nbtree VACUUM: cope with topparent inconsistencies.

Avoid "right sibling %u of block %u is not next child" errors when
vacuuming a corrupt nbtree index.  Just LOG the issue and press on.
That way VACUUM will have a decent chance of finishing off all required
processing for the index (and for the table as a whole).

This is similar to recent work from commit 5abff197, as well as work
from commit 5b861baa (later backpatched as commit 43e409ce), which
taught nbtree VACUUM to keep going when its "re-find" check fails.  The
hardening added by this commit takes place directly after the "re-find"
check, right before the critical section for the first stage of page
deletion.

Author: Peter Geoghegan <pg@bowt.ie>
Discussion: https://postgr.es/m/CAH2-Wz=dayg0vjs4+er84TS9ami=csdzjpuiCGbEw=idhwqhzQ@mail.gmail.com
Backpatch: 11- (all supported versions).

diff --git a/src/backend/access/nbtree/nbtpage.c b/src/backend/access/nbtree/nbtpage.c
index 0f4901b614c2f24df29a73534c29a07b4587cd05..878c096c97fc7107e00b1ea387d6226ee1a29cd3 100644 (file)
--- a/src/backend/access/nbtree/nbtpage.c
+++ b/src/backend/access/nbtree/nbtpage.c
@@ -1772,12 +1772,6 @@ _bt_mark_page_halfdead(Relation rel, Buffer leafbuf, BTStack stack)
                                 &topparent, &topparentrightsib))
        return false;
 
-   /*
-    * Check that the parent-page index items we're about to delete/overwrite
-    * in subtree parent page contain what we expect.  This can fail if the
-    * index has become corrupt for some reason.  We want to throw any error
-    * before entering the critical section --- otherwise it'd be a PANIC.
-    */
    page = BufferGetPage(subtreeparent);
    opaque = (BTPageOpaque) PageGetSpecialPointer(page);
 
@@ -1795,8 +1789,17 @@ _bt_mark_page_halfdead(Relation rel, Buffer leafbuf, BTStack stack)
    nextoffset = OffsetNumberNext(poffset);
    itemid = PageGetItemId(page, nextoffset);
    itup = (IndexTuple) PageGetItem(page, itemid);
+
+   /*
+    * Check that the parent-page index items we're about to delete/overwrite
+    * in subtree parent page contain what we expect.  This can fail if the
+    * index has become corrupt for some reason.  When that happens we back
+    * out of deletion of the leafbuf subtree.  (This is just like the case
+    * where _bt_lock_subtree_parent() cannot "re-find" leafbuf's downlink.)
+    */
    if (BTreeTupleGetDownLink(itup) != topparentrightsib)
-       ereport(ERROR,
+   {
+       ereport(LOG,
                (errcode(ERRCODE_INDEX_CORRUPTED),
                 errmsg_internal("right sibling %u of block %u is not next child %u of block %u in index \"%s\"",
                                 topparentrightsib, topparent,
@@ -1804,6 +1807,11 @@ _bt_mark_page_halfdead(Relation rel, Buffer leafbuf, BTStack stack)
                                 BufferGetBlockNumber(subtreeparent),
                                 RelationGetRelationName(rel))));
 
+       _bt_relbuf(rel, subtreeparent);
+       Assert(false);
+       return false;
+   }
+
    /*
     * Any insert which would have gone on the leaf block will now go to its
     * right sibling.  In other words, the key space moves right.
@@ -2428,6 +2436,7 @@ _bt_lock_subtree_parent(Relation rel, BlockNumber child, BTStack stack,
                (errcode(ERRCODE_INDEX_CORRUPTED),
                 errmsg_internal("failed to re-find parent key in index \"%s\" for deletion target page %u",
                                 RelationGetRelationName(rel), child)));
+       Assert(false);
        return false;
    }