Don't clear btpo_cycleid during _bt_vacuum_one_page.

When "vacuuming" a single btree page by removing LP_DEAD tuples, we are not
actually within a vacuum operation, but rather in an ordinary insertion
process that could well be running concurrently with a vacuum.  So clearing
the cycleid is incorrect, and could cause the concurrent vacuum to miss
removing tuples that it needs to remove.  This is a longstanding bug
introduced by commit e6284649b9e30372b3990107a082bc7520325676 of
2006-07-25.  I believe it explains Maxim Boguk's recent report of index
corruption, and probably some other previously unexplained reports.

In 9.0 and up this is a one-line fix; before that we need to introduce a
flag to tell _bt_delitems what to do.

diff --git a/src/backend/access/nbtree/nbtinsert.c b/src/backend/access/nbtree/nbtinsert.c
index 2d9c89cb6c1917dcc15500b4d17d809e2e60fc51..df2f1fd7837148f99e3b34afca3c646eb4c088aa 100644 (file)
--- a/src/backend/access/nbtree/nbtinsert.c
+++ b/src/backend/access/nbtree/nbtinsert.c
@@ -1974,7 +1974,7 @@ _bt_vacuum_one_page(Relation rel, Buffer buffer)
        }
 
        if (ndeletable > 0)
-               _bt_delitems(rel, buffer, deletable, ndeletable);
+               _bt_delitems(rel, buffer, deletable, ndeletable, false);
 
        /*
         * Note: if we didn't find any LP_DEAD items, then the page's

diff --git a/src/backend/access/nbtree/nbtpage.c b/src/backend/access/nbtree/nbtpage.c
index d1a496339360321404c28c94ec245748bdce2b4b..a89e3ba84f472cfbf1a155f10733aa6e3e89716b 100644 (file)
--- a/src/backend/access/nbtree/nbtpage.c
+++ b/src/backend/access/nbtree/nbtpage.c
@@ -650,7 +650,8 @@ _bt_page_recyclable(Page page)
  */
 void
 _bt_delitems(Relation rel, Buffer buf,
-                        OffsetNumber *itemnos, int nitems)
+                        OffsetNumber *itemnos, int nitems,
+                        bool inVacuum)
 {
        Page            page = BufferGetPage(buf);
        BTPageOpaque opaque;
@@ -662,11 +663,12 @@ _bt_delitems(Relation rel, Buffer buf,
        PageIndexMultiDelete(page, itemnos, nitems);
 
        /*
-        * We can clear the vacuum cycle ID since this page has certainly been
-        * processed by the current vacuum scan.
+        * If this is within VACUUM, we can clear the vacuum cycle ID since this
+        * page has certainly been processed by the current vacuum scan.
         */
        opaque = (BTPageOpaque) PageGetSpecialPointer(page);
-       opaque->btpo_cycleid = 0;
+       if (inVacuum)
+               opaque->btpo_cycleid = 0;
 
        /*
         * Mark the page as not containing any LP_DEAD items.  This is not

diff --git a/src/backend/access/nbtree/nbtree.c b/src/backend/access/nbtree/nbtree.c
index 8b30331be4b0b2ab6e8a01fec2f5606f6bb5dc3e..6226f2cd3c7db54b776cddafdab81bf0f55f5df4 100644 (file)
--- a/src/backend/access/nbtree/nbtree.c
+++ b/src/backend/access/nbtree/nbtree.c
@@ -889,7 +889,7 @@ restart:
                 */
                if (ndeletable > 0)
                {
-                       _bt_delitems(rel, buf, deletable, ndeletable);
+                       _bt_delitems(rel, buf, deletable, ndeletable, true);
                        stats->tuples_removed += ndeletable;
                        /* must recompute maxoff */
                        maxoff = PageGetMaxOffsetNumber(page);

diff --git a/src/include/access/nbtree.h b/src/include/access/nbtree.h
index fc8b30b2ea91a001798577c8e0f2a5af0e43a916..29024ce7e557ffc7944ef8460678dd7726537101 100644 (file)
--- a/src/include/access/nbtree.h
+++ b/src/include/access/nbtree.h
@@ -539,7 +539,7 @@ extern void _bt_relbuf(Relation rel, Buffer buf);
 extern void _bt_pageinit(Page page, Size size);
 extern bool _bt_page_recyclable(Page page);
 extern void _bt_delitems(Relation rel, Buffer buf,
-                        OffsetNumber *itemnos, int nitems);
+                        OffsetNumber *itemnos, int nitems, bool inVacuum);
 extern int _bt_pagedel(Relation rel, Buffer buf,
                        BTStack stack, bool vacuum_full);