path = KeyRequest {
+            path,
+            tenant_id: auth.tenant_id,
+            timeline_id: auth.timeline_id,
+            endpoint_id: auth.endpoint_id,
+        };
+        assert_eq!(S3Path::try_from(&key_path).unwrap(), s3_path(key_path.path));
+
+        key_path.path = "we/can/have/nested/paths".to_string();
+        assert_eq!(S3Path::try_from(&key_path).unwrap(), s3_path(key_path.path));
+
+        key_path.path = "../error/hello/../".to_string();
+        assert!(S3Path::try_from(&key_path).is_err());
+    }
+
+    #[test]
+    fn prefix_s3_path() {
+        let mut path = PrefixKeyPath {
+            tenant_id: TENANT_ID,
+            timeline_id: None,
+            endpoint_id: None,
+        };
+        let prefix_path = |s: String| RemotePath::from_string(&s).unwrap();
+        assert_eq!(
+            PrefixS3Path::from(&path).path,
+            prefix_path(format!("{TENANT_ID}"))
+        );
+
+        path.timeline_id = Some(TIMELINE_ID);
+        assert_eq!(
+            PrefixS3Path::from(&path).path,
+            prefix_path(format!("{TENANT_ID}/{TIMELINE_ID}"))
+        );
+
+        path.endpoint_id = Some(ENDPOINT_ID.into());
+        assert_eq!(
+            PrefixS3Path::from(&path).path,
+            prefix_path(format!("{TENANT_ID}/{TIMELINE_ID}/{ENDPOINT_ID}"))
+        );
+    }
+}
--- a/object_storage/src/main.rs
+++ b/object_storage/src/main.rs
@@ -0,0 +1,65 @@
+//! `object_storage` is a service which provides API for uploading and downloading
+//! files. It is used by compute and control plane for accessing LFC prewarm data.
+//! This service is deployed either as a separate component or as part of compute image
+//! for large computes.
+mod app;
+use anyhow::Context;
+use tracing::info;
+use utils::logging;
+
+//see set()
+const fn max_upload_file_limit() -> usize {
+    100 * 1024 * 1024
+}
+
+#[derive(serde::Deserialize)]
+#[serde(tag = "type")]
+struct Config {
+    listen: std::net::SocketAddr,
+    pemfile: camino::Utf8PathBuf,
+    #[serde(flatten)]
+    storage_config: remote_storage::RemoteStorageConfig,
+    #[serde(default = "max_upload_file_limit")]
+    max_upload_file_limit: usize,
+}
+
+#[tokio::main]
+async fn main() -> anyhow::Result<()> {
+    logging::init(
+        logging::LogFormat::Plain,
+        logging::TracingErrorLayerEnablement::EnableWithRustLogFilter,
+        logging::Output::Stdout,
+    )?;
+
+    let config: String = std::env::args().skip(1).take(1).collect();
+    if config.is_empty() {
+        anyhow::bail!("Usage: object_storage config.json")
+    }
+    info!("Reading config from {config}");
+    let config = std::fs::read_to_string(config.clone())?;
+    let config: Config = serde_json::from_str(&config).context("parsing config")?;
+    info!("Reading pemfile from {}", config.pemfile.clone());
+    let pemfile = std::fs::read(config.pemfile.clone())?;
+    info!("Loading public key from {}", config.pemfile.clone());
+    let auth = object_storage::JwtAuth::new(&pemfile)?;
+
+    let listener = tokio::net::TcpListener::bind(config.listen).await.unwrap();
+    info!("listening on {}", listener.local_addr().unwrap());
+
+    let storage = remote_storage::GenericRemoteStorage::from_config(&config.storage_config).await?;
+    let cancel = tokio_util::sync::CancellationToken::new();
+    app::check_storage_permissions(&storage, cancel.clone()).await?;
+
+    let proxy = std::sync::Arc::new(object_storage::Storage {
+        auth,
+        storage,
+        cancel: cancel.clone(),
+        max_upload_file_limit: config.max_upload_file_limit,
+    });
+
+    tokio::spawn(utils::signals::signal_handler(cancel.clone()));
+    axum::serve(listener, app::app(proxy))
+        .with_graceful_shutdown(async move { cancel.cancelled().await })
+        .await?;
+    Ok(())
+}
--- a/pageserver/Cargo.toml
+++ b/pageserver/Cargo.toml
@@ -10,6 +10,8 @@ default = []
 # which adds some runtime cost to run tests on outage conditions
 testing = ["fail/failpoints", "pageserver_api/testing", "wal_decoder/testing", "pageserver_client/testing"]

+fuzz-read-path = ["testing"]
+
 [dependencies]
 anyhow.workspace = true
 arc-swap.workspace = true
--- a/pageserver/benches/bench_ingest.rs
+++ b/pageserver/benches/bench_ingest.rs
@@ -126,7 +126,7 @@ async fn ingest(
            max_concurrency: NonZeroUsize::new(1).unwrap(),
        });
        let (_desc, path) = layer
-            .write_to_disk(&ctx, None, l0_flush_state.inner())
+            .write_to_disk(&ctx, None, l0_flush_state.inner(), &gate, cancel.clone())
            .await?
            .unwrap();
        tokio::fs::remove_file(path).await?;
--- a/pageserver/benches/bench_walredo.rs
+++ b/pageserver/benches/bench_walredo.rs
@@ -65,7 +65,7 @@ use bytes::{Buf, Bytes};
 use criterion::{BenchmarkId, Criterion};
 use once_cell::sync::Lazy;
 use pageserver::config::PageServerConf;
-use pageserver::walredo::PostgresRedoManager;
+use pageserver::walredo::{PostgresRedoManager, RedoAttemptType};
 use pageserver_api::key::Key;
 use pageserver_api::record::NeonWalRecord;
 use pageserver_api::shard::TenantShardId;
@@ -223,7 +223,14 @@ impl Request {

        // TODO: avoid these clones
        manager
-            .request_redo(*key, *lsn, base_img.clone(), records.clone(), *pg_version)
+            .request_redo(
+                *key,
+                *lsn,
+                base_img.clone(),
+                records.clone(),
+                *pg_version,
+                RedoAttemptType::ReadPage,
+            )
            .await
            .context("request_redo")
    }
--- a/pageserver/src/basebackup.rs
+++ b/pageserver/src/basebackup.rs
@@ -34,7 +34,7 @@ use utils::lsn::Lsn;
 use crate::context::RequestContext;
 use crate::pgdatadir_mapping::Version;
 use crate::tenant::storage_layer::IoConcurrency;
-use crate::tenant::timeline::GetVectoredError;
+use crate::tenant::timeline::{GetVectoredError, VersionedKeySpaceQuery};
 use crate::tenant::{PageReconstructError, Timeline};

 #[derive(Debug, thiserror::Error)]
@@ -353,9 +353,10 @@ where
            let mut slru_builder = SlruSegmentsBuilder::new(&mut self.ar);

            for part in slru_partitions.parts {
+                let query = VersionedKeySpaceQuery::uniform(part, self.lsn);
                let blocks = self
                    .timeline
-                    .get_vectored(part, self.lsn, self.io_concurrency.clone(), self.ctx)
+                    .get_vectored(query, self.io_concurrency.clone(), self.ctx)
                    .await?;

                for (key, block) in blocks {
--- a/pageserver/src/bin/pageserver.rs
+++ b/pageserver/src/bin/pageserver.rs
@@ -16,7 +16,7 @@ use http_utils::tls_certs::ReloadingCertificateResolver;
 use metrics::launch_timestamp::{LaunchTimestamp, set_launch_timestamp_metric};
 use metrics::set_build_info_metric;
 use nix::sys::socket::{setsockopt, sockopt};
-use pageserver::config::{PageServerConf, PageserverIdentity};
+use pageserver::config::{PageServerConf, PageserverIdentity, ignored_fields};
 use pageserver::controller_upcall_client::StorageControllerUpcallClient;
 use pageserver::deletion_queue::DeletionQueue;
 use pageserver::disk_usage_eviction_task::{self, launch_disk_usage_global_eviction_task};
@@ -31,7 +31,6 @@ use pageserver::{
 };
 use postgres_backend::AuthType;
 use remote_storage::GenericRemoteStorage;
-use tokio::signal::unix::SignalKind;
 use tokio::time::Instant;
 use tokio_util::sync::CancellationToken;
 use tracing::*;
@@ -98,7 +97,7 @@ fn main() -> anyhow::Result<()> {
    env::set_current_dir(&workdir)
        .with_context(|| format!("Failed to set application's current dir to '{workdir}'"))?;

-    let conf = initialize_config(&identity_file_path, &cfg_file_path, &workdir)?;
+    let (conf, ignored) = initialize_config(&identity_file_path, &cfg_file_path, &workdir)?;

    // Initialize logging.
    //
@@ -144,7 +143,17 @@ fn main() -> anyhow::Result<()> {
        &[("node_id", &conf.id.to_string())],
    );

-    // after setting up logging, log the effective IO engine choice and read path implementations
+    // Warn about ignored config items; see pageserver_api::config::ConfigToml
+    // doc comment for rationale why we prefer this over serde(deny_unknown_fields).
+    {
+        let ignored_fields::Paths { paths } = &ignored;
+        for path in paths {
+            warn!(?path, "ignoring unknown configuration item");
+        }
+    }
+
+    // Log configuration items for feature-flag-like config
+    // (maybe we should automate this with a visitor?).
    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");
@@ -207,7 +216,7 @@ fn main() -> anyhow::Result<()> {
    tracing::info!("Initializing page_cache...");
    page_cache::init(conf.page_cache_size);

-    start_pageserver(launch_ts, conf, otel_guard).context("Failed to start pageserver")?;
+    start_pageserver(launch_ts, conf, ignored, otel_guard).context("Failed to start pageserver")?;

    scenario.teardown();
    Ok(())
@@ -217,7 +226,7 @@ fn initialize_config(
    identity_file_path: &Utf8Path,
    cfg_file_path: &Utf8Path,
    workdir: &Utf8Path,
-) -> anyhow::Result<&'static PageServerConf> {
+) -> anyhow::Result<(&'static PageServerConf, ignored_fields::Paths)> {
    // The deployment orchestrator writes out an indentity file containing the node id
    // for all pageservers. This file is the source of truth for the node id. In order
    // to allow for rolling back pageserver releases, the node id is also included in
@@ -246,16 +255,36 @@ fn initialize_config(

    let config_file_contents =
        std::fs::read_to_string(cfg_file_path).context("read config file from filesystem")?;
-    let config_toml = serde_path_to_error::deserialize(
-        toml_edit::de::Deserializer::from_str(&config_file_contents)
-            .context("build toml deserializer")?,
-    )
-    .context("deserialize config toml")?;

+    // Deserialize the config file contents into a ConfigToml.
+    let config_toml: pageserver_api::config::ConfigToml = {
+        let deserializer = toml_edit::de::Deserializer::from_str(&config_file_contents)
+            .context("build toml deserializer")?;
+        let mut path_to_error_track = serde_path_to_error::Track::new();
+        let deserializer =
+            serde_path_to_error::Deserializer::new(deserializer, &mut path_to_error_track);
+        serde::Deserialize::deserialize(deserializer).context("deserialize config toml")?
+    };
+
+    // Find unknown fields by re-serializing the parsed ConfigToml and comparing it to the on-disk file.
+    // Any fields that are only in the on-disk version are unknown.
+    // (The assumption here is that the ConfigToml doesn't to skip_serializing_if.)
+    // (Make sure to read the ConfigToml doc comment on why we only want to warn about, but not fail startup, on unknown fields).
+    let ignored = {
+        let ondisk_toml = config_file_contents
+            .parse::<toml_edit::DocumentMut>()
+            .context("parse original config as toml document")?;
+        let parsed_toml = toml_edit::ser::to_document(&config_toml)
+            .context("re-serialize config to toml document")?;
+        pageserver::config::ignored_fields::find(ondisk_toml, parsed_toml)
+    };
+
+    // Construct the runtime god object (it's called PageServerConf but actually is just global shared state).
    let conf = PageServerConf::parse_and_validate(identity.id, config_toml, workdir)
        .context("runtime-validation of config toml")?;
+    let conf = Box::leak(Box::new(conf));

-    Ok(Box::leak(Box::new(conf)))
+    Ok((conf, ignored))
 }

 struct WaitForPhaseResult<F: std::future::Future + Unpin> {
@@ -306,6 +335,7 @@ fn startup_checkpoint(started_at: Instant, phase: &str, human_phase: &str) {
 fn start_pageserver(
    launch_ts: &'static LaunchTimestamp,
    conf: &'static PageServerConf,
+    ignored: ignored_fields::Paths,
    otel_guard: Option<OtelGuard>,
 ) -> anyhow::Result<()> {
    // Monotonic time for later calculating startup duration
@@ -329,7 +359,7 @@ fn start_pageserver(
        pageserver::metrics::tokio_epoll_uring::Collector::new(),
    ))
    .unwrap();
-    pageserver::preinitialize_metrics(conf);
+    pageserver::preinitialize_metrics(conf, ignored);

    // If any failpoints were set from FAILPOINTS environment variable,
    // print them to the log for debugging purposes
@@ -422,6 +452,24 @@ fn start_pageserver(
    info!("Using auth for http API: {:#?}", conf.http_auth_type);
    info!("Using auth for pg connections: {:#?}", conf.pg_auth_type);

+    let tls_server_config = if conf.listen_https_addr.is_some() || conf.enable_tls_page_service_api
+    {
+        let resolver = BACKGROUND_RUNTIME.block_on(ReloadingCertificateResolver::new(
+            "main",
+            &conf.ssl_key_file,
+            &conf.ssl_cert_file,
+            conf.ssl_cert_reload_period,
+        ))?;
+
+        let server_config = rustls::ServerConfig::builder()
+            .with_no_client_auth()
+            .with_cert_resolver(resolver);
+
+        Some(Arc::new(server_config))
+    } else {
+        None
+    };
+
    match var("NEON_AUTH_TOKEN") {
        Ok(v) => {
            info!("Loaded JWT token for authentication with Safekeeper");
@@ -640,17 +688,11 @@ fn start_pageserver(

        let https_task = match https_listener {
            Some(https_listener) => {
-                let resolver = MGMT_REQUEST_RUNTIME.block_on(ReloadingCertificateResolver::new(
-                    &conf.ssl_key_file,
-                    &conf.ssl_cert_file,
-                    conf.ssl_cert_reload_period,
-                ))?;
+                let tls_server_config = tls_server_config
+                    .clone()
+                    .expect("tls_server_config is set earlier if https is enabled");

-                let server_config = rustls::ServerConfig::builder()
-                    .with_no_client_auth()
-                    .with_cert_resolver(resolver);
-
-                let tls_acceptor = tokio_rustls::TlsAcceptor::from(Arc::new(server_config));
+                let tls_acceptor = tokio_rustls::TlsAcceptor::from(tls_server_config);

                let server =
                    http_utils::server::Server::new(service, https_listener, Some(tls_acceptor))?;
@@ -706,6 +748,11 @@ fn start_pageserver(
            tokio::net::TcpListener::from_std(pageserver_listener)
                .context("create tokio listener")?
        },
+        if conf.enable_tls_page_service_api {
+            tls_server_config
+        } else {
+            None
+        },
    );

    // All started up! Now just sit and wait for shutdown signal.
@@ -713,32 +760,7 @@ fn start_pageserver(
        let signal_token = CancellationToken::new();
        let signal_cancel = signal_token.child_token();

-        // Spawn signal handlers. Runs in a loop since we want to be responsive to multiple signals
-        // even after triggering shutdown (e.g. a SIGQUIT after a slow SIGTERM shutdown). See:
-        // https://github.com/neondatabase/neon/issues/9740.
-        tokio::spawn(async move {
-            let mut sigint = tokio::signal::unix::signal(SignalKind::interrupt()).unwrap();
-            let mut sigterm = tokio::signal::unix::signal(SignalKind::terminate()).unwrap();
-            let mut sigquit = tokio::signal::unix::signal(SignalKind::quit()).unwrap();
-
-            loop {
-                let signal = tokio::select! {
-                    _ = sigquit.recv() => {
-                        info!("Got signal SIGQUIT. Terminating in immediate shutdown mode.");
-                        std::process::exit(111);
-                    }
-                    _ = sigint.recv() => "SIGINT",
-                    _ = sigterm.recv() => "SIGTERM",
-                };
-
-                if !signal_token.is_cancelled() {
-                    info!("Got signal {signal}. Terminating gracefully in fast shutdown mode.");
-                    signal_token.cancel();
-                } else {
-                    info!("Got signal {signal}. Already shutting down.");
-                }
-            }
-        });
+        tokio::spawn(utils::signals::signal_handler(signal_token));

        // Wait for cancellation signal and shut down the pageserver.
        //
--- a/pageserver/src/config.rs
+++ b/pageserver/src/config.rs
@@ -4,6 +4,8 @@
 //! file, or on the command line.
 //! See also `settings.md` for better description on every parameter.

+pub mod ignored_fields;
+
 use std::env;
 use std::num::NonZeroUsize;
 use std::sync::Arc;
@@ -217,6 +219,11 @@ pub struct PageServerConf {
    pub generate_unarchival_heatmap: bool,

    pub tracing: Option<pageserver_api::config::Tracing>,
+
+    /// Enable TLS in page service API.
+    /// Does not force TLS: the client negotiates TLS usage during the handshake.
+    /// Uses key and certificate from ssl_key_file/ssl_cert_file.
+    pub enable_tls_page_service_api: bool,
 }

 /// Token for authentication to safekeepers
@@ -389,6 +396,7 @@ impl PageServerConf {
            load_previous_heatmap,
            generate_unarchival_heatmap,
            tracing,
+            enable_tls_page_service_api,
        } = config_toml;

        let mut conf = PageServerConf {
@@ -439,6 +447,7 @@ impl PageServerConf {
            page_service_pipelining,
            get_vectored_concurrent_io,
            tracing,
+            enable_tls_page_service_api,

            // ------------------------------------------------------------
            // fields that require additional validation or custom handling
@@ -560,7 +569,6 @@ impl PageServerConf {
 }

 #[derive(serde::Deserialize, serde::Serialize)]
-#[serde(deny_unknown_fields)]
 pub struct PageserverIdentity {
    pub id: NodeId,
 }
@@ -632,82 +640,4 @@ mod tests {
        PageServerConf::parse_and_validate(NodeId(0), config_toml, &workdir)
            .expect("parse_and_validate");
    }
-
-    /// If there's a typo in the pageserver config, we'd rather catch that typo
-    /// and fail pageserver startup than silently ignoring the typo, leaving whoever
-    /// made it in the believe that their config change is effective.
-    ///
-    /// The default in serde is to allow unknown fields, so, we rely
-    /// on developer+review discipline to add `deny_unknown_fields` when adding
-    /// new structs to the config, and these tests here as a regression test.
-    ///
-    /// The alternative to all of this would be to allow unknown fields in the config.
-    /// To catch them, we could have a config check tool or mgmt API endpoint that
-    /// compares the effective config with the TOML on disk and makes sure that
-    /// the on-disk TOML is a strict subset of the effective config.
-    mod unknown_fields_handling {
-        macro_rules! test {
-            ($short_name:ident, $input:expr) => {
-                #[test]
-                fn $short_name() {
-                    let input = $input;
-                    let err = toml_edit::de::from_str::<pageserver_api::config::ConfigToml>(&input)
-                        .expect_err("some_invalid_field is an invalid field");
-                    dbg!(&err);
-                    assert!(err.to_string().contains("some_invalid_field"));
-                }
-            };
-        }
-        use indoc::indoc;
-
-        test!(
-            toplevel,
-            indoc! {r#"
-                some_invalid_field = 23
-            "#}
-        );
-
-        test!(
-            toplevel_nested,
-            indoc! {r#"
-                [some_invalid_field]
-                foo = 23
-            "#}
-        );
-
-        test!(
-            disk_usage_based_eviction,
-            indoc! {r#"
-                [disk_usage_based_eviction]
-                some_invalid_field = 23
-            "#}
-        );
-
-        test!(
-            tenant_config,
-            indoc! {r#"
-                [tenant_config]
-                some_invalid_field = 23
-            "#}
-        );
-
-        test!(
-            l0_flush,
-            indoc! {r#"
-                [l0_flush]
-                mode = "direct"
-                some_invalid_field = 23
-            "#}
-        );
-
-        // TODO: fix this => https://github.com/neondatabase/neon/issues/8915
-        // test!(
-        //     remote_storage_config,
-        //     indoc! {r#"
-        //         [remote_storage_config]
-        //         local_path = "/nonexistent"
-        //         some_invalid_field = 23
-        //     "#}
-        // );
-    }
 }
--- a/pageserver/src/config/ignored_fields.rs
+++ b/pageserver/src/config/ignored_fields.rs
@@ -0,0 +1,179 @@
+//! Check for fields in the on-disk config file that were ignored when
+//! deserializing [`pageserver_api::config::ConfigToml`].
+//!
+//! This could have been part of the [`pageserver_api::config`] module,
+//! but the way we identify unused fields in this module
+//! is specific to the format (TOML) and the implementation of the
+//! deserialization for that format ([`toml_edit`]).
+
+use std::collections::HashSet;
+
+use itertools::Itertools;
+
+/// Pass in the user-specified config and the re-serialized [`pageserver_api::config::ConfigToml`].
+/// The returned [`Paths`] contains the paths to the fields that were ignored by deserialization
+/// of the [`pageserver_api::config::ConfigToml`].
+pub fn find(user_specified: toml_edit::DocumentMut, reserialized: toml_edit::DocumentMut) -> Paths {
+    let user_specified = paths(user_specified);
+    let reserialized = paths(reserialized);
+    fn paths(doc: toml_edit::DocumentMut) -> HashSet<String> {
+        let mut out = Vec::new();
+        let mut visitor = PathsVisitor::new(&mut out);
+        visitor.visit_table_like(doc.as_table());
+        HashSet::from_iter(out)
+    }
+
+    let mut ignored = HashSet::new();
+
+    // O(n) because of HashSet
+    for path in user_specified {
+        if !reserialized.contains(&path) {
+            ignored.insert(path);
+        }
+    }
+
+    Paths {
+        paths: ignored
+            .into_iter()
+            // sort lexicographically for deterministic output
+            .sorted()
+            .collect(),
+    }
+}
+
+pub struct Paths {
+    pub paths: Vec<String>,
+}
+
+struct PathsVisitor<'a> {
+    stack: Vec<String>,
+    out: &'a mut Vec<String>,
+}
+
+impl<'a> PathsVisitor<'a> {
+    fn new(out: &'a mut Vec<String>) -> Self {
+        Self {
+            stack: Vec::new(),
+            out,
+        }
+    }
+
+    fn visit_table_like(&mut self, table_like: &dyn toml_edit::TableLike) {
+        for (entry, item) in table_like.iter() {
+            self.stack.push(entry.to_string());
+            self.visit_item(item);
+            self.stack.pop();
+        }
+    }
+
+    fn visit_item(&mut self, item: &toml_edit::Item) {
+        match item {
+            toml_edit::Item::None => (),
+            toml_edit::Item::Value(value) => self.visit_value(value),
+            toml_edit::Item::Table(table) => {
+                self.visit_table_like(table);
+            }
+            toml_edit::Item::ArrayOfTables(array_of_tables) => {
+                for (i, table) in array_of_tables.iter().enumerate() {
+                    self.stack.push(format!("[{i}]"));
+                    self.visit_table_like(table);
+                    self.stack.pop();
+                }
+            }
+        }
+    }
+
+    fn visit_value(&mut self, value: &toml_edit::Value) {
+        match value {
+            toml_edit::Value::String(_)
+            | toml_edit::Value::Integer(_)
+            | toml_edit::Value::Float(_)
+            | toml_edit::Value::Boolean(_)
+            | toml_edit::Value::Datetime(_) => self.out.push(self.stack.join(".")),
+            toml_edit::Value::Array(array) => {
+                for (i, value) in array.iter().enumerate() {
+                    self.stack.push(format!("[{i}]"));
+                    self.visit_value(value);
+                    self.stack.pop();
+                }
+            }
+            toml_edit::Value::InlineTable(inline_table) => self.visit_table_like(inline_table),
+        }
+    }
+}
+
+#[cfg(test)]
+pub(crate) mod tests {
+
+    fn test_impl(original: &str, parsed: &str, expect: [&str; 1]) {
+        let original: toml_edit::DocumentMut = original.parse().expect("parse original config");
+        let parsed: toml_edit::DocumentMut = parsed.parse().expect("parse re-serialized config");
+
+        let super::Paths { paths: actual } = super::find(original, parsed);
+        assert_eq!(actual, &expect);
+    }
+
+    #[test]
+    fn top_level() {
+        test_impl(
+            r#"
+                [a]
+                b = 1
+                c = 2
+                d = 3
+            "#,
+            r#"
+                [a]
+                b = 1
+                c = 2
+            "#,
+            ["a.d"],
+        );
+    }
+
+    #[test]
+    fn nested() {
+        test_impl(
+            r#"
+                [a.b.c]
+                d = 23
+            "#,
+            r#"
+                [a]
+                e = 42
+            "#,
+            ["a.b.c.d"],
+        );
+    }
+
+    #[test]
+    fn array_of_tables() {
+        test_impl(
+            r#"
+                [[a]]
+                b = 1
+                c = 2
+                d = 3
+            "#,
+            r#"
+                [[a]]
+                b = 1
+                c = 2
+            "#,
+            ["a.[0].d"],
+        );
+    }
+
+    #[test]
+    fn array() {
+        test_impl(
+            r#"
+            foo = [ {bar = 23} ]
+            "#,
+            r#"
+            foo = [ { blup = 42 }]
+            "#,
+            ["foo.[0].bar"],
+        );
+    }
+}
--- a/pageserver/src/context.rs
+++ b/pageserver/src/context.rs
@@ -89,7 +89,7 @@
 //! [`RequestContext`] argument. Functions in the middle of the call chain
 //! only need to pass it on.

-use std::sync::Arc;
+use std::{sync::Arc, time::Duration};

 use once_cell::sync::Lazy;
 use tracing::warn;
@@ -566,22 +566,37 @@ impl RequestContext {
        }
    }

-    pub(crate) fn perf_follows_from(&self, from: &RequestContext) {
-        if let (Some(span), Some(from_span)) = (&self.perf_span, &from.perf_span) {
-            span.inner().follows_from(from_span.inner());
+    pub(crate) fn ondemand_download_wait_observe(&self, duration: Duration) {
+        if duration == Duration::ZERO {
+            return;
+        }
+
+        match &self.scope {
+            Scope::Timeline { arc_arc } => arc_arc
+                .wait_ondemand_download_time
+                .observe(self.task_kind, duration),
+            _ => {
+                use once_cell::sync::Lazy;
+                use std::sync::Mutex;
+                use std::time::Duration;
+                use utils::rate_limit::RateLimit;
+                static LIMIT: Lazy<Mutex<RateLimit>> =
+                    Lazy::new(|| Mutex::new(RateLimit::new(Duration::from_secs(1))));
+                let mut guard = LIMIT.lock().unwrap();
+                guard.call2(|rate_limit_stats| {
+                    warn!(
+                        %rate_limit_stats,
+                        backtrace=%std::backtrace::Backtrace::force_capture(),
+                        "ondemand downloads should always happen within timeline scope",
+                    );
+                });
+            }
        }
    }

-    pub(crate) fn perf_span_record<
-        Q: tracing::field::AsField + ?Sized,
-        V: tracing::field::Value,
-    >(
-        &self,
-        field: &Q,
-        value: V,
-    ) {
-        if let Some(span) = &self.perf_span {
-            span.record(field, value);
+    pub(crate) fn perf_follows_from(&self, from: &RequestContext) {
+        if let (Some(span), Some(from_span)) = (&self.perf_span, &from.perf_span) {
+            span.inner().follows_from(from_span.inner());
        }
    }

--- a/pageserver/src/http/openapi_spec.yml
+++ b/pageserver/src/http/openapi_spec.yml
@@ -212,6 +212,12 @@ paths:
              schema:
                type: string
                format: date-time
+        "412":
+          description: No timestamp is found for given LSN, e.g. if there had been no commits till LSN
+          content:
+            application/json:
+              schema:
+                $ref: "#/components/schemas/PreconditionFailedError"

  /v1/tenant/{tenant_id}/timeline/{timeline_id}/get_lsn_by_timestamp:
    parameters:
--- a/pageserver/src/http/routes.rs
+++ b/pageserver/src/http/routes.rs
@@ -67,7 +67,7 @@ use crate::tenant::mgr::{
 };
 use crate::tenant::remote_timeline_client::index::GcCompactionState;
 use crate::tenant::remote_timeline_client::{
-    download_index_part, list_remote_tenant_shards, list_remote_timelines,
+    download_index_part, download_tenant_manifest, list_remote_tenant_shards, list_remote_timelines,
 };
 use crate::tenant::secondary::SecondaryController;
 use crate::tenant::size::ModelInputs;
@@ -989,7 +989,7 @@ async fn get_lsn_by_timestamp_handler(
    if !tenant_shard_id.is_shard_zero() {
        // Requires SLRU contents, which are only stored on shard zero
        return Err(ApiError::BadRequest(anyhow!(
-            "Size calculations are only available on shard zero"
+            "Lsn calculations by timestamp are only available on shard zero"
        )));
    }

@@ -1064,7 +1064,7 @@ async fn get_timestamp_of_lsn_handler(
    if !tenant_shard_id.is_shard_zero() {
        // Requires SLRU contents, which are only stored on shard zero
        return Err(ApiError::BadRequest(anyhow!(
-            "Size calculations are only available on shard zero"
+            "Timestamp calculations by lsn are only available on shard zero"
        )));
    }

@@ -1090,8 +1090,8 @@ async fn get_timestamp_of_lsn_handler(
            .to_string();
            json_response(StatusCode::OK, time)
        }
-        None => Err(ApiError::NotFound(
-            anyhow::anyhow!("Timestamp for lsn {} not found", lsn).into(),
+        None => Err(ApiError::PreconditionFailed(
+            format!("Timestamp for lsn {} not found", lsn).into(),
        )),
    }
 }
@@ -2274,6 +2274,7 @@ async fn timeline_compact_handler(
    if Some(true) == parse_query_param::<_, bool>(&request, "dry_run")? {
        flags |= CompactFlags::DryRun;
    }
+    // Manual compaction does not yield for L0.

    let wait_until_uploaded =
        parse_query_param::<_, bool>(&request, "wait_until_uploaded")?.unwrap_or(false);
@@ -2911,9 +2912,22 @@ async fn tenant_scan_remote_handler(
            };
        }

+        let result =
+            download_tenant_manifest(&state.remote_storage, &tenant_shard_id, generation, &cancel)
+                .instrument(info_span!("download_tenant_manifest",
+                            tenant_id=%tenant_shard_id.tenant_id,
+                            shard_id=%tenant_shard_id.shard_slug()))
+                .await;
+        let stripe_size = match result {
+            Ok((manifest, _, _)) => manifest.stripe_size,
+            Err(DownloadError::NotFound) => None,
+            Err(err) => return Err(ApiError::InternalServerError(anyhow!(err))),
+        };
+
        response.shards.push(TenantScanRemoteStorageShard {
            tenant_shard_id,
            generation: generation.into(),
+            stripe_size,
        });
    }

@@ -3239,7 +3253,7 @@ async fn ingest_aux_files(
        modification
            .put_file(&fname, content.as_bytes(), &ctx)
            .await
-            .map_err(ApiError::InternalServerError)?;
+            .map_err(|e| ApiError::InternalServerError(e.into()))?;
    }
    modification
        .commit(&ctx)
@@ -3368,11 +3382,11 @@ async fn put_tenant_timeline_import_basebackup(

        let broker_client = state.broker_client.clone();

-        let mut body = StreamReader::new(request.into_body().map(|res| {
-            res.map_err(|error| {
-                std::io::Error::new(std::io::ErrorKind::Other, anyhow::anyhow!(error))
-            })
-        }));
+        let mut body = StreamReader::new(
+            request
+                .into_body()
+                .map(|res| res.map_err(|error| std::io::Error::other(anyhow::anyhow!(error)))),
+        );

        tenant.wait_to_become_active(ACTIVE_TENANT_TIMEOUT).await?;

@@ -3446,7 +3460,7 @@ async fn put_tenant_timeline_import_wal(

        let mut body = StreamReader::new(request.into_body().map(|res| {
            res.map_err(|error| {
-                std::io::Error::new(std::io::ErrorKind::Other, anyhow::anyhow!(error))
+                std::io::Error::other( anyhow::anyhow!(error))
            })
        }));

--- a/pageserver/src/import_datadir.rs
+++ b/pageserver/src/import_datadir.rs
@@ -27,7 +27,7 @@ use crate::context::RequestContext;
 use crate::metrics::WAL_INGEST;
 use crate::pgdatadir_mapping::*;
 use crate::tenant::Timeline;
-use crate::walingest::WalIngest;
+use crate::walingest::{WalIngest, WalIngestErrorKind};

 // Returns checkpoint LSN from controlfile
 pub fn get_lsn_from_controlfile(path: &Utf8Path) -> Result<Lsn> {
@@ -157,9 +157,9 @@ async fn import_rel(
        .put_rel_creation(rel, nblocks as u32, ctx)
        .await
    {
-        match e {
-            RelationError::AlreadyExists => {
-                debug!("Relation {} already exist. We must be extending it.", rel)
+        match e.kind {
+            WalIngestErrorKind::RelationAlreadyExists(rel) => {
+                debug!("Relation {rel} already exists. We must be extending it.")
            }
            _ => return Err(e.into()),
        }
--- a/pageserver/src/metrics.rs
+++ b/pageserver/src/metrics.rs
@@ -1,10 +1,8 @@
 use std::collections::HashMap;
 use std::num::NonZeroUsize;
 use std::os::fd::RawFd;
-use std::pin::Pin;
 use std::sync::atomic::AtomicU64;
 use std::sync::{Arc, Mutex};
-use std::task::{Context, Poll};
 use std::time::{Duration, Instant};

 use enum_map::{Enum as _, EnumMap};
@@ -19,17 +17,17 @@ use metrics::{
 use once_cell::sync::Lazy;
 use pageserver_api::config::{
    PageServicePipeliningConfig, PageServicePipeliningConfigPipelined,
-    PageServiceProtocolPipelinedExecutionStrategy,
+    PageServiceProtocolPipelinedBatchingStrategy, PageServiceProtocolPipelinedExecutionStrategy,
 };
 use pageserver_api::models::InMemoryLayerInfo;
 use pageserver_api::shard::TenantShardId;
-use pin_project_lite::pin_project;
 use postgres_backend::{QueryError, is_expected_io_error};
 use pq_proto::framed::ConnectionError;
 use strum::{EnumCount, IntoEnumIterator as _, VariantNames};
 use strum_macros::{IntoStaticStr, VariantNames};
 use utils::id::TimelineId;

+use crate::config;
 use crate::config::PageServerConf;
 use crate::context::{PageContentKind, RequestContext};
 use crate::pgdatadir_mapping::DatadirModificationStats;
@@ -499,6 +497,100 @@ pub(crate) static WAIT_LSN_IN_PROGRESS_GLOBAL_MICROS: Lazy<IntCounter> = Lazy::n
    .expect("failed to define a metric")
 });

+pub(crate) mod wait_ondemand_download_time {
+    use super::*;
+    const WAIT_ONDEMAND_DOWNLOAD_TIME_BUCKETS: &[f64] = &[
+        0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, // 10 ms - 100ms
+        0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, // 100ms to 1s
+        1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, // 1s to 10s
+        10.0, 20.0, 30.0, 40.0, 50.0, 60.0, // 10s to 1m
+    ];
+
+    /// The task kinds for which we want to track wait times for on-demand downloads.
+    /// Other task kinds' wait times are accumulated in label value `unknown`.
+    pub(crate) const WAIT_ONDEMAND_DOWNLOAD_METRIC_TASK_KINDS: [TaskKind; 2] = [
+        TaskKind::PageRequestHandler,
+        TaskKind::WalReceiverConnectionHandler,
+    ];
+
+    pub(crate) static WAIT_ONDEMAND_DOWNLOAD_TIME_GLOBAL: Lazy<Vec<Histogram>> = Lazy::new(|| {
+        let histo = register_histogram_vec!(
+            "pageserver_wait_ondemand_download_seconds_global",
+            "Observations are individual tasks' wait times for on-demand downloads. \
+         If N tasks coalesce on an on-demand download, and it takes 10s, than we observe N * 10s.",
+            &["task_kind"],
+            WAIT_ONDEMAND_DOWNLOAD_TIME_BUCKETS.into(),
+        )
+        .expect("failed to define a metric");
+        WAIT_ONDEMAND_DOWNLOAD_METRIC_TASK_KINDS
+            .iter()
+            .map(|task_kind| histo.with_label_values(&[task_kind.into()]))
+            .collect::<Vec<_>>()
+    });
+
+    pub(crate) static WAIT_ONDEMAND_DOWNLOAD_TIME_SUM: Lazy<CounterVec> = Lazy::new(|| {
+        register_counter_vec!(
+            // use a name that _could_ be evolved into a per-timeline histogram later
+            "pageserver_wait_ondemand_download_seconds_sum",
+            "Like `pageserver_wait_ondemand_download_seconds_global` but per timeline",
+            &["tenant_id", "shard_id", "timeline_id", "task_kind"],
+        )
+        .unwrap()
+    });
+
+    pub struct WaitOndemandDownloadTimeSum {
+        counters: [Counter; WAIT_ONDEMAND_DOWNLOAD_METRIC_TASK_KINDS.len()],
+    }
+
+    impl WaitOndemandDownloadTimeSum {
+        pub(crate) fn new(tenant_id: &str, shard_id: &str, timeline_id: &str) -> Self {
+            let counters = WAIT_ONDEMAND_DOWNLOAD_METRIC_TASK_KINDS
+                .iter()
+                .map(|task_kind| {
+                    WAIT_ONDEMAND_DOWNLOAD_TIME_SUM
+                        .get_metric_with_label_values(&[
+                            tenant_id,
+                            shard_id,
+                            timeline_id,
+                            task_kind.into(),
+                        ])
+                        .unwrap()
+                })
+                .collect::<Vec<_>>();
+            Self {
+                counters: counters.try_into().unwrap(),
+            }
+        }
+        pub(crate) fn observe(&self, task_kind: TaskKind, duration: Duration) {
+            let maybe = WAIT_ONDEMAND_DOWNLOAD_METRIC_TASK_KINDS
+                .iter()
+                .enumerate()
+                .find(|(_, kind)| **kind == task_kind);
+            let Some((idx, _)) = maybe else {
+                return;
+            };
+            WAIT_ONDEMAND_DOWNLOAD_TIME_GLOBAL[idx].observe(duration.as_secs_f64());
+            let counter = &self.counters[idx];
+            counter.inc_by(duration.as_secs_f64());
+        }
+    }
+
+    pub(crate) fn shutdown_timeline(tenant_id: &str, shard_id: &str, timeline_id: &str) {
+        for task_kind in WAIT_ONDEMAND_DOWNLOAD_METRIC_TASK_KINDS {
+            let _ = WAIT_ONDEMAND_DOWNLOAD_TIME_SUM.remove_label_values(&[
+                tenant_id,
+                shard_id,
+                timeline_id,
+                task_kind.into(),
+            ]);
+        }
+    }
+
+    pub(crate) fn preinitialize_global_metrics() {
+        Lazy::force(&WAIT_ONDEMAND_DOWNLOAD_TIME_GLOBAL);
+    }
+}
+
 static LAST_RECORD_LSN: Lazy<IntGaugeVec> = Lazy::new(|| {
    register_int_gauge_vec!(
        "pageserver_last_record_lsn",
@@ -1248,13 +1340,13 @@ pub(crate) static STORAGE_IO_TIME_METRIC: Lazy<StorageIoTime> = Lazy::new(Storag

 #[derive(Clone, Copy)]
 #[repr(usize)]
-enum StorageIoSizeOperation {
+pub(crate) enum StorageIoSizeOperation {
    Read,
    Write,
 }

 impl StorageIoSizeOperation {
-    const VARIANTS: &'static [&'static str] = &["read", "write"];
+    pub(crate) const VARIANTS: &'static [&'static str] = &["read", "write"];

    fn as_str(&self) -> &'static str {
        Self::VARIANTS[*self as usize]
@@ -1262,7 +1354,7 @@ impl StorageIoSizeOperation {
 }

 // Needed for the https://neonprod.grafana.net/d/5uK9tHL4k/picking-tenant-for-relocation?orgId=1
-static STORAGE_IO_SIZE: Lazy<UIntGaugeVec> = Lazy::new(|| {
+pub(crate) static STORAGE_IO_SIZE: Lazy<UIntGaugeVec> = Lazy::new(|| {
    register_uint_gauge_vec!(
        "pageserver_io_operations_bytes_total",
        "Total amount of bytes read/written in IO operations",
@@ -1622,6 +1714,28 @@ pub enum SmgrQueryType {
    Test,
 }

+#[derive(
+    Debug,
+    Clone,
+    Copy,
+    IntoStaticStr,
+    strum_macros::EnumCount,
+    strum_macros::EnumIter,
+    strum_macros::FromRepr,
+    enum_map::Enum,
+)]
+#[strum(serialize_all = "snake_case")]
+pub enum GetPageBatchBreakReason {
+    BatchFull,
+    NonBatchableRequest,
+    NonUniformLsn,
+    SamePageAtDifferentLsn,
+    NonUniformTimeline,
+    ExecutorSteal,
+    #[cfg(feature = "testing")]
+    NonUniformKey,
+}
+
 pub(crate) struct SmgrQueryTimePerTimeline {
    global_started: [IntCounter; SmgrQueryType::COUNT],
    global_latency: [Histogram; SmgrQueryType::COUNT],
@@ -1633,6 +1747,8 @@ pub(crate) struct SmgrQueryTimePerTimeline {
    per_timeline_flush_in_progress_micros: IntCounter,
    global_batch_wait_time: Histogram,
    per_timeline_batch_wait_time: Histogram,
+    global_batch_break_reason: [IntCounter; GetPageBatchBreakReason::COUNT],
+    per_timeline_batch_break_reason: GetPageBatchBreakReasonTimelineMetrics,
    throttling: Arc<tenant_throttling::Pagestream>,
 }

@@ -1766,12 +1882,55 @@ static PAGE_SERVICE_BATCH_SIZE_PER_TENANT_TIMELINE: Lazy<HistogramVec> = Lazy::n
    .expect("failed to define a metric")
 });

+static PAGE_SERVICE_BATCH_BREAK_REASON_GLOBAL: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        // it's a counter, but, name is prepared to extend it to a histogram of queue depth
+        "pageserver_page_service_batch_break_reason_global",
+        "Reason for breaking batches of get page requests",
+        &["reason"],
+    )
+    .expect("failed to define a metric")
+});
+
+struct GetPageBatchBreakReasonTimelineMetrics {
+    map: EnumMap<GetPageBatchBreakReason, IntCounter>,
+}
+
+impl GetPageBatchBreakReasonTimelineMetrics {
+    fn new(tenant_id: &str, shard_slug: &str, timeline_id: &str) -> Self {
+        GetPageBatchBreakReasonTimelineMetrics {
+            map: EnumMap::from_array(std::array::from_fn(|reason_idx| {
+                let reason = GetPageBatchBreakReason::from_usize(reason_idx);
+                PAGE_SERVICE_BATCH_BREAK_REASON_PER_TENANT_TIMELINE.with_label_values(&[
+                    tenant_id,
+                    shard_slug,
+                    timeline_id,
+                    reason.into(),
+                ])
+            })),
+        }
+    }
+
+    fn inc(&self, reason: GetPageBatchBreakReason) {
+        self.map[reason].inc()
+    }
+}
+
+static PAGE_SERVICE_BATCH_BREAK_REASON_PER_TENANT_TIMELINE: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "pageserver_page_service_batch_break_reason",
+        "Reason for breaking batches of get page requests",
+        &["tenant_id", "shard_id", "timeline_id", "reason"],
+    )
+    .expect("failed to define a metric")
+});
+
 pub(crate) static PAGE_SERVICE_CONFIG_MAX_BATCH_SIZE: Lazy<IntGaugeVec> = Lazy::new(|| {
    register_int_gauge_vec!(
        "pageserver_page_service_config_max_batch_size",
        "Configured maximum batch size for the server-side batching functionality of page_service. \
         Labels expose more of the configuration parameters.",
-        &["mode", "execution"]
+        &["mode", "execution", "batching"]
    )
    .expect("failed to define a metric")
 });
@@ -1779,10 +1938,11 @@ pub(crate) static PAGE_SERVICE_CONFIG_MAX_BATCH_SIZE: Lazy<IntGaugeVec> = Lazy::
 fn set_page_service_config_max_batch_size(conf: &PageServicePipeliningConfig) {
    PAGE_SERVICE_CONFIG_MAX_BATCH_SIZE.reset();
    let (label_values, value) = match conf {
-        PageServicePipeliningConfig::Serial => (["serial", "-"], 1),
+        PageServicePipeliningConfig::Serial => (["serial", "-", "-"], 1),
        PageServicePipeliningConfig::Pipelined(PageServicePipeliningConfigPipelined {
            max_batch_size,
            execution,
+            batching,
        }) => {
            let mode = "pipelined";
            let execution = match execution {
@@ -1791,7 +1951,12 @@ fn set_page_service_config_max_batch_size(conf: &PageServicePipeliningConfig) {
                }
                PageServiceProtocolPipelinedExecutionStrategy::Tasks => "tasks",
            };
-            ([mode, execution], max_batch_size.get())
+            let batching = match batching {
+                PageServiceProtocolPipelinedBatchingStrategy::UniformLsn => "uniform-lsn",
+                PageServiceProtocolPipelinedBatchingStrategy::ScatteredLsn => "scattered-lsn",
+            };
+
+            ([mode, execution, batching], max_batch_size.get())
        }
    };
    PAGE_SERVICE_CONFIG_MAX_BATCH_SIZE
@@ -1887,6 +2052,15 @@ impl SmgrQueryTimePerTimeline {
            .get_metric_with_label_values(&[&tenant_id, &shard_slug, &timeline_id])
            .unwrap();

+        let global_batch_break_reason = std::array::from_fn(|i| {
+            let reason = GetPageBatchBreakReason::from_usize(i);
+            PAGE_SERVICE_BATCH_BREAK_REASON_GLOBAL
+                .get_metric_with_label_values(&[reason.into()])
+                .unwrap()
+        });
+        let per_timeline_batch_break_reason =
+            GetPageBatchBreakReasonTimelineMetrics::new(&tenant_id, &shard_slug, &timeline_id);
+
        let global_flush_in_progress_micros =
            PAGE_SERVICE_SMGR_FLUSH_INPROGRESS_MICROS_GLOBAL.clone();
        let per_timeline_flush_in_progress_micros = PAGE_SERVICE_SMGR_FLUSH_INPROGRESS_MICROS
@@ -1904,6 +2078,8 @@ impl SmgrQueryTimePerTimeline {
            per_timeline_flush_in_progress_micros,
            global_batch_wait_time,
            per_timeline_batch_wait_time,
+            global_batch_break_reason,
+            per_timeline_batch_break_reason,
            throttling: pagestream_throttle_metrics,
        }
    }
@@ -1932,9 +2108,16 @@ impl SmgrQueryTimePerTimeline {
    }

    /// TODO: do something about this? seems odd, we have a similar call on SmgrOpTimer
-    pub(crate) fn observe_getpage_batch_start(&self, batch_size: usize) {
+    pub(crate) fn observe_getpage_batch_start(
+        &self,
+        batch_size: usize,
+        break_reason: GetPageBatchBreakReason,
+    ) {
        self.global_batch_size.observe(batch_size as f64);
        self.per_timeline_batch_size.observe(batch_size as f64);
+
+        self.global_batch_break_reason[break_reason.into_usize()].inc();
+        self.per_timeline_batch_break_reason.inc(break_reason);
    }
 }

@@ -2314,13 +2497,18 @@ impl RemoteOpFileKind {
    }
 }

-pub(crate) static REMOTE_OPERATION_TIME: Lazy<HistogramVec> = Lazy::new(|| {
+pub(crate) static REMOTE_TIMELINE_CLIENT_COMPLETION_LATENCY: Lazy<HistogramVec> = Lazy::new(|| {
    register_histogram_vec!(
-        "pageserver_remote_operation_seconds",
-        "Time spent on remote storage operations. \
-        Grouped by tenant, timeline, operation_kind and status. \
+        "pageserver_remote_timeline_client_seconds_global",
+        "Time spent on remote timeline client operations. \
+        Grouped by task_kind, file_kind, operation_kind and status. \
+        The task_kind is \
+          - for layer downloads, populated from RequestContext (primary objective of having the label) \
+          - for index downloads, set to 'unknown' \
+          - for any upload operation, set to 'RemoteUploadTask' \
+        This keeps dimensionality at bay. \
        Does not account for time spent waiting in remote timeline client's queues.",
-        &["file_kind", "op_kind", "status"]
+        &["task_kind", "file_kind", "op_kind", "status"]
    )
    .expect("failed to define a metric")
 });
@@ -2882,6 +3070,7 @@ pub(crate) struct TimelineMetrics {
    pub storage_io_size: StorageIoSizeMetrics,
    pub wait_lsn_in_progress_micros: GlobalAndPerTenantIntCounter,
    pub wait_lsn_start_finish_counterpair: IntCounterPair,
+    pub wait_ondemand_download_time: wait_ondemand_download_time::WaitOndemandDownloadTimeSum,
    shutdown: std::sync::atomic::AtomicBool,
 }

@@ -3027,6 +3216,13 @@ impl TimelineMetrics {
            .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
            .unwrap();

+        let wait_ondemand_download_time =
+            wait_ondemand_download_time::WaitOndemandDownloadTimeSum::new(
+                &tenant_id,
+                &shard_id,
+                &timeline_id,
+            );
+
        TimelineMetrics {
            tenant_id,
            shard_id,
@@ -3060,6 +3256,7 @@ impl TimelineMetrics {
            wal_records_received,
            wait_lsn_in_progress_micros,
            wait_lsn_start_finish_counterpair,
+            wait_ondemand_download_time,
            shutdown: std::sync::atomic::AtomicBool::default(),
        }
    }
@@ -3252,6 +3449,8 @@ impl TimelineMetrics {
                .remove_label_values(&mut res, &[tenant_id, shard_id, timeline_id]);
        }

+        wait_ondemand_download_time::shutdown_timeline(tenant_id, shard_id, timeline_id);
+
        let _ = SMGR_QUERY_STARTED_PER_TENANT_TIMELINE.remove_label_values(&[
            SmgrQueryType::GetPageAtLsn.into(),
            tenant_id,
@@ -3284,6 +3483,15 @@ impl TimelineMetrics {
            shard_id,
            timeline_id,
        ]);
+
+        for reason in GetPageBatchBreakReason::iter() {
+            let _ = PAGE_SERVICE_BATCH_BREAK_REASON_PER_TENANT_TIMELINE.remove_label_values(&[
+                tenant_id,
+                shard_id,
+                timeline_id,
+                reason.into(),
+            ]);
+        }
    }
 }

@@ -3373,13 +3581,18 @@ impl RemoteTimelineClientMetrics {

    pub fn remote_operation_time(
        &self,
+        task_kind: Option<TaskKind>,
        file_kind: &RemoteOpFileKind,
        op_kind: &RemoteOpKind,
        status: &'static str,
    ) -> Histogram {
-        let key = (file_kind.as_str(), op_kind.as_str(), status);
-        REMOTE_OPERATION_TIME
-            .get_metric_with_label_values(&[key.0, key.1, key.2])
+        REMOTE_TIMELINE_CLIENT_COMPLETION_LATENCY
+            .get_metric_with_label_values(&[
+                task_kind.as_ref().map(|tk| tk.into()).unwrap_or("unknown"),
+                file_kind.as_str(),
+                op_kind.as_str(),
+                status,
+            ])
            .unwrap()
    }

@@ -3624,54 +3837,26 @@ impl Drop for RemoteTimelineClientMetrics {

 /// Wrapper future that measures the time spent by a remote storage operation,
 /// and records the time and success/failure as a prometheus metric.
-pub(crate) trait MeasureRemoteOp: Sized {
-    fn measure_remote_op(
+pub(crate) trait MeasureRemoteOp<O, E>: Sized + Future<Output = Result<O, E>> {
+    async fn measure_remote_op(
        self,
+        task_kind: Option<TaskKind>, // not all caller contexts have a RequestContext / TaskKind handy
        file_kind: RemoteOpFileKind,
        op: RemoteOpKind,
        metrics: Arc<RemoteTimelineClientMetrics>,
-    ) -> MeasuredRemoteOp<Self> {
+    ) -> Result<O, E> {
        let start = Instant::now();
-        MeasuredRemoteOp {
-            inner: self,
-            file_kind,
-            op,
-            start,
-            metrics,
-        }
+        let res = self.await;
+        let duration = start.elapsed();
+        let status = if res.is_ok() { &"success" } else { &"failure" };
+        metrics
+            .remote_operation_time(task_kind, &file_kind, &op, status)
+            .observe(duration.as_secs_f64());
+        res
    }
 }

-impl<T: Sized> MeasureRemoteOp for T {}
-
-pin_project! {
-    pub(crate) struct MeasuredRemoteOp<F>
-    {
-        #[pin]
-        inner: F,
-        file_kind: RemoteOpFileKind,
-        op: RemoteOpKind,
-        start: Instant,
-        metrics: Arc<RemoteTimelineClientMetrics>,
-    }
-}
-
-impl<F: Future<Output = Result<O, E>>, O, E> Future for MeasuredRemoteOp<F> {
-    type Output = Result<O, E>;
-
-    fn poll(self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<Self::Output> {
-        let this = self.project();
-        let poll_result = this.inner.poll(cx);
-        if let Poll::Ready(ref res) = poll_result {
-            let duration = this.start.elapsed();
-            let status = if res.is_ok() { &"success" } else { &"failure" };
-            this.metrics
-                .remote_operation_time(this.file_kind, this.op, status)
-                .observe(duration.as_secs_f64());
-        }
-        poll_result
-    }
-}
+impl<Fut, O, E> MeasureRemoteOp<O, E> for Fut where Fut: Sized + Future<Output = Result<O, E>> {}

 pub mod tokio_epoll_uring {
    use std::collections::HashMap;
@@ -4107,9 +4292,33 @@ pub(crate) fn set_tokio_runtime_setup(setup: &str, num_threads: NonZeroUsize) {
        .set(u64::try_from(num_threads.get()).unwrap());
 }

-pub fn preinitialize_metrics(conf: &'static PageServerConf) {
+static PAGESERVER_CONFIG_IGNORED_ITEMS: Lazy<UIntGaugeVec> = Lazy::new(|| {
+    register_uint_gauge_vec!(
+        "pageserver_config_ignored_items",
+        "TOML items present in the on-disk configuration file but ignored by the pageserver config parser.\
+         The `item` label is the dot-separated path of the ignored item in the on-disk configuration file.\
+         The value for an unknown config item is always 1.\
+         There is a special label value \"\", which is 0, so that there is always a metric exposed (simplifies dashboards).",
+        &["item"]
+    )
+    .unwrap()
+});
+
+pub fn preinitialize_metrics(
+    conf: &'static PageServerConf,
+    ignored: config::ignored_fields::Paths,
+) {
    set_page_service_config_max_batch_size(&conf.page_service_pipelining);

+    PAGESERVER_CONFIG_IGNORED_ITEMS
+        .with_label_values(&[""])
+        .set(0);
+    for path in &ignored.paths {
+        PAGESERVER_CONFIG_IGNORED_ITEMS
+            .with_label_values(&[path])
+            .set(1);
+    }
+
    // Python tests need these and on some we do alerting.
    //
    // FIXME(4813): make it so that we have no top level metrics as this fn will easily fall out of
@@ -4161,6 +4370,7 @@ pub fn preinitialize_metrics(conf: &'static PageServerConf) {
    [
        &BACKGROUND_LOOP_PERIOD_OVERRUN_COUNT,
        &SMGR_QUERY_STARTED_GLOBAL,
+        &PAGE_SERVICE_BATCH_BREAK_REASON_GLOBAL,
    ]
    .into_iter()
    .for_each(|c| {
@@ -4195,4 +4405,5 @@ pub fn preinitialize_metrics(conf: &'static PageServerConf) {
    Lazy::force(&tokio_epoll_uring::THREAD_LOCAL_METRICS_STORAGE);

    tenant_throttling::preinitialize_global_metrics();
+    wait_ondemand_download_time::preinitialize_global_metrics();
 }
--- a/Show More
+++ b/Show More