Alvaro Herrera <alvherre@xxxxxxxxxxxxxxx> writes:
> These are all shared catalogs. There are others, so you may still see
> more. We got another report for pg_database
> https://www.postgresql.org/message-id/A9D40BB7-CFD6-46AF-A0A1-249F04878A2A%40amazon.com
> so I suppose there really is a bug. I don't know what's going on there.
I think it's pretty obvious: autovacuum.c's rule for detecting whether
some other worker is already processing table X is wrong when X is a
shared table. I propose the attached patch.
regards, tom lane
diff --git a/src/backend/postmaster/autovacuum.c b/src/backend/postmaster/autovacuum.c
index e2859df..4bc5d42 100644
*** a/src/backend/postmaster/autovacuum.c
--- b/src/backend/postmaster/autovacuum.c
*************** typedef struct autovac_table
*** 203,215 ****
* wi_links entry into free list or running list
* wi_dboid OID of the database this worker is supposed to work on
* wi_tableoid OID of the table currently being vacuumed, if any
* wi_proc pointer to PGPROC of the running worker, NULL if not started
* wi_launchtime Time at which this worker was launched
* wi_cost_* Vacuum cost-based delay parameters current in this worker
*
! * All fields are protected by AutovacuumLock, except for wi_tableoid which is
! * protected by AutovacuumScheduleLock (which is read-only for everyone except
! * that worker itself).
*-------------
*/
typedef struct WorkerInfoData
--- 203,216 ----
* wi_links entry into free list or running list
* wi_dboid OID of the database this worker is supposed to work on
* wi_tableoid OID of the table currently being vacuumed, if any
+ * wi_tableshared true if the table currently being vacuumed is a shared rel
* wi_proc pointer to PGPROC of the running worker, NULL if not started
* wi_launchtime Time at which this worker was launched
* wi_cost_* Vacuum cost-based delay parameters current in this worker
*
! * All fields are protected by AutovacuumLock, except for wi_tableoid and
! * wi_tableshared which are protected by AutovacuumScheduleLock (and are
! * read-only for everyone except that worker itself).
*-------------
*/
typedef struct WorkerInfoData
*************** typedef struct WorkerInfoData
*** 217,222 ****
--- 218,224 ----
dlist_node wi_links;
Oid wi_dboid;
Oid wi_tableoid;
+ bool wi_tableshared;
PGPROC *wi_proc;
TimestampTz wi_launchtime;
bool wi_dobalance;
*************** autovac_balance_cost(void)
*** 1791,1798 ****
}
if (worker->wi_proc != NULL)
! elog(DEBUG2, "autovac_balance_cost(pid=%u db=%u, rel=%u, dobalance=%s cost_limit=%d, cost_limit_base=%d, cost_delay=%d)",
worker->wi_proc->pid, worker->wi_dboid, worker->wi_tableoid,
worker->wi_dobalance ? "yes" : "no",
worker->wi_cost_limit, worker->wi_cost_limit_base,
worker->wi_cost_delay);
--- 1793,1801 ----
}
if (worker->wi_proc != NULL)
! elog(DEBUG2, "autovac_balance_cost(pid=%u db=%u, rel=%u%s, dobalance=%s cost_limit=%d, cost_limit_base=%d, cost_delay=%d)",
worker->wi_proc->pid, worker->wi_dboid, worker->wi_tableoid,
+ worker->wi_tableshared ? " (shared)" : "",
worker->wi_dobalance ? "yes" : "no",
worker->wi_cost_limit, worker->wi_cost_limit_base,
worker->wi_cost_delay);
*************** do_autovacuum(void)
*** 1885,1893 ****
HeapScanDesc relScan;
Form_pg_database dbForm;
List *table_oids = NIL;
HASHCTL ctl;
HTAB *table_toast_map;
! ListCell *volatile cell;
PgStat_StatDBEntry *shared;
PgStat_StatDBEntry *dbentry;
BufferAccessStrategy bstrategy;
--- 1888,1898 ----
HeapScanDesc relScan;
Form_pg_database dbForm;
List *table_oids = NIL;
+ List *table_shares = NIL;
HASHCTL ctl;
HTAB *table_toast_map;
! ListCell *volatile lco;
! ListCell *volatile lcs;
PgStat_StatDBEntry *shared;
PgStat_StatDBEntry *dbentry;
BufferAccessStrategy bstrategy;
*************** do_autovacuum(void)
*** 2004,2009 ****
--- 2009,2015 ----
PgStat_StatTabEntry *tabentry;
AutoVacOpts *relopts;
Oid relid;
+ bool relisshared;
bool dovacuum;
bool doanalyze;
bool wraparound;
*************** do_autovacuum(void)
*** 2013,2022 ****
continue;
relid = HeapTupleGetOid(tuple);
/* Fetch reloptions and the pgstat entry for this table */
relopts = extract_autovac_opts(tuple, pg_class_desc);
! tabentry = get_pgstat_tabentry_relid(relid, classForm->relisshared,
shared, dbentry);
/* Check if it needs vacuum or analyze */
--- 2019,2029 ----
continue;
relid = HeapTupleGetOid(tuple);
+ relisshared = classForm->relisshared;
/* Fetch reloptions and the pgstat entry for this table */
relopts = extract_autovac_opts(tuple, pg_class_desc);
! tabentry = get_pgstat_tabentry_relid(relid, relisshared,
shared, dbentry);
/* Check if it needs vacuum or analyze */
*************** do_autovacuum(void)
*** 2069,2077 ****
}
else
{
! /* relations that need work are added to table_oids */
if (dovacuum || doanalyze)
table_oids = lappend_oid(table_oids, relid);
/*
* Remember the association for the second pass. Note: we must do
--- 2076,2087 ----
}
else
{
! /* relations that need work are added to table_oids/table_shares */
if (dovacuum || doanalyze)
+ {
table_oids = lappend_oid(table_oids, relid);
+ table_shares = lappend_int(table_shares, relisshared);
+ }
/*
* Remember the association for the second pass. Note: we must do
*************** do_autovacuum(void)
*** 2117,2122 ****
--- 2127,2133 ----
Form_pg_class classForm = (Form_pg_class) GETSTRUCT(tuple);
PgStat_StatTabEntry *tabentry;
Oid relid;
+ bool relisshared;
AutoVacOpts *relopts = NULL;
bool dovacuum;
bool doanalyze;
*************** do_autovacuum(void)
*** 2129,2134 ****
--- 2140,2146 ----
continue;
relid = HeapTupleGetOid(tuple);
+ relisshared = classForm->relisshared;
/*
* fetch reloptions -- if this toast table does not have them, try the
*************** do_autovacuum(void)
*** 2146,2152 ****
}
/* Fetch the pgstat entry for this table */
! tabentry = get_pgstat_tabentry_relid(relid, classForm->relisshared,
shared, dbentry);
relation_needs_vacanalyze(relid, relopts, classForm, tabentry,
--- 2158,2164 ----
}
/* Fetch the pgstat entry for this table */
! tabentry = get_pgstat_tabentry_relid(relid, relisshared,
shared, dbentry);
relation_needs_vacanalyze(relid, relopts, classForm, tabentry,
*************** do_autovacuum(void)
*** 2155,2161 ****
--- 2167,2176 ----
/* ignore analyze for toast tables */
if (dovacuum)
+ {
table_oids = lappend_oid(table_oids, relid);
+ table_shares = lappend_int(table_shares, relisshared);
+ }
}
heap_endscan(relScan);
*************** do_autovacuum(void)
*** 2181,2189 ****
/*
* Perform operations on collected tables.
*/
! foreach(cell, table_oids)
{
! Oid relid = lfirst_oid(cell);
autovac_table *tab;
bool skipit;
int stdVacuumCostDelay;
--- 2196,2205 ----
/*
* Perform operations on collected tables.
*/
! forboth(lco, table_oids, lcs, table_shares)
{
! Oid relid = lfirst_oid(lco);
! bool relisshared = (bool) lfirst_int(lcs);
autovac_table *tab;
bool skipit;
int stdVacuumCostDelay;
*************** do_autovacuum(void)
*** 2229,2243 ****
if (worker == MyWorkerInfo)
continue;
! /* ignore workers in other databases */
! if (worker->wi_dboid != MyDatabaseId)
continue;
! if (worker->wi_tableoid == relid)
! {
! skipit = true;
! break;
! }
}
LWLockRelease(AutovacuumLock);
if (skipit)
--- 2245,2261 ----
if (worker == MyWorkerInfo)
continue;
! /* not a match if tableoid or shared-rel flag don't match */
! if (worker->wi_tableoid != relid ||
! worker->wi_tableshared != relisshared)
continue;
! /* for unshared table, not a match unless same database */
! if (!relisshared && worker->wi_dboid != MyDatabaseId)
! continue;
!
! skipit = true;
! break;
}
LWLockRelease(AutovacuumLock);
if (skipit)
*************** do_autovacuum(void)
*** 2271,2276 ****
--- 2289,2295 ----
* the lock so that other workers don't vacuum it concurrently.
*/
MyWorkerInfo->wi_tableoid = relid;
+ MyWorkerInfo->wi_tableshared = relisshared;
LWLockRelease(AutovacuumScheduleLock);
/*
--
Sent via pgsql-admin mailing list (pgsql-admin@xxxxxxxxxxxxxx)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin
