Fix inconsistent code with shared invalidations of snapshots
authorMichael Paquier <[email protected]>
Mon, 28 Dec 2020 13:16:49 +0000 (22:16 +0900)
committerMichael Paquier <[email protected]>
Mon, 28 Dec 2020 13:16:49 +0000 (22:16 +0900)
The code in charge of processing a single invalidation message has been
using since 568d413 the structure for relation mapping messages.  This
had fortunately no consequence as both locate the database ID at the
same location, but it could become a problem in the future if this area
of the code changes.

Author: Konstantin Knizhnik
Discussion: https://postgr.es/m/8044c223-4d3a-2cdb-42bf-29940840ce94@postgrespro.ru
Back-through: 9.5

src/backend/utils/cache/inval.c

index 628d6f5d0cceb592a1ac3038a4c0359c1072214b..e7279d06a3ddde17162216eccf0b17769a9f84bf 100644 (file)
@@ -622,9 +622,9 @@ LocalExecuteInvalidationMessage(SharedInvalidationMessage *msg)
    else if (msg->id == SHAREDINVALSNAPSHOT_ID)
    {
        /* We only care about our own database and shared catalogs */
-       if (msg->rm.dbId == InvalidOid)
+       if (msg->sn.dbId == InvalidOid)
            InvalidateCatalogSnapshot();
-       else if (msg->rm.dbId == MyDatabaseId)
+       else if (msg->sn.dbId == MyDatabaseId)
            InvalidateCatalogSnapshot();
    }
    else