database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-15 19:21:59 +03:00
Code Activity

Fix vacuum so that autovacuum is really not cancelled when doing an emergency

Browse Source 
job (i.e. to prevent Xid wraparound problems.)  Bug reported by ITAGAKI
Takahiro in 20080314103837.63D3.52131E4D@oss.ntt.co.jp, though I didn't use his
patch.
This commit is contained in:
Alvaro Herrera
2008-03-14 17:25:59 +00:00
parent be5d6df346
commit adc4e1e635
4 changed files with 26 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
src/backend/commands/vacuum.c
View File

@ -13,7 +13,7 @@
*
*
* IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/commands/vacuum.c,v 1.366 2008/03/10 02:04:08 tgl Exp $
* $PostgreSQL: pgsql/src/backend/commands/vacuum.c,v 1.367 2008/03/14 17:25:58 alvherre Exp $
*
*-------------------------------------------------------------------------
*/
@ -209,7 +209,8 @@ static BufferAccessStrategy vac_strategy;
static List *get_rel_oids(List *relids, const RangeVar *vacrel,
const char *stmttype);
static void vac_truncate_clog(TransactionId frozenXID);
static void vacuum_rel(Oid relid, VacuumStmt *vacstmt, char expected_relkind);
static void vacuum_rel(Oid relid, VacuumStmt *vacstmt, char expected_relkind,
bool for_wraparound);
static void full_vacuum_rel(Relation onerel, VacuumStmt *vacstmt);
static void scan_heap(VRelStats *vacrelstats, Relation onerel,
VacPageList vacuum_pages, VacPageList fraged_pages);
@ -263,6 +264,9 @@ static Size PageGetFreeSpaceWithFillFactor(Relation relation, Page page);
* relation OIDs to be processed, and vacstmt->relation is ignored.
* (The non-NIL case is currently only used by autovacuum.)
*
* for_wraparound is used by autovacuum to let us know when it's forcing
* a vacuum for wraparound, which should not be auto-cancelled.
*
* bstrategy is normally given as NULL, but in autovacuum it can be passed
* in to use the same buffer strategy object across multiple vacuum() calls.
*
@ -274,7 +278,7 @@ static Size PageGetFreeSpaceWithFillFactor(Relation relation, Page page);
*/
void
vacuum(VacuumStmt *vacstmt, List *relids,
BufferAccessStrategy bstrategy, bool isTopLevel)
BufferAccessStrategy bstrategy, bool for_wraparound, bool isTopLevel)
{
const char *stmttype = vacstmt->vacuum ? "VACUUM" : "ANALYZE";
volatile MemoryContext anl_context = NULL;
@ -421,7 +425,7 @@ vacuum(VacuumStmt *vacstmt, List *relids,
Oid relid = lfirst_oid(cur);
if (vacstmt->vacuum)
vacuum_rel(relid, vacstmt, RELKIND_RELATION);
vacuum_rel(relid, vacstmt, RELKIND_RELATION, for_wraparound);
if (vacstmt->analyze)
{
@ -966,7 +970,8 @@ vac_truncate_clog(TransactionId frozenXID)
* At entry and exit, we are not inside a transaction.
*/
static void
vacuum_rel(Oid relid, VacuumStmt *vacstmt, char expected_relkind)
vacuum_rel(Oid relid, VacuumStmt *vacstmt, char expected_relkind,
bool for_wraparound)
{
LOCKMODE lmode;
Relation onerel;
@ -999,6 +1004,10 @@ vacuum_rel(Oid relid, VacuumStmt *vacstmt, char expected_relkind)
* contents of other tables is arguably broken, but we won't break it
* here by violating transaction semantics.)
*
* We also set the VACUUM_FOR_WRAPAROUND flag, which is passed down
* by autovacuum; it's used to avoid cancelling a vacuum that was
* invoked in an emergency.
*
* Note: this flag remains set until CommitTransaction or
* AbortTransaction. We don't want to clear it until we reset
* MyProc->xid/xmin, else OldestXmin might appear to go backwards,
@ -1006,6 +1015,8 @@ vacuum_rel(Oid relid, VacuumStmt *vacstmt, char expected_relkind)
*/
LWLockAcquire(ProcArrayLock, LW_EXCLUSIVE);
MyProc->vacuumFlags |= PROC_IN_VACUUM;
if (for_wraparound)
MyProc->vacuumFlags |= PROC_VACUUM_FOR_WRAPAROUND;
LWLockRelease(ProcArrayLock);
}
@ -1147,7 +1158,7 @@ vacuum_rel(Oid relid, VacuumStmt *vacstmt, char expected_relkind)
* totally unimportant for toast relations.
*/
if (toast_relid != InvalidOid)
vacuum_rel(toast_relid, vacstmt, RELKIND_TOASTVALUE);
vacuum_rel(toast_relid, vacstmt, RELKIND_TOASTVALUE, for_wraparound);
/*
* Now release the session-level lock on the master table.