diff --git a/.ci/java-versions.properties b/.ci/java-versions.properties
index 16b760507324d..7e1032df49851 100644
--- a/.ci/java-versions.properties
+++ b/.ci/java-versions.properties
@@ -7,5 +7,6 @@
 ES_BUILD_JAVA=openjdk12
 ES_RUNTIME_JAVA=java11
 GRADLE_TASK=build
+GRADLE_EXTRA_ARGS=-Dtests.bwc.refspec.7.x=peer-recovery-retention-leases-7.x
 
 
diff --git a/server/src/main/java/org/elasticsearch/index/seqno/ReplicationTracker.java b/server/src/main/java/org/elasticsearch/index/seqno/ReplicationTracker.java
index 1fb91d6abeadd..256fe571a24e5 100644
--- a/server/src/main/java/org/elasticsearch/index/seqno/ReplicationTracker.java
+++ b/server/src/main/java/org/elasticsearch/index/seqno/ReplicationTracker.java
@@ -489,7 +489,7 @@ public synchronized void renewPeerRecoveryRetentionLeases() {
                      * create peer recovery retention leases for every shard copy. TODO create leases lazily in that situation.
                      */
                     assert checkpoints.get(shardRouting.allocationId().getId()).tracked == false
-                        || indexSettings.getIndexVersionCreated().before(Version.V_8_0_0);
+                        || indexSettings.getIndexVersionCreated().before(Version.V_7_4_0);
                     return false;
                 }
                 return retentionLease.timestamp() <= renewalTimeMillis
@@ -742,7 +742,7 @@ private boolean invariant() {
         if (primaryMode
             && indexSettings.isSoftDeleteEnabled()
             && indexSettings.getIndexMetaData().getState() == IndexMetaData.State.OPEN
-            && indexSettings.getIndexVersionCreated().onOrAfter(Version.V_8_0_0)) {
+            && indexSettings.getIndexVersionCreated().onOrAfter(Version.V_7_4_0)) {
             // all tracked shard copies have a corresponding peer-recovery retention lease
             for (final ShardRouting shardRouting : routingTable.assignedShards()) {
                 if (checkpoints.get(shardRouting.allocationId().getId()).tracked) {
@@ -910,7 +910,7 @@ public synchronized void activatePrimaryMode(final long localCheckpoint) {
                  * We might have got here here via a rolling upgrade from an older version that doesn't create peer recovery retention
                  * leases for every shard copy, but in this case we do not expect any leases to exist.
                  */
-                if (indexSettings.getIndexVersionCreated().onOrAfter(Version.V_8_0_0)) {
+                if (indexSettings.getIndexVersionCreated().onOrAfter(Version.V_7_4_0)) {
                     // We are starting up the whole replication group from scratch: if we were not (i.e. this is a replica promotion) then
                     // this copy must already be in-sync and active and therefore holds a retention lease for itself.
                     assert routingTable.activeShards().equals(Collections.singletonList(primaryShard)) : routingTable.activeShards();