Skip to content
Snippets Groups Projects
Commit beb850e1 authored by Robert Haas's avatar Robert Haas
Browse files

Properly set relpersistence for fake relcache entries.

This can result in buffers failing to be properly flushed at
checkpoint time, leading to data loss.

Report, diagnosis, and patch by Jeff Davis.
parent 9afc6481
Branches
Tags
No related merge requests found
......@@ -394,6 +394,8 @@ CreateFakeRelcacheEntry(RelFileNode rnode)
FakeRelCacheEntry fakeentry;
Relation rel;
Assert(InRecovery);
/* Allocate the Relation struct and all related space in one block. */
fakeentry = palloc0(sizeof(FakeRelCacheEntryData));
rel = (Relation) fakeentry;
......@@ -403,6 +405,9 @@ CreateFakeRelcacheEntry(RelFileNode rnode)
/* We will never be working with temp rels during recovery */
rel->rd_backend = InvalidBackendId;
/* It must be a permanent table if we're in recovery. */
rel->rd_rel->relpersistence = RELPERSISTENCE_PERMANENT;
/* We don't know the name of the relation; use relfilenode instead */
sprintf(RelationGetRelationName(rel), "%u", rnode.relNode);
......
......@@ -271,6 +271,8 @@ ReadBufferWithoutRelcache(RelFileNode rnode, ForkNumber forkNum,
SMgrRelation smgr = smgropen(rnode, InvalidBackendId);
Assert(InRecovery);
return ReadBuffer_common(smgr, RELPERSISTENCE_PERMANENT, forkNum, blockNum,
mode, strategy, &hit);
}
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Please register or to comment