From 73d19271a9de503a964fb6b8a158e5a5d883af43 Mon Sep 17 00:00:00 2001
From: Armin Braun <me@obrown.io>
Date: Fri, 18 Sep 2020 12:20:23 +0200
Subject: [PATCH] Fix Races in
 testQueuedSnapshotOperationsAndBrokenRepoOnMasterFailOverMultipleRepos
 (#62431) (#62614)

This test (in-part) verifies that snapshot creation is not
retried on master fail-over once a snaphot has been started already.
Unless we wait for the snapshot creation to show up in the cluster
state before failing the master node though, we could run into a
race where the snapshot wasn't yet in the cluster state and a retry goes through
successfully.
---
 .../org/elasticsearch/snapshots/ConcurrentSnapshotsIT.java    | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/server/src/internalClusterTest/java/org/elasticsearch/snapshots/ConcurrentSnapshotsIT.java b/server/src/internalClusterTest/java/org/elasticsearch/snapshots/ConcurrentSnapshotsIT.java
index 538548fc9f0..8f71eadf0b2 100644
--- a/server/src/internalClusterTest/java/org/elasticsearch/snapshots/ConcurrentSnapshotsIT.java
+++ b/server/src/internalClusterTest/java/org/elasticsearch/snapshots/ConcurrentSnapshotsIT.java
@@ -784,17 +784,21 @@ public class ConcurrentSnapshotsIT extends AbstractSnapshotIntegTestCase {
         blockNodeOnAnyFiles(blockedRepoName, masterNode);
         final ActionFuture<AcknowledgedResponse> deleteFuture = startDeleteFromNonMasterClient(blockedRepoName, "*");
         waitForBlock(masterNode, blockedRepoName, TimeValue.timeValueSeconds(30L));
+        awaitNDeletionsInProgress(1);
         final ActionFuture<CreateSnapshotResponse> createBlockedSnapshot =
             startFullSnapshotFromNonMasterClient(blockedRepoName, "queued-snapshot");
+        awaitNSnapshotsInProgress(1);
 
         final long generation = getRepositoryData(repoName).getGenId();
         blockNodeOnAnyFiles(repoName, masterNode);
         final ActionFuture<CreateSnapshotResponse> snapshotThree = startFullSnapshotFromNonMasterClient(repoName, "snapshot-three");
         waitForBlock(masterNode, repoName, TimeValue.timeValueSeconds(30L));
+        awaitNSnapshotsInProgress(2);
 
         corruptIndexN(repoPath, generation);
 
         final ActionFuture<CreateSnapshotResponse> snapshotFour = startFullSnapshotFromNonMasterClient(repoName, "snapshot-four");
+        awaitNSnapshotsInProgress(3);
         internalCluster().stopCurrentMasterNode();
         ensureStableCluster(3);