Add block hash to primary keys in transactions, receipts and logs tables (#100)

* Add block hash to primary keys in transactions, receipts and logs tables

* Add block hash in postgraphile triggers for transactions, receipts and logs tables

* Make indexes on transaction cid and mh_key non-unique

.github/workflows/on-main.yaml

@@ -1,25 +0,0 @@
-name: Docker Compose Build
-
-on:
-  push:
-    branches:
-      - main
-
-jobs:
-  build:
-    name: Run docker build
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v2
-      - name: Get the version
-        id: vars
-        run: echo ::set-output name=sha::$(echo ${GITHUB_SHA:0:7})
-      - name: Run docker build
-        run: make docker-build
-      - name: Tag docker image
-        run: docker tag vulcanize/ipld-eth-db docker.pkg.github.com/vulcanize/ipld-eth-db/ipld-eth-db:${{steps.vars.outputs.sha}}
-      - name: Docker Login
-        run: echo ${{ secrets.GITHUB_TOKEN }} | docker login https://docker.pkg.github.com -u vulcanize --password-stdin
-      - name: Docker Push
-        run: docker push docker.pkg.github.com/vulcanize/ipld-eth-db/ipld-eth-db:${{steps.vars.outputs.sha}}
-

.github/workflows/on-pr.yaml

@@ -3,6 +3,21 @@ name: Docker Build
 on: [pull_request]
 
 jobs:
+  build:
+    name: Run docker build
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - name: Run docker build
+        run: make docker-build
+  test:
+    # Add a dummy job to avoid failing GitHub CI checks.
+    # Other checks to be added later.
+    name: Dummy job
+    runs-on: ubuntu-latest
+    steps:
+      - name: Run dummy job
+        run: echo "Empty dummy job"
   # concise_migration_diff:
   #   name: Verify concise migration and generated schema
   #   runs-on: ubuntu-latest

.github/workflows/publish.yaml

@@ -1,11 +1,28 @@
 name: Publish Docker image
 on:
   release:
-    types: [published]
+    types: [published, edited]
 jobs:
+  build:
+    name: Run docker build
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - name: Get the version
+        id: vars
+        run: echo ::set-output name=sha::$(echo ${GITHUB_SHA:0:7})
+      - name: Run docker build
+        run: make docker-build
+      - name: Tag docker image
+        run: docker tag vulcanize/ipld-eth-db docker.pkg.github.com/vulcanize/ipld-eth-db/ipld-eth-db:${{steps.vars.outputs.sha}}
+      - name: Docker Login
+        run: echo ${{ secrets.GITHUB_TOKEN }} | docker login https://docker.pkg.github.com -u vulcanize --password-stdin
+      - name: Docker Push
+        run: docker push docker.pkg.github.com/vulcanize/ipld-eth-db/ipld-eth-db:${{steps.vars.outputs.sha}}
   push_to_registries:
     name: Push Docker image to Docker Hub
     runs-on: ubuntu-latest
+    needs: build
     steps:
       - name: Get the version
         id: vars
@@ -21,4 +38,4 @@ jobs:
       - name: Tag docker image
         run: docker tag docker.pkg.github.com/vulcanize/ipld-eth-db/ipld-eth-db:${{steps.vars.outputs.sha}} vulcanize/ipld-eth-db:${{steps.vars.outputs.tag}}
       - name: Docker Push to Docker Hub
-        run: docker push vulcanize/ipld-eth-db:${{steps.vars.outputs.tag}}
+        run: docker push vulcanize/ipld-eth-db:${{steps.vars.outputs.tag}}

.gitignore

@@ -1 +1,2 @@
-.idea/
+.idea/
+.vscode

Dockerfile

@@ -1,4 +1,4 @@
-FROM golang:1.16-alpine as builder
+FROM golang:1.18-alpine as builder
 
 RUN apk --update --no-cache add make git g++ linux-headers
 
@@ -17,11 +17,6 @@ WORKDIR /app
 
 COPY --from=builder /go/src/github.com/vulcanize/ipld-eth-db/scripts/startup_script.sh .
 
-# copy over files for multi-node setup
-COPY --from=builder /go/src/github.com/vulcanize/ipld-eth-db/docker-compose.test.yml docker-multi-node/docker-compose.test.yml
-COPY --from=builder /go/src/github.com/vulcanize/ipld-eth-db/scripts/init-access-node.sh docker-multi-node/scripts/init-access-node.sh
-COPY --from=builder /go/src/github.com/vulcanize/ipld-eth-db/scripts/init-data-node.sh docker-multi-node/scripts/init-data-node.sh
-
 COPY --from=builder /go/src/github.com/pressly/goose/cmd/goose/goose goose
 COPY --from=builder /go/src/github.com/vulcanize/ipld-eth-db/db/migrations migrations/vulcanizedb
 

Makefile

@@ -8,7 +8,7 @@ BIN = $(GOPATH)/bin
 ## Migration tool
 GOOSE = $(BIN)/goose
 $(BIN)/goose:
-	GO111MODULE=off go get -u github.com/pressly/goose/cmd/goose
+	go get -u github.com/pressly/goose/cmd/goose
 
 .PHONY: installtools
 installtools: | $(GOOSE)
@@ -115,7 +115,7 @@ import:
 ## Build docker image with schema
 .PHONY: docker-build
 docker-build:
-	docker-compose build
+	docker-compose -f docker-compose.test.yml build
 
 ## Build docker image for migration
 .PHONY: docker-concise-migration-build

README.md

@@ -12,34 +12,24 @@ Schemas and utils for IPLD ETH Postgres database
   docker-compose down -v --remove-orphans
   ```
 
-* Spin up an access node and three data nodes using [docker-compose.test.yml](./docker-compose.test.yml):
+* Spin up `ipld-eth-db` using an existing image:
 
-  ```bash
+  * Update image source used for running the migrations in [docker-compose.yml](./docker-compose.yml) (if required).
-  docker-compose -f docker-compose.test.yml up
-  ```
 
-  Following final output should be seen on all the nodes:
+  * Run:
 
     ```
-    LOG:  TimescaleDB background worker launcher connected to shared catalogs
+    docker-compose -f docker-compose.yml up
     ```
 
-* Edit [startup_script.sh](./scripts/startup_script.sh) to change the number of migrations to be run:
+* Spin up `ipld-eth-db` using a locally built image:
 
-  ```bash
+  * Update [Dockerfile](./Dockerfile) (if required).
-  ./goose -dir migrations/vulcanizedb postgres "$VDB_PG_CONNECT" up-to 21
-  ```
 
-* In another `ipld-eth-db` terminal window, build an image `migrations-test` using [Dockerfile](./db/Dockerfile):
+  * Update build context used for running the migrations in [docker-compose.test.yml](./docker-compose.test.yml) (if required).
 
-  ```bash
+  * Run:
-  docker build -t migrations-test -f ./db/Dockerfile .
-  ```
 
-* After the access and data nodes have spun up, start a container using `migrations-test` image to run the db migrations:
+    ```
-
+    docker-compose -f docker-compose.test.yml up --build
-  ```bash
+    ```
-  # Here, we are running the container using host network.
-  # So connect to access node on 127.0.0.1:8066
-  docker run --rm --network host -e DATABASE_USER=postgres -e DATABASE_PASSWORD=password -e DATABASE_HOSTNAME=127.0.0.1 -e DATABASE_PORT=8066 -e DATABASE_NAME=vulcanize_testing_v4 migrations-test
-  ```

db/Dockerfile

@@ -1,4 +1,4 @@
-FROM golang:1.16-alpine as builder
+FROM golang:1.18-alpine as builder
 
 RUN apk --update --no-cache add make git g++ linux-headers
 

db/migrations/00006_create_eth_transaction_cids_table.sql

@@ -11,7 +11,7 @@ CREATE TABLE IF NOT EXISTS eth.transaction_cids (
     tx_data               BYTEA,
     tx_type               INTEGER,
     value                 NUMERIC,
-    PRIMARY KEY (tx_hash, block_number)
+    PRIMARY KEY (tx_hash, header_id, block_number)
 );
 
 -- +goose Down

db/migrations/00007_create_eth_receipt_cids_table.sql

@@ -1,6 +1,7 @@
 -- +goose Up
 CREATE TABLE IF NOT EXISTS eth.receipt_cids (
     block_number          BIGINT NOT NULL,
+    header_id             VARCHAR(66) NOT NULL,
     tx_id                 VARCHAR(66) NOT NULL,
     leaf_cid              TEXT NOT NULL,
     contract              VARCHAR(66),
@@ -9,7 +10,7 @@ CREATE TABLE IF NOT EXISTS eth.receipt_cids (
     post_state            VARCHAR(66),
     post_status           INTEGER,
     log_root              VARCHAR(66),
-    PRIMARY KEY (tx_id, block_number)
+    PRIMARY KEY (tx_id, header_id, block_number)
 );
 
 -- +goose Down

db/migrations/00012_create_eth_log_cids_table.sql

@@ -1,6 +1,7 @@
 -- +goose Up
 CREATE TABLE IF NOT EXISTS eth.log_cids (
     block_number        BIGINT NOT NULL,
+    header_id           VARCHAR(66) NOT NULL,
     leaf_cid            TEXT NOT NULL,
     leaf_mh_key         TEXT NOT NULL,
     rct_id              VARCHAR(66) NOT NULL,
@@ -11,7 +12,7 @@ CREATE TABLE IF NOT EXISTS eth.log_cids (
     topic2              VARCHAR(66),
     topic3              VARCHAR(66),
     log_data            BYTEA,
-    PRIMARY KEY (rct_id, index, block_number)
+    PRIMARY KEY (rct_id, index, header_id, block_number)
 );
 
 -- +goose Down

db/migrations/00014_create_cid_indexes.sql

@@ -14,13 +14,14 @@ CREATE INDEX uncle_header_id_index ON eth.uncle_cids USING btree (header_id);
 -- transaction indexes
 CREATE INDEX tx_block_number_index ON eth.transaction_cids USING brin (block_number);
 CREATE INDEX tx_header_id_index ON eth.transaction_cids USING btree (header_id);
-CREATE UNIQUE INDEX tx_cid_index ON eth.transaction_cids USING btree (cid, block_number);
+CREATE INDEX tx_cid_index ON eth.transaction_cids USING btree (cid, block_number);
-CREATE UNIQUE INDEX tx_mh_block_number_index ON eth.transaction_cids USING btree (mh_key, block_number);
+CREATE INDEX tx_mh_block_number_index ON eth.transaction_cids USING btree (mh_key, block_number);
 CREATE INDEX tx_dst_index ON eth.transaction_cids USING btree (dst);
 CREATE INDEX tx_src_index ON eth.transaction_cids USING btree (src);
 
 -- receipt indexes
 CREATE INDEX rct_block_number_index ON eth.receipt_cids USING brin (block_number);
+CREATE INDEX rct_header_id_index ON eth.receipt_cids USING btree (header_id);
 CREATE INDEX rct_leaf_cid_index ON eth.receipt_cids USING btree (leaf_cid);
 CREATE INDEX rct_leaf_mh_block_number_index ON eth.receipt_cids USING btree (leaf_mh_key, block_number);
 CREATE INDEX rct_contract_index ON eth.receipt_cids USING btree (contract);
@@ -55,6 +56,7 @@ CREATE INDEX access_list_storage_keys_index ON eth.access_list_elements USING gi
 
 -- log indexes
 CREATE INDEX log_block_number_index ON eth.log_cids USING brin (block_number);
+CREATE INDEX log_header_id_index ON eth.log_cids USING btree (header_id);
 CREATE INDEX log_leaf_mh_block_number_index ON eth.log_cids USING btree (leaf_mh_key, block_number);
 CREATE INDEX log_cid_index ON  eth.log_cids USING btree (leaf_cid);
 CREATE INDEX log_address_index ON eth.log_cids USING btree (address);
@@ -72,6 +74,7 @@ DROP INDEX eth.log_topic0_index;
 DROP INDEX eth.log_address_index;
 DROP INDEX eth.log_cid_index;
 DROP INDEX eth.log_leaf_mh_block_number_index;
+DROP INDEX eth.log_header_id_index;
 DROP INDEX eth.log_block_number_index;
 
 -- access list indexes
@@ -106,6 +109,7 @@ DROP INDEX eth.rct_contract_hash_index;
 DROP INDEX eth.rct_contract_index;
 DROP INDEX eth.rct_leaf_mh_block_number_index;
 DROP INDEX eth.rct_leaf_cid_index;
+DROP INDEX eth.rct_header_id_index;
 DROP INDEX eth.rct_block_number_index;
 
 -- transaction indexes

db/migrations/00015_create_postgraphile_triggers.sql

@@ -22,17 +22,20 @@ BEGIN
     ELSIF (TG_TABLE_NAME = 'log_cids') THEN
          obj := json_build_array(
                     TG_TABLE_NAME,
+                    NEW.header_id,
                     NEW.rct_id,
                     NEW.index
                 );
     ELSIF (TG_TABLE_NAME = 'receipt_cids') THEN
          obj := json_build_array(
                     TG_TABLE_NAME,
+                    NEW.header_id,
                     NEW.tx_id
                 );
     ELSIF (TG_TABLE_NAME = 'transaction_cids') THEN
          obj := json_build_array(
                     TG_TABLE_NAME,
+                    NEW.header_id,
                     NEW.tx_hash
                 );
     ELSIF (TG_TABLE_NAME = 'access_list_elements') THEN

db/migrations/00017_create_eth_meta_schema.sql

@@ -0,0 +1,5 @@
+-- +goose Up
+CREATE SCHEMA eth_meta;
+
+-- +goose Down
+DROP SCHEMA eth_meta;

db/migrations/00018_add_data_nodes.sql

@@ -1,16 +0,0 @@
--- +goose NO TRANSACTION
--- +goose Up
--- this is generated by a script
-SELECT add_data_node('dn3', host => 'data-node-3', port => 5432, database => 'vulcanize_testing_v4', password => 'password');
-SELECT add_data_node('dn2', host => 'data-node-2', port => 5432, database => 'vulcanize_testing_v4', password => 'password');
-SELECT add_data_node('dn1', host => 'data-node-1', port => 5432, database => 'vulcanize_testing_v4', password => 'password');
-
-CALL distributed_exec($$ CREATE SCHEMA eth $$);
-
--- +goose Down
--- this is generated by a script
-CALL distributed_exec($$ DROP SCHEMA eth $$);
-
-SELECT delete_data_node('dn1', force => true, if_attached => true);
-SELECT delete_data_node('dn2', force => true, if_attached => true);
-SELECT delete_data_node('dn3', force => true, if_attached => true);

db/migrations/00018_create_watched_addresses_table.sql

@@ -0,0 +1,10 @@
+-- +goose Up
+CREATE TABLE eth_meta.watched_addresses (
+    address                 VARCHAR(66) PRIMARY KEY,
+    created_at              BIGINT NOT NULL,
+    watched_at              BIGINT NOT NULL,
+    last_filled_at          BIGINT NOT NULL DEFAULT 0
+);
+
+-- +goose Down
+DROP TABLE eth_meta.watched_addresses;

db/migrations/00019_convert_to_distributed_hypertables.sql

@@ -1,130 +0,0 @@
--- +goose Up
--- creating distributed hypertables from duplicate tables for now as we are getting the following error (while running geth
--- unit tests) if regular tables are directly converted to distributed hypertables
--- error: "cannot PREPARE a transaction that has executed LISTEN, UNLISTEN, or NOTIFY"
-
--- create new regular tables
-CREATE TABLE eth.log_cids_i (LIKE eth.log_cids INCLUDING ALL);
-CREATE TABLE eth.access_list_elements_i (LIKE eth.access_list_elements INCLUDING ALL);
-CREATE TABLE eth.state_accounts_i (LIKE eth.state_accounts INCLUDING ALL);
-CREATE TABLE eth.storage_cids_i (LIKE eth.storage_cids INCLUDING ALL);
-CREATE TABLE eth.state_cids_i (LIKE eth.state_cids INCLUDING ALL);
-CREATE TABLE eth.receipt_cids_i (LIKE eth.receipt_cids INCLUDING ALL);
-CREATE TABLE eth.transaction_cids_i (LIKE eth.transaction_cids INCLUDING ALL);
-CREATE TABLE eth.uncle_cids_i (LIKE eth.uncle_cids INCLUDING ALL);
-CREATE TABLE eth.header_cids_i (LIKE eth.header_cids INCLUDING ALL);
-CREATE TABLE public.blocks_i (LIKE public.blocks INCLUDING ALL);
-
--- turn them into distributed hypertables
-SELECT create_distributed_hypertable('public.blocks_i', 'block_number', chunk_time_interval => 32768);
-SELECT create_distributed_hypertable('eth.header_cids_i', 'block_number', chunk_time_interval => 32768);
-SELECT create_distributed_hypertable('eth.uncle_cids_i', 'block_number', chunk_time_interval => 32768);
-SELECT create_distributed_hypertable('eth.transaction_cids_i', 'block_number', chunk_time_interval => 32768);
-SELECT create_distributed_hypertable('eth.receipt_cids_i', 'block_number', chunk_time_interval => 32768);
-SELECT create_distributed_hypertable('eth.state_cids_i', 'block_number', chunk_time_interval => 32768);
-SELECT create_distributed_hypertable('eth.storage_cids_i', 'block_number', chunk_time_interval => 32768);
-SELECT create_distributed_hypertable('eth.state_accounts_i', 'block_number', chunk_time_interval => 32768);
-SELECT create_distributed_hypertable('eth.access_list_elements_i', 'block_number', chunk_time_interval => 32768);
-SELECT create_distributed_hypertable('eth.log_cids_i', 'block_number', chunk_time_interval => 32768);
-
--- migrate data
-INSERT INTO eth.log_cids_i (SELECT * FROM eth.log_cids);
-INSERT INTO eth.access_list_elements_i (SELECT * FROM eth.access_list_elements);
-INSERT INTO eth.state_accounts_i (SELECT * FROM eth.state_accounts);
-INSERT INTO eth.storage_cids_i (SELECT * FROM eth.storage_cids);
-INSERT INTO eth.state_cids_i (SELECT * FROM eth.state_cids);
-INSERT INTO eth.receipt_cids_i (SELECT * FROM eth.receipt_cids);
-INSERT INTO eth.transaction_cids_i (SELECT * FROM eth.transaction_cids);
-INSERT INTO eth.uncle_cids_i (SELECT * FROM eth.uncle_cids);
-INSERT INTO eth.header_cids_i (SELECT * FROM eth.header_cids);
-INSERT INTO public.blocks_i (SELECT * FROM public.blocks);
-
--- drop tables
-DROP TABLE eth.log_cids;
-DROP TABLE eth.access_list_elements;
-DROP TABLE eth.state_accounts;
-DROP TABLE eth.storage_cids;
-DROP TABLE eth.state_cids;
-DROP TABLE eth.receipt_cids;
-DROP TABLE eth.transaction_cids;
-DROP TABLE eth.uncle_cids;
-DROP TABLE eth.header_cids;
-DROP TABLE public.blocks;
-
--- rename distributed hypertables
-ALTER TABLE eth.log_cids_i RENAME TO log_cids;
-ALTER TABLE eth.access_list_elements_i RENAME TO access_list_elements;
-ALTER TABLE eth.state_accounts_i RENAME TO state_accounts;
-ALTER TABLE eth.storage_cids_i RENAME TO storage_cids;
-ALTER TABLE eth.state_cids_i RENAME TO state_cids;
-ALTER TABLE eth.receipt_cids_i RENAME TO receipt_cids;
-ALTER TABLE eth.transaction_cids_i RENAME TO transaction_cids;
-ALTER TABLE eth.uncle_cids_i RENAME TO uncle_cids;
-ALTER TABLE eth.header_cids_i RENAME TO header_cids;
-ALTER TABLE public.blocks_i RENAME TO blocks;
-CALL distributed_exec('ALTER TABLE eth.log_cids_i RENAME TO log_cids');
-CALL distributed_exec('ALTER TABLE eth.access_list_elements_i RENAME TO access_list_elements');
-CALL distributed_exec('ALTER TABLE eth.state_accounts_i RENAME TO state_accounts');
-CALL distributed_exec('ALTER TABLE eth.storage_cids_i RENAME TO storage_cids');
-CALL distributed_exec('ALTER TABLE eth.state_cids_i RENAME TO state_cids');
-CALL distributed_exec('ALTER TABLE eth.receipt_cids_i RENAME TO receipt_cids');
-CALL distributed_exec('ALTER TABLE eth.transaction_cids_i RENAME TO transaction_cids');
-CALL distributed_exec('ALTER TABLE eth.uncle_cids_i RENAME TO uncle_cids');
-CALL distributed_exec('ALTER TABLE eth.header_cids_i RENAME TO header_cids');
-CALL distributed_exec('ALTER TABLE public.blocks_i RENAME TO blocks');
-
--- update version
-INSERT INTO public.db_version (singleton, version) VALUES (true, 'v4.0.00-dh')
-    ON CONFLICT (singleton) DO UPDATE SET (version, tstamp) = ('v4.0.0-dh', NOW());
-
--- +goose Down
-INSERT INTO public.db_version (singleton, version) VALUES (true, 'v4.0.0')
-    ON CONFLICT (singleton) DO UPDATE SET (version, tstamp) = ('v4.0.0', NOW());
--- reversing conversion to hypertables requires migrating all data from every chunk back to a single table
--- create new regular tables
-CREATE TABLE eth.log_cids_i (LIKE eth.log_cids INCLUDING ALL);
-CREATE TABLE eth.access_list_elements_i (LIKE eth.access_list_elements INCLUDING ALL);
-CREATE TABLE eth.state_accounts_i (LIKE eth.state_accounts INCLUDING ALL);
-CREATE TABLE eth.storage_cids_i (LIKE eth.storage_cids INCLUDING ALL);
-CREATE TABLE eth.state_cids_i (LIKE eth.state_cids INCLUDING ALL);
-CREATE TABLE eth.receipt_cids_i (LIKE eth.receipt_cids INCLUDING ALL);
-CREATE TABLE eth.transaction_cids_i (LIKE eth.transaction_cids INCLUDING ALL);
-CREATE TABLE eth.uncle_cids_i (LIKE eth.uncle_cids INCLUDING ALL);
-CREATE TABLE eth.header_cids_i (LIKE eth.header_cids INCLUDING ALL);
-CREATE TABLE public.blocks_i (LIKE public.blocks INCLUDING ALL);
-
--- migrate data
-INSERT INTO eth.log_cids_i (SELECT * FROM eth.log_cids);
-INSERT INTO eth.access_list_elements_i (SELECT * FROM eth.access_list_elements);
-INSERT INTO eth.state_accounts_i (SELECT * FROM eth.state_accounts);
-INSERT INTO eth.storage_cids_i (SELECT * FROM eth.storage_cids);
-INSERT INTO eth.state_cids_i (SELECT * FROM eth.state_cids);
-INSERT INTO eth.receipt_cids_i (SELECT * FROM eth.receipt_cids);
-INSERT INTO eth.transaction_cids_i (SELECT * FROM eth.transaction_cids);
-INSERT INTO eth.uncle_cids_i (SELECT * FROM eth.uncle_cids);
-INSERT INTO eth.header_cids_i (SELECT * FROM eth.header_cids);
-INSERT INTO public.blocks_i (SELECT * FROM public.blocks);
-
--- drop distributed hypertables
-DROP TABLE eth.log_cids;
-DROP TABLE eth.access_list_elements;
-DROP TABLE eth.state_accounts;
-DROP TABLE eth.storage_cids;
-DROP TABLE eth.state_cids;
-DROP TABLE eth.receipt_cids;
-DROP TABLE eth.transaction_cids;
-DROP TABLE eth.uncle_cids;
-DROP TABLE eth.header_cids;
-DROP TABLE public.blocks;
-
--- rename hypertable tables
-ALTER TABLE eth.log_cids_i RENAME TO log_cids;
-ALTER TABLE eth.access_list_elements_i RENAME TO access_list_elements;
-ALTER TABLE eth.state_accounts_i RENAME TO state_accounts;
-ALTER TABLE eth.storage_cids_i RENAME TO storage_cids;
-ALTER TABLE eth.state_cids_i RENAME TO state_cids;
-ALTER TABLE eth.receipt_cids_i RENAME TO receipt_cids;
-ALTER TABLE eth.transaction_cids_i RENAME TO transaction_cids;
-ALTER TABLE eth.uncle_cids_i RENAME TO uncle_cids;
-ALTER TABLE eth.header_cids_i RENAME TO header_cids;
-ALTER TABLE public.blocks_i RENAME TO blocks;

db/migrations/00019_create_known_gaps_table.sql

@@ -0,0 +1,10 @@
+-- +goose Up
+CREATE TABLE eth_meta.known_gaps (
+    starting_block_number bigint PRIMARY KEY,
+    ending_block_number bigint,
+    checked_out boolean,
+    processing_key bigint
+);
+
+-- +goose Down
+DROP TABLE eth_meta.known_gaps;

db/migrations/00020_attach_data_nodes.sql

@@ -1,5 +0,0 @@
--- +goose Up
--- to be generated by a script
-
--- +goose Down
--- to be generated by a script

db/migrations/00021_convert_to_hypertables.sql

@@ -0,0 +1,68 @@
+-- +goose Up
+SELECT create_hypertable('public.blocks', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+SELECT create_hypertable('eth.header_cids', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+SELECT create_hypertable('eth.uncle_cids', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+SELECT create_hypertable('eth.transaction_cids', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+SELECT create_hypertable('eth.receipt_cids', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+SELECT create_hypertable('eth.state_cids', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+SELECT create_hypertable('eth.storage_cids', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+SELECT create_hypertable('eth.state_accounts', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+SELECT create_hypertable('eth.access_list_elements', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+SELECT create_hypertable('eth.log_cids', 'block_number', migrate_data => true, chunk_time_interval => 32768);
+
+-- update version
+INSERT INTO public.db_version (singleton, version) VALUES (true, 'v4.0.0-h')
+    ON CONFLICT (singleton) DO UPDATE SET (version, tstamp) = ('v4.0.0-h', NOW());
+
+-- +goose Down
+INSERT INTO public.db_version (singleton, version) VALUES (true, 'v4.0.0')
+    ON CONFLICT (singleton) DO UPDATE SET (version, tstamp) = ('v4.0.0', NOW());
+
+-- reversing conversion to hypertable requires migrating all data from every chunk back to a single table
+-- create new regular tables
+CREATE TABLE eth.log_cids_i (LIKE eth.log_cids INCLUDING ALL);
+CREATE TABLE eth.access_list_elements_i (LIKE eth.access_list_elements INCLUDING ALL);
+CREATE TABLE eth.state_accounts_i (LIKE eth.state_accounts INCLUDING ALL);
+CREATE TABLE eth.storage_cids_i (LIKE eth.storage_cids INCLUDING ALL);
+CREATE TABLE eth.state_cids_i (LIKE eth.state_cids INCLUDING ALL);
+CREATE TABLE eth.receipt_cids_i (LIKE eth.receipt_cids INCLUDING ALL);
+CREATE TABLE eth.transaction_cids_i (LIKE eth.transaction_cids INCLUDING ALL);
+CREATE TABLE eth.uncle_cids_i (LIKE eth.uncle_cids INCLUDING ALL);
+CREATE TABLE eth.header_cids_i (LIKE eth.header_cids INCLUDING ALL);
+CREATE TABLE public.blocks_i (LIKE public.blocks INCLUDING ALL);
+
+-- migrate data
+INSERT INTO eth.log_cids_i (SELECT * FROM eth.log_cids);
+INSERT INTO eth.access_list_elements_i (SELECT * FROM eth.access_list_elements);
+INSERT INTO eth.state_accounts_i (SELECT * FROM eth.state_accounts);
+INSERT INTO eth.storage_cids_i (SELECT * FROM eth.storage_cids);
+INSERT INTO eth.state_cids_i (SELECT * FROM eth.state_cids);
+INSERT INTO eth.receipt_cids_i (SELECT * FROM eth.receipt_cids);
+INSERT INTO eth.transaction_cids_i (SELECT * FROM eth.transaction_cids);
+INSERT INTO eth.uncle_cids_i (SELECT * FROM eth.uncle_cids);
+INSERT INTO eth.header_cids_i (SELECT * FROM eth.header_cids);
+INSERT INTO public.blocks_i (SELECT * FROM public.blocks);
+
+-- drop hypertables
+DROP TABLE eth.log_cids;
+DROP TABLE eth.access_list_elements;
+DROP TABLE eth.state_accounts;
+DROP TABLE eth.storage_cids;
+DROP TABLE eth.state_cids;
+DROP TABLE eth.receipt_cids;
+DROP TABLE eth.transaction_cids;
+DROP TABLE eth.uncle_cids;
+DROP TABLE eth.header_cids;
+DROP TABLE public.blocks;
+
+-- rename new tables
+ALTER TABLE eth.log_cids_i RENAME TO log_cids;
+ALTER TABLE eth.access_list_elements_i RENAME TO access_list_elements;
+ALTER TABLE eth.state_accounts_i RENAME TO state_accounts;
+ALTER TABLE eth.storage_cids_i RENAME TO storage_cids;
+ALTER TABLE eth.state_cids_i RENAME TO state_cids;
+ALTER TABLE eth.receipt_cids_i RENAME TO receipt_cids;
+ALTER TABLE eth.transaction_cids_i RENAME TO transaction_cids;
+ALTER TABLE eth.uncle_cids_i RENAME TO uncle_cids;
+ALTER TABLE eth.header_cids_i RENAME TO header_cids;
+ALTER TABLE public.blocks_i RENAME TO blocks;

db/migrations/00022_create_stored_functions.sql

@@ -155,26 +155,26 @@ BEGIN
     END IF;
 
     -- select all of the state nodes for this snapshot: the latest state node record at every unique path
-    SELECT DISTINCT ON (state_path) blocks.data, state_cids.state_leaf_key, state_cids.cid, state_cids.state_path,
+    SELECT ARRAY (SELECT DISTINCT ON (state_path) ROW (blocks.data, state_cids.state_leaf_key, state_cids.cid, state_cids.state_path,
-        state_cids.node_type, state_cids.mh_key
+        state_cids.node_type, state_cids.mh_key)
-    INTO results
     FROM eth.state_cids
         INNER JOIN public.blocks
             ON (state_cids.mh_key, state_cids.block_number) = (blocks.key, blocks.block_number)
     WHERE state_cids.block_number BETWEEN starting_height AND ending_height
-    ORDER BY state_path, block_number DESC;
+    ORDER BY state_path, state_cids.block_number DESC)
+    INTO results;
 
     -- from the set returned above, insert public.block records at the ending_height block number
     INSERT INTO public.blocks (block_number, key, data)
     SELECT ending_height, r.mh_key, r.data
-    FROM results r;
+    FROM unnest(results) r;
 
     -- from the set returned above, insert eth.state_cids records at the ending_height block number
     -- anchoring all the records to the canonical header found at ending_height
     INSERT INTO eth.state_cids (block_number, header_id, state_leaf_key, cid, state_path, node_type, diff, mh_key)
     SELECT ending_height, canonical_hash, r.state_leaf_key, r.cid, r.state_path, r.node_type, false, r.mh_key
-    FROM results r
+    FROM unnest(results) r
-    ON CONFLICT (state_path, header_id) DO NOTHING;
+    ON CONFLICT (state_path, header_id, block_number) DO NOTHING;
 END
 $BODY$
 LANGUAGE 'plpgsql';
@@ -208,29 +208,29 @@ BEGIN
     END IF;
 
     -- select all of the storage nodes for this snapshot: the latest storage node record at every unique state leaf key
-    SELECT DISTINCT ON (state_leaf_key, storage_path) block.data, storage_cids.state_path, storage_cids.storage_leaf_key,
+    SELECT ARRAY (SELECT DISTINCT ON (state_leaf_key, storage_path) ROW (blocks.data, storage_cids.state_path, storage_cids.storage_leaf_key,
-     storage_cids.cid, storage_cids.storage_path, storage_cids.node_type, storage_cids.mh_key
+     storage_cids.cid, storage_cids.storage_path, storage_cids.node_type, storage_cids.mh_key)
-    INTO results
     FROM eth.storage_cids
         INNER JOIN public.blocks
         ON (storage_cids.mh_key, storage_cids.block_number) = (blocks.key, blocks.block_number)
         INNER JOIN eth.state_cids
         ON (storage_cids.state_path, storage_cids.header_id) = (state_cids.state_path, state_cids.header_id)
     WHERE storage_cids.block_number BETWEEN starting_height AND ending_height
-    ORDER BY state_path, storage_path, block_number DESC;
+    ORDER BY state_leaf_key, storage_path, storage_cids.state_path, storage_cids.block_number DESC)
+    INTO results;
 
     -- from the set returned above, insert public.block records at the ending_height block number
     INSERT INTO public.blocks (block_number, key, data)
     SELECT ending_height, r.mh_key, r.data
-    FROM results r;
+    FROM unnest(results) r;
 
     -- from the set returned above, insert eth.state_cids records at the ending_height block number
     -- anchoring all the records to the canonical header found at ending_height
     INSERT INTO eth.storage_cids (block_number, header_id, state_path, storage_leaf_key, cid, storage_path,
                               node_type, diff, mh_key)
     SELECT ending_height, canonical_hash, r.state_path, r.storage_leaf_key, r.cid, r.storage_path, r.node_type, false, r.mh_key
-    FROM results r
+    FROM unnest(results) r
-    ON CONFLICT (storage_path, state_path, header_id) DO NOTHING;
+    ON CONFLICT (storage_path, state_path, header_id, block_number) DO NOTHING;
 END
 $BODY$
 LANGUAGE 'plpgsql';

db/migrations/00022_set_replication_factor.sql

@@ -1,27 +0,0 @@
--- +goose Up
--- to be generated by a script
-
-SELECT set_replication_factor('public.blocks', 3);
-SELECT set_replication_factor('eth.header_cids', 3);
-SELECT set_replication_factor('eth.uncle_cids', 3);
-SELECT set_replication_factor('eth.transaction_cids', 3);
-SELECT set_replication_factor('eth.receipt_cids', 3);
-SELECT set_replication_factor('eth.state_cids', 3);
-SELECT set_replication_factor('eth.storage_cids', 3);
-SELECT set_replication_factor('eth.state_accounts', 3);
-SELECT set_replication_factor('eth.access_list_elements', 3);
-SELECT set_replication_factor('eth.log_cids', 3);
-
--- +goose Down
--- to be generated by a script
-
-SELECT set_replication_factor('public.blocks', 1);
-SELECT set_replication_factor('eth.header_cids', 1);
-SELECT set_replication_factor('eth.uncle_cids', 1);
-SELECT set_replication_factor('eth.transaction_cids', 1);
-SELECT set_replication_factor('eth.receipt_cids', 1);
-SELECT set_replication_factor('eth.state_cids', 1);
-SELECT set_replication_factor('eth.storage_cids', 1);
-SELECT set_replication_factor('eth.state_accounts', 1);
-SELECT set_replication_factor('eth.access_list_elements', 1);
-SELECT set_replication_factor('eth.log_cids', 1);

docker-compose.test.yml

@@ -1,62 +1,27 @@
 version: '3.2'
 
 services:
-  access-node:
+  migrations:
+    restart: on-failure
+    depends_on:
+      - ipld-eth-db
+    image: vulcanize/ipld-eth-db
+    # Build image using local context
+    build: .
+    environment:
+      DATABASE_USER: "vdbm"
+      DATABASE_NAME: "vulcanize_testing"
+      DATABASE_PASSWORD: "password"
+      DATABASE_HOSTNAME: "ipld-eth-db"
+      DATABASE_PORT: 5432
+
+  ipld-eth-db:
     image: timescale/timescaledb:latest-pg14
     restart: always
-    container_name: access-node
-    depends_on:
-      - data-node-1
-      - data-node-2
-      - data-node-3
     command: ["postgres", "-c", "log_statement=all"]
     environment:
-      POSTGRES_USER: "postgres"
+      POSTGRES_USER: "vdbm"
-      POSTGRES_DB: "vulcanize_testing_v4"
+      POSTGRES_DB: "vulcanize_testing"
       POSTGRES_PASSWORD: "password"
     ports:
-      - "127.0.0.1:8066:5432"
+      - "127.0.0.1:8077:5432"
-    volumes:
-      - ./scripts/init-access-node.sh:/docker-entrypoint-initdb.d/init-access-node.sh
-
-  data-node-1:
-    image: timescale/timescaledb:latest-pg14
-    restart: unless-stopped
-    container_name: data-node-1
-    command: ["postgres", "-c", "log_statement=all"]
-    environment:
-      POSTGRES_USER: "postgres"
-      POSTGRES_DB: "vulcanize_testing_v4"
-      POSTGRES_PASSWORD: "password"
-    ports:
-      - "127.0.0.1:8067:5432"
-    volumes:
-      - ./scripts/init-data-node.sh:/docker-entrypoint-initdb.d/init-data-node.sh
-
-  data-node-2:
-    image: timescale/timescaledb:latest-pg14
-    restart: unless-stopped
-    container_name: data-node-2
-    command: ["postgres", "-c", "log_statement=all"]
-    environment:
-      POSTGRES_USER: "postgres"
-      POSTGRES_DB: "vulcanize_testing_v4"
-      POSTGRES_PASSWORD: "password"
-    ports:
-      - "127.0.0.1:8068:5432"
-    volumes:
-      - ./scripts/init-data-node.sh:/docker-entrypoint-initdb.d/init-data-node.sh
-
-  data-node-3:
-    image: timescale/timescaledb:latest-pg14
-    restart: unless-stopped
-    container_name: data-node-3
-    command: ["postgres", "-c", "log_statement=all"]
-    environment:
-      POSTGRES_USER: "postgres"
-      POSTGRES_DB: "vulcanize_testing_v4"
-      POSTGRES_PASSWORD: "password"
-    ports:
-      - "127.0.0.1:8069:5432"
-    volumes:
-      - ./scripts/init-data-node.sh:/docker-entrypoint-initdb.d/init-data-node.sh

docker-compose.yml

@@ -1,14 +1,26 @@
 version: '3.2'
 
 services:
+  migrations:
+    restart: on-failure
+    depends_on:
+      - ipld-eth-db
+    # Use an existing image
+    image: vulcanize/ipld-eth-db:v4.1.4-alpha
+    environment:
+      DATABASE_USER: "vdbm"
+      DATABASE_NAME: "vulcanize_testing"
+      DATABASE_PASSWORD: "password"
+      DATABASE_HOSTNAME: "ipld-eth-db"
+      DATABASE_PORT: 5432
+
   ipld-eth-db:
+    image: timescale/timescaledb:latest-pg14
     restart: always
-    image: vulcanize/ipld-eth-db
+    command: ["postgres", "-c", "log_statement=all"]
-    build: .
     environment:
       POSTGRES_USER: "vdbm"
       POSTGRES_DB: "vulcanize_testing"
       POSTGRES_PASSWORD: "password"
-      hostname: db
     ports:
-      - "127.0.0.1:8077:5432"
+      - "127.0.0.1:8077:5432"

schema.sql

@@ -37,6 +37,13 @@ COMMENT ON EXTENSION timescaledb IS 'Enables scalable inserts and complex querie
 CREATE SCHEMA eth;
 
 
+--
+-- Name: eth_meta; Type: SCHEMA; Schema: -; Owner: -
+--
+
+CREATE SCHEMA eth_meta;
+
+
 SET default_tablespace = '';
 
 SET default_table_access_method = heap;
@@ -534,6 +541,30 @@ CREATE TABLE eth.uncle_cids (
 );
 
 
+--
+-- Name: known_gaps; Type: TABLE; Schema: eth_meta; Owner: -
+--
+
+CREATE TABLE eth_meta.known_gaps (
+    starting_block_number bigint NOT NULL,
+    ending_block_number bigint,
+    checked_out boolean,
+    processing_key bigint
+);
+
+
+--
+-- Name: watched_addresses; Type: TABLE; Schema: eth_meta; Owner: -
+--
+
+CREATE TABLE eth_meta.watched_addresses (
+    address character varying(66) NOT NULL,
+    created_at bigint NOT NULL,
+    watched_at bigint NOT NULL,
+    last_filled_at bigint DEFAULT 0 NOT NULL
+);
+
+
 --
 -- Name: blocks; Type: TABLE; Schema: public; Owner: -
 --
@@ -695,6 +726,22 @@ ALTER TABLE ONLY eth.uncle_cids
     ADD CONSTRAINT uncle_cids_pkey PRIMARY KEY (block_hash, block_number);
 
 
+--
+-- Name: known_gaps known_gaps_pkey; Type: CONSTRAINT; Schema: eth_meta; Owner: -
+--
+
+ALTER TABLE ONLY eth_meta.known_gaps
+    ADD CONSTRAINT known_gaps_pkey PRIMARY KEY (starting_block_number);
+
+
+--
+-- Name: watched_addresses watched_addresses_pkey; Type: CONSTRAINT; Schema: eth_meta; Owner: -
+--
+
+ALTER TABLE ONLY eth_meta.watched_addresses
+    ADD CONSTRAINT watched_addresses_pkey PRIMARY KEY (address);
+
+
 --
 -- Name: blocks blocks_pkey; Type: CONSTRAINT; Schema: public; Owner: -
 --

scripts/add_nodes.sh

@@ -1,138 +0,0 @@
-#!/bin/bash
-# Guards
-if [ -z "$1" ] || [ -z "$2" ] || [ -z "$3" ] || [ -z "$4" ] || [ -z "$5" ]
-  then
-    echo "Env variables not provided"
-    echo "Usage: ./add_nodes.sh <MIGRATION_FILE_PATH:string> <NODE_NAME:string> <NODE_HOST:string> <NODE_PORT:numeric/string> <NODE_DATABASE:string> <EXECUTE_SQL:bool>"
-    echo "Only <EXECUTE_SQL> is optional"
-    exit 1
-fi
-if [ -z "$6" ]
-  then
-    echo "EXECUTE_SQL not set, will not run statements against an access server"
-  else
-    echo "EXECUTE_SQL is set, will run stataments against an access server"
-    echo "Note: this mode is not recommended except in the case when the migration has already been applied with previous
-    nodes and we need/want to add more while still recording them in the existing migration (and adding their Down statements to said migration)"
-    echo "Expected environment variables:"
-    echo "DATABASE_HOSTNAME, DATABASE_NAME, DATABASE_PORT, DATABASE_USER"
-    echo "For now, DATABASE_PASSWORD will be prompted for on each psql call"
-fi
-
-# Remote DB node info
-export MIGRATION_FILE_PATH=$1
-export NODE_NAME=$2
-export NODE_HOST=$3
-export NODE_PORT=$4
-export NODE_DATABASE=$5
-printf "Enter the ${NODE_HOST} database password:\n"
-stty -echo
-read NODE_PASSWORD
-stty echo
-export NODE_PASSWORD
-
-if ! [ -z "$6" ]
-  then
-    # Access DB info
-    echo "heeeeey"
-    export DATABASE_HOSTNAME=localhost
-    export DATABASE_PORT=5432
-    export DATABASE_USER=vdbm
-    export DATABASE_NAME=vulcanize_db
-fi
-
-# Array of distributed hypertable names
-declare -a tables_names=("public.blocks" "eth.header_cid" "eth.uncle_cids" "eth.transaction_cids"
-                "eth.receipt_cids" "eth.state_cid" "eth.storage_cids" "eth.state_accounts"
-                "eth.access_list_elements" "eth.log_cids"
-                )
-# Array to append Up statements to for later (optional) execution
-declare -a up_stmts=()
-
-echo "Writing Up and Down statements to provided migration file at ${migration_file_path}"
-
-# Create add node statement
-up_add_pg_str="SELECT add_data_node('${NODE_NAME}', host => '${NODE_HOST}', port => ${NODE_PORT}, database => '${NODE_DATABASE}', password => '${NODE_PASSWORD}');"
-up_stmts+=(${up_add_pg_str})
-
-# Insert at the 3rd line of the file
-sed -i.bak '3 i\
-'"${up_add_pg_str}"'
-' "${MIGRATION_FILE_PATH}"
-
-# Check for error
-if [[ $? -eq 0 ]]; then
-    echo "Wrote Up add node statement ${up_add_pg_str}"
-else
-    echo "Could not write Up add node statement ${up_add_pg_str}. Is the migration file path correct?"
-    exit 1
-fi
-
-# Create attach node statements
-for table_name in "${tables_names[@]}"
-do
-  up_attach_pg_str="SELECT attach_data_node('${NODE_NAME}', '${table_name}', if_not_attached => true);"
-  up_stmts+=(${up_attach_pg_str})
-  # Insert at the 4th line of the file
-  sed -i.bak '4 i\
-'"${up_attach_pg_str}"'
-' "${MIGRATION_FILE_PATH}"
-  # Check for error
-  if [[ $? -eq 0 ]]; then
-      echo "Wrote Up attach node statement ${up_attach_pg_str}"
-  else
-      echo "Could not write Up attach node statement ${up_attach_pg_str}. Is the migration file path correct?"
-      exit 1
-  fi
-done
-
-## Create detach and remove node statement
-down_attach_pg_str="SELECT detach_data_node('${NODE_NAME}', force => true, if_attached = true);"
-down_add_pg_str="SELECT delete_data_node('${NODE_NAME}', force => true, if_attached => true);"
-
-# Append them at the last line in the file
-sed -i.bak '$ a\
-'"${down_attach_pg_str}"'
-' "${MIGRATION_FILE_PATH}"
-# Check for error
-if [[ $? -eq 0 ]]; then
-    echo "Wrote Down attach node statement ${down_attach_pg_str}"
-else
-    echo "Could not write Down attach node statement ${down_attach_pg_str}. Is the migration file path correct?"
-    exit 1
-fi
-# Append them at the last line in the file
-sed -i.bak '$ a\
-'"${down_add_pg_str}"'
-' "${MIGRATION_FILE_PATH}"
-# Check for error
-if [[ $? -eq 0 ]]; then
-    echo "Wrote Down add node statement ${down_add_pg_str}"
-else
-    echo "Could not write Down add node statement ${down_add_pg_str}. Is the migration file path correct?"
-    exit 1
-fi
-
-# Execute Up statements on the server if we are in that mode
-if [ -z "$6" ]
-  then
-    echo "Done!"
-    exit 0
-fi
-
-echo "Executing Up statements against provided server"
-
-for up_stmt in "${up_stmts[@]}"
-do
-  psql -c '\x' -c "${up_stmt}" -h $DATABASE_HOSTNAME -p $DATABASE_PORT -U $DATABASE_USER -d $DATABASE_NAME -W
-  # Check for error
-  if [[ $? -eq 0 ]]; then
-      echo "Executed Up statement ${up_stmt}}"
-  else
-      echo "Could not execute Up statement ${up_stmt}. Is the migration file path correct?"
-      exit 1
-  fi
-done
-
-echo "Done!"
-exit 0

scripts/init-access-node.sh

@@ -1,29 +0,0 @@
-#!/bin/sh
-set -e
-
-# https://docs.timescale.com/timescaledb/latest/how-to-guides/multinode-timescaledb/multinode-config/
-
-# To achieve good query performance you need to enable partition-wise aggregation on the access node. This pushes down aggregation queries to the data nodes.
-# https://www.postgresql.org/docs/12/runtime-config-query.html#enable_partitionwise_aggregate
-sed -ri "s!^#?(enable_partitionwise_aggregate)\s*=.*!\1 = on!" /var/lib/postgresql/data/postgresql.conf
-grep "enable_partitionwise_aggregate" /var/lib/postgresql/data/postgresql.conf
-
-# JIT should be set to off on the access node as JIT currently doesn't work well with distributed queries.
-# https://www.postgresql.org/docs/12/runtime-config-query.html#jit
-sed -ri "s!^#?(jit)\s*=.*!\1 = off!" /var/lib/postgresql/data/postgresql.conf
-grep "jit" /var/lib/postgresql/data/postgresql.conf
-
-# https://docs.timescale.com/timescaledb/latest/how-to-guides/multinode-timescaledb/multinode-auth/
-# https://docs.timescale.com/timescaledb/latest/how-to-guides/multinode-timescaledb/multinode-auth/#password-authentication
-
-# Set password_encryption = 'scram-sha-256' in postgresql.conf on the access node.
-sed -ri "s!^#?(password_encryption)\s*=.*!\1 = 'scram-sha-256'!" /var/lib/postgresql/data/postgresql.conf
-grep "password_encryption" /var/lib/postgresql/data/postgresql.conf
-
-# Append to data/passfile *:*:*:ROLE:ROLE_PASSWORD
-# This file stores the passwords for each role that the access node connects to on the data nodes.
-echo "*:*:*:postgres:password">>/var/lib/postgresql/data/passfile
-chmod 0600 /var/lib/postgresql/data/passfile
-
-# Add "host  all  all  ACCESS_NODE_IP  scram-sha-256" pg_hba.conf on the data nodes.
-# Skipped. Using default "host  all  all  all  scram-sha-256" for now.

scripts/init-data-node.sh

@@ -1,27 +0,0 @@
-#!/bin/sh
-set -e
-
-# https://docs.timescale.com/timescaledb/latest/how-to-guides/configuration/timescaledb-config/#timescaledb-last-tuned-string
-# https://docs.timescale.com/timescaledb/latest/how-to-guides/multi-node-setup/required-configuration/
-
-# It is necessary to change the parameter max_prepared_transactions to a non-zero value ('150' is recommended).
-# https://www.postgresql.org/docs/12/runtime-config-resource.html#max_prepared_transactions
-sed -ri "s!^#?(max_prepared_transactions)\s*=.*!\1 = 150!" /var/lib/postgresql/data/postgresql.conf
-grep "max_prepared_transactions" /var/lib/postgresql/data/postgresql.conf
-
-# Statement timeout should be disabled on the data nodes and managed through the access node configuration if desired.
-# https://www.postgresql.org/docs/12/runtime-config-client.html#statement_timeout
-sed -ri "s!^#?(statement_timeout)\s*=.*!\1 = 0!" /var/lib/postgresql/data/postgresql.conf
-grep "statement_timeout" /var/lib/postgresql/data/postgresql.conf
-
-# On the data nodes, set the wal_level to logical or higher to move or copy chunks between data nodes.
-# https://www.postgresql.org/docs/14/runtime-config-wal.html
-sed -ri "s!^#?(wal_level)\s*=.*!\1 = 'logical'!" /var/lib/postgresql/data/postgresql.conf
-grep "wal_level" /var/lib/postgresql/data/postgresql.conf
-
-# https://docs.timescale.com/timescaledb/latest/how-to-guides/multinode-timescaledb/multinode-auth/
-# https://docs.timescale.com/timescaledb/latest/how-to-guides/multinode-timescaledb/multinode-auth/#password-authentication
-
-# Set password_encryption = 'scram-sha-256' in postgresql.conf on the data node.
-sed -ri "s!^#?(password_encryption)\s*=.*!\1 = 'scram-sha-256'!" /var/lib/postgresql/data/postgresql.conf
-grep "password_encryption" /var/lib/postgresql/data/postgresql.conf

scripts/startup_script.sh

@@ -7,13 +7,13 @@ VDB_PG_CONNECT=postgresql://$DATABASE_USER:$DATABASE_PASSWORD@$DATABASE_HOSTNAME
 
 # Run the DB migrations
 echo "Connecting with: $VDB_PG_CONNECT"
-sleep 15
 echo "Running database migrations"
-./goose -dir migrations/vulcanizedb postgres "$VDB_PG_CONNECT" up-to 21
+./goose -dir migrations/vulcanizedb postgres "$VDB_PG_CONNECT" up
 
 # If the db migrations ran without err
 if [[ $? -eq 0 ]]; then
     echo "Migration process ran successfully"
+    tail -f /dev/null
 else
     echo "Could not run migrations. Are the database details correct?"
     exit 1