Postgres 14 (#2607)

* Postgres 14

Signed-off-by: wslulciuc <willy@datakin.com>

* fix migration backfill for psql14

Signed-off-by: Pawel Leszczynski <leszczynski.pawel@gmail.com>

---------

Signed-off-by: wslulciuc <willy@datakin.com>
Signed-off-by: Pawel Leszczynski <leszczynski.pawel@gmail.com>
Co-authored-by: Pawel Leszczynski <leszczynski.pawel@gmail.com>

Author: wslulciuc

.circleci/db-migration.sh

@@ -11,7 +11,7 @@
 # Usage: $ ./db-migration.sh
 
 # Version of PostgreSQL
-readonly POSTGRES_VERSION="12.1"
+readonly POSTGRES_VERSION="14"
 # Version of Marquez
 readonly MARQUEZ_VERSION=0.40.0
 # Build version of Marquez

README.md

@@ -100,7 +100,7 @@ Marquez uses a _multi_-project structure and contains the following modules:
 ## Requirements
 
 * [Java 17](https://adoptium.net)
-* [PostgreSQL 12.1](https://www.postgresql.org/download)
+* [PostgreSQL 14](https://www.postgresql.org/download)
 
 > **Note:** To connect to your running PostgreSQL instance, you will need the standard [`psql`](https://www.postgresql.org/docs/9.6/app-psql.html) tool.
 

api/src/main/java/marquez/db/migrations/V57_1__BackfillFacets.java

@@ -128,6 +128,18 @@ public void migrate(Context context) throws Exception {
     }
 
     int estimatedEventsCount = estimateCountLineageEvents();
+
+    if (estimatedEventsCount < 0) {
+      // https://www.postgresql.org/docs/current/catalog-pg-class.html
+      // -1 indicating that the row count is unknown.
+      // This happens when lineage_events table is empty.
+      log.info("Vacuuming lineage_events table");
+      jdbi.withHandle(h -> h.execute("VACUUM lineage_events;"));
+      log.info("Vacuuming lineage_events table finished");
+
+      estimatedEventsCount = estimateCountLineageEvents();
+    }
+
     log.info("Estimating {} events in lineage_events table", estimatedEventsCount);
     if (estimatedEventsCount == 0 && countLineageEvents() == 0) {
       // lineage_events table is empty -> no need to run migration

api/src/test/java/marquez/PostgresContainer.java

@@ -11,7 +11,7 @@
 import org.testcontainers.utility.DockerImageName;
 
 public final class PostgresContainer extends PostgreSQLContainer<PostgresContainer> {
-  private static final DockerImageName POSTGRES = DockerImageName.parse("postgres:12.12");
+  private static final DockerImageName POSTGRES = DockerImageName.parse("postgres:14");
   private static final int JDBC = 5;
 
   private static final Map<String, PostgresContainer> containers = new HashMap<>();

api/src/test/java/marquez/db/DbTest.java

@@ -33,7 +33,7 @@
 @Tag("DataAccessTests")
 @Testcontainers
 class DbTest {
-  private static final DockerImageName POSTGRES_12_1 = DockerImageName.parse("postgres:12.1");
+  private static final DockerImageName POSTGRES_12_1 = DockerImageName.parse("postgres:14");
 
   @Container
   private static final PostgreSQLContainer<?> DB_CONTAINER =

docker-compose.db.yml

@@ -1,7 +1,7 @@
 version: "3.7"
 services:
   db:
-    image: postgres:12.1
+    image: postgres:14
     container_name: marquez-db
     ports:
       - "5432:5432"

docker-compose.yml

@@ -18,7 +18,7 @@ services:
     entrypoint: ["/opt/marquez/wait-for-it.sh", "db:5432", "--", "./entrypoint.sh"]
 
   db:
-    image: postgres:12.1
+    image: postgres:14
     container_name: marquez-db
     ports:
       - "5432:5432"