Allow VACUUM to be run with index cleanup disabled.

This commit adds a new reloption, vacuum_index_cleanup, which
controls whether index cleanup is performed for a particular
relation by default.  It also adds a new option to the VACUUM
command, INDEX_CLEANUP, which can be used to override the
reloption.  If neither the reloption nor the VACUUM option is
used, the default is true, as before.

Masahiko Sawada, reviewed and tested by Nathan Bossart, Alvaro
Herrera, Kyotaro Horiguchi, Darafei Praliaskouski, and me.
The wording of the documentation is mostly due to me.

Discussion: http://postgr.es/m/CAD21AoAt5R3DNUZSjOoXDUY=naYPUOuffVsRzuTYMz29yLzQCA@mail.gmail.com
This commit is contained in:
Robert Haas 2019-04-04 14:58:53 -04:00
parent 74eb2176bf
commit a96c41feec
11 changed files with 193 additions and 25 deletions

View File

@ -1389,6 +1389,21 @@ WITH ( MODULUS <replaceable class="parameter">numeric_literal</replaceable>, REM
</listitem>
</varlistentry>
<varlistentry>
<term><literal>vacuum_index_cleanup</literal> (<type>boolean</type>)</term>
<listitem>
<para>
Enables or disables index cleanup when <command>VACUUM</command> is
run on this table. The default value is <literal>true</literal>.
Disabling index cleanup can speed up <command>VACUUM</command> very
significantly, but may also lead to severely bloated indexes if table
modifications are frequent. The <literal>INDEX_CLEANUP</literal>
parameter to <xref linkend="sql-vacuum"/>, if specified, overrides
the value of this option.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><literal>autovacuum_vacuum_threshold</literal>, <literal>toast.autovacuum_vacuum_threshold</literal> (<type>integer</type>)</term>
<listitem>

View File

@ -32,6 +32,7 @@ VACUUM [ FULL ] [ FREEZE ] [ VERBOSE ] [ ANALYZE ] [ <replaceable class="paramet
ANALYZE [ <replaceable class="parameter">boolean</replaceable> ]
DISABLE_PAGE_SKIPPING [ <replaceable class="parameter">boolean</replaceable> ]
SKIP_LOCKED [ <replaceable class="parameter">boolean</replaceable> ]
INDEX_CLEANUP [ <replaceable class="parameter">boolean</replaceable> ]
<phrase>and <replaceable class="parameter">table_and_columns</replaceable> is:</phrase>
@ -181,6 +182,28 @@ VACUUM [ FULL ] [ FREEZE ] [ VERBOSE ] [ ANALYZE ] [ <replaceable class="paramet
</listitem>
</varlistentry>
<varlistentry>
<term><literal>INDEX_CLEANUP</literal></term>
<listitem>
<para>
Specifies that <command>VACUUM</command> should attempt to remove
index entries pointing to dead tuples. This is normally the desired
behavior and is the default unless the
<literal>vacuum_index_cleanup</literal> option has been set to false
for the table to be vacuumed. Setting this option to false may be
useful when it is necessary to make vacuum run as quickly as possible,
for example to avoid imminent transaction ID wraparound
(see <xref linkend="vacuum-for-wraparound"/>). However, if index
cleanup is not performed regularly, performance may suffer, because
as the table is modified, indexes will accumulate dead tuples
and the table itself will accumulate dead line pointers that cannot be
removed until index cleanup is completed. This option has no effect
for tables that do not have an index and is ignored if the
<literal>FULL</literal> is used.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term><replaceable class="parameter">boolean</replaceable></term>
<listitem>

View File

@ -138,6 +138,15 @@ static relopt_bool boolRelOpts[] =
},
false
},
{
{
"vacuum_index_cleanup",
"Enables index vacuuming and index cleanup",
RELOPT_KIND_HEAP,
ShareUpdateExclusiveLock
},
true
},
/* list terminator */
{{NULL}}
};
@ -1388,7 +1397,9 @@ default_reloptions(Datum reloptions, bool validate, relopt_kind kind)
{"parallel_workers", RELOPT_TYPE_INT,
offsetof(StdRdOptions, parallel_workers)},
{"vacuum_cleanup_index_scale_factor", RELOPT_TYPE_REAL,
offsetof(StdRdOptions, vacuum_cleanup_index_scale_factor)}
offsetof(StdRdOptions, vacuum_cleanup_index_scale_factor)},
{"vacuum_index_cleanup", RELOPT_TYPE_BOOL,
offsetof(StdRdOptions, vacuum_index_cleanup)}
};
options = parseRelOptions(reloptions, validate, kind, &numoptions);

View File

@ -112,8 +112,8 @@
typedef struct LVRelStats
{
/* hasindex = true means two-pass strategy; false means one-pass */
bool hasindex;
/* useindex = true means two-pass strategy; false means one-pass */
bool useindex;
/* Overall statistics about rel */
BlockNumber old_rel_pages; /* previous value of pg_class.relpages */
BlockNumber rel_pages; /* total number of pages */
@ -125,6 +125,8 @@ typedef struct LVRelStats
double new_rel_tuples; /* new estimated total # of tuples */
double new_live_tuples; /* new estimated total # of live tuples */
double new_dead_tuples; /* new estimated total # of dead tuples */
double nleft_dead_tuples; /* # of dead tuples we left */
double nleft_dead_itemids; /* # of dead item pointers we left */
BlockNumber pages_removed;
double tuples_deleted;
BlockNumber nonempty_pages; /* actually, last nonempty page + 1 */
@ -150,7 +152,7 @@ static BufferAccessStrategy vac_strategy;
/* non-export function prototypes */
static void lazy_scan_heap(Relation onerel, int options,
static void lazy_scan_heap(Relation onerel, VacuumParams *params,
LVRelStats *vacrelstats, Relation *Irel, int nindexes,
bool aggressive);
static void lazy_vacuum_heap(Relation onerel, LVRelStats *vacrelstats, BlockNumber nblocks);
@ -209,6 +211,7 @@ heap_vacuum_rel(Relation onerel, VacuumParams *params,
MultiXactId new_min_multi;
Assert(params != NULL);
Assert(params->index_cleanup != VACOPT_TERNARY_DEFAULT);
/* measure elapsed time iff autovacuum logging requires it */
if (IsAutoVacuumWorkerProcess() && params->log_min_duration >= 0)
@ -275,10 +278,11 @@ heap_vacuum_rel(Relation onerel, VacuumParams *params,
/* Open all indexes of the relation */
vac_open_indexes(onerel, RowExclusiveLock, &nindexes, &Irel);
vacrelstats->hasindex = (nindexes > 0);
vacrelstats->useindex = (nindexes > 0 &&
params->index_cleanup == VACOPT_TERNARY_ENABLED);
/* Do the vacuuming */
lazy_scan_heap(onerel, params->options, vacrelstats, Irel, nindexes, aggressive);
lazy_scan_heap(onerel, params, vacrelstats, Irel, nindexes, aggressive);
/* Done with indexes */
vac_close_indexes(nindexes, Irel, NoLock);
@ -349,7 +353,7 @@ heap_vacuum_rel(Relation onerel, VacuumParams *params,
new_rel_pages,
new_live_tuples,
new_rel_allvisible,
vacrelstats->hasindex,
nindexes > 0,
new_frozen_xid,
new_min_multi,
false);
@ -419,6 +423,12 @@ heap_vacuum_rel(Relation onerel, VacuumParams *params,
vacrelstats->new_rel_tuples,
vacrelstats->new_dead_tuples,
OldestXmin);
if (vacrelstats->nleft_dead_tuples > 0 ||
vacrelstats->nleft_dead_itemids > 0)
appendStringInfo(&buf,
_("%.0f tuples and %.0f item identifiers are left as dead.\n"),
vacrelstats->nleft_dead_tuples,
vacrelstats->nleft_dead_itemids);
appendStringInfo(&buf,
_("buffer usage: %d hits, %d misses, %d dirtied\n"),
VacuumPageHit,
@ -485,7 +495,7 @@ vacuum_log_cleanup_info(Relation rel, LVRelStats *vacrelstats)
* reference them have been killed.
*/
static void
lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
lazy_scan_heap(Relation onerel, VacuumParams *params, LVRelStats *vacrelstats,
Relation *Irel, int nindexes, bool aggressive)
{
BlockNumber nblocks,
@ -501,7 +511,10 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
live_tuples, /* live tuples (reltuples estimate) */
tups_vacuumed, /* tuples cleaned up by vacuum */
nkeep, /* dead-but-not-removable tuples */
nunused; /* unused item pointers */
nunused, /* unused item pointers */
nleft_dead_tuples, /* tuples we left as dead */
nleft_dead_itemids; /* item pointers we left as dead,
* includes nleft_dead_tuples. */
IndexBulkDeleteResult **indstats;
int i;
PGRUsage ru0;
@ -534,6 +547,7 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
empty_pages = vacuumed_pages = 0;
next_fsm_block_to_vacuum = (BlockNumber) 0;
num_tuples = live_tuples = tups_vacuumed = nkeep = nunused = 0;
nleft_dead_itemids = nleft_dead_tuples = 0;
indstats = (IndexBulkDeleteResult **)
palloc0(nindexes * sizeof(IndexBulkDeleteResult *));
@ -599,7 +613,7 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
* be replayed on any hot standby, where it can be disruptive.
*/
next_unskippable_block = 0;
if ((options & VACOPT_DISABLE_PAGE_SKIPPING) == 0)
if ((params->options & VACOPT_DISABLE_PAGE_SKIPPING) == 0)
{
while (next_unskippable_block < nblocks)
{
@ -654,7 +668,7 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
{
/* Time to advance next_unskippable_block */
next_unskippable_block++;
if ((options & VACOPT_DISABLE_PAGE_SKIPPING) == 0)
if ((params->options & VACOPT_DISABLE_PAGE_SKIPPING) == 0)
{
while (next_unskippable_block < nblocks)
{
@ -1070,7 +1084,17 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
HeapTupleIsHeapOnly(&tuple))
nkeep += 1;
else
{
tupgone = true; /* we can delete the tuple */
/*
* Since this dead tuple will not be vacuumed and
* ignored when index cleanup is disabled we count
* count it for reporting.
*/
if (params->index_cleanup == VACOPT_TERNARY_ENABLED)
nleft_dead_tuples++;
}
all_visible = false;
break;
case HEAPTUPLE_LIVE:
@ -1222,15 +1246,33 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
}
/*
* If there are no indexes then we can vacuum the page right now
* instead of doing a second scan.
* If there are no indexes we can vacuum the page right now instead of
* doing a second scan. Also we don't do that but forget dead tuples
* when index cleanup is disabled.
*/
if (nindexes == 0 &&
vacrelstats->num_dead_tuples > 0)
if (!vacrelstats->useindex && vacrelstats->num_dead_tuples > 0)
{
/* Remove tuples from heap */
lazy_vacuum_page(onerel, blkno, buf, 0, vacrelstats, &vmbuffer);
has_dead_tuples = false;
if (nindexes == 0)
{
/* Remove tuples from heap if the table has no index */
lazy_vacuum_page(onerel, blkno, buf, 0, vacrelstats, &vmbuffer);
vacuumed_pages++;
has_dead_tuples = false;
}
else
{
/*
* Here, we have indexes but index cleanup is disabled. Instead of
* vacuuming the dead tuples on the heap, we just forget them.
*
* Note that vacrelstats->dead_tuples could have tuples which
* became dead after HOT-pruning but are not marked dead yet.
* We do not process them because it's a very rare condition, and
* the next vacuum will process them anyway.
*/
Assert(params->index_cleanup == VACOPT_TERNARY_DISABLED);
nleft_dead_itemids += vacrelstats->num_dead_tuples;
}
/*
* Forget the now-vacuumed tuples, and press on, but be careful
@ -1238,7 +1280,6 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
* valid.
*/
vacrelstats->num_dead_tuples = 0;
vacuumed_pages++;
/*
* Periodically do incremental FSM vacuuming to make newly-freed
@ -1357,6 +1398,11 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
RecordPageWithFreeSpace(onerel, blkno, freespace, nblocks);
}
/* No dead tuples should be left if index cleanup is enabled */
Assert((params->index_cleanup == VACOPT_TERNARY_ENABLED &&
nleft_dead_tuples == 0 && nleft_dead_itemids == 0) ||
params->index_cleanup == VACOPT_TERNARY_DISABLED);
/* report that everything is scanned and vacuumed */
pgstat_progress_update_param(PROGRESS_VACUUM_HEAP_BLKS_SCANNED, blkno);
@ -1364,7 +1410,9 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
/* save stats for use later */
vacrelstats->tuples_deleted = tups_vacuumed;
vacrelstats->new_dead_tuples = nkeep;
vacrelstats->new_dead_tuples = nkeep + nleft_dead_tuples;
vacrelstats->nleft_dead_tuples = nleft_dead_tuples;
vacrelstats->nleft_dead_itemids = nleft_dead_itemids;
/* now we can compute the new value for pg_class.reltuples */
vacrelstats->new_live_tuples = vac_estimate_reltuples(onerel,
@ -1433,8 +1481,11 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
PROGRESS_VACUUM_PHASE_INDEX_CLEANUP);
/* Do post-vacuum cleanup and statistics update for each index */
for (i = 0; i < nindexes; i++)
lazy_cleanup_index(Irel[i], indstats[i], vacrelstats);
if (vacrelstats->useindex)
{
for (i = 0; i < nindexes; i++)
lazy_cleanup_index(Irel[i], indstats[i], vacrelstats);
}
/* If no indexes, make log report that lazy_vacuum_heap would've made */
if (vacuumed_pages)
@ -1465,6 +1516,8 @@ lazy_scan_heap(Relation onerel, int options, LVRelStats *vacrelstats,
"%u pages are entirely empty.\n",
empty_pages),
empty_pages);
appendStringInfo(&buf, "%.0f tuples and %.0f item identifiers are left as dead.\n",
nleft_dead_tuples, nleft_dead_itemids);
appendStringInfo(&buf, _("%s."), pg_rusage_show(&ru0));
ereport(elevel,
@ -2110,7 +2163,7 @@ lazy_space_alloc(LVRelStats *vacrelstats, BlockNumber relblocks)
autovacuum_work_mem != -1 ?
autovacuum_work_mem : maintenance_work_mem;
if (vacrelstats->hasindex)
if (vacrelstats->useindex)
{
maxtuples = (vac_work_mem * 1024L) / sizeof(ItemPointerData);
maxtuples = Min(maxtuples, INT_MAX);

View File

@ -76,6 +76,7 @@ static void vac_truncate_clog(TransactionId frozenXID,
TransactionId lastSaneFrozenXid,
MultiXactId lastSaneMinMulti);
static bool vacuum_rel(Oid relid, RangeVar *relation, VacuumParams *params);
static VacOptTernaryValue get_vacopt_ternary_value(DefElem *def);
/*
* Primary entry point for manual VACUUM and ANALYZE commands
@ -95,6 +96,9 @@ ExecVacuum(ParseState *pstate, VacuumStmt *vacstmt, bool isTopLevel)
bool disable_page_skipping = false;
ListCell *lc;
/* Set default value */
params.index_cleanup = VACOPT_TERNARY_DEFAULT;
/* Parse options list */
foreach(lc, vacstmt->options)
{
@ -120,6 +124,8 @@ ExecVacuum(ParseState *pstate, VacuumStmt *vacstmt, bool isTopLevel)
full = defGetBoolean(opt);
else if (strcmp(opt->defname, "disable_page_skipping") == 0)
disable_page_skipping = defGetBoolean(opt);
else if (strcmp(opt->defname, "index_cleanup") == 0)
params.index_cleanup = get_vacopt_ternary_value(opt);
else
ereport(ERROR,
(errcode(ERRCODE_SYNTAX_ERROR),
@ -1719,6 +1725,16 @@ vacuum_rel(Oid relid, RangeVar *relation, VacuumParams *params)
onerelid = onerel->rd_lockInfo.lockRelId;
LockRelationIdForSession(&onerelid, lmode);
/* Set index cleanup option based on reloptions if not yet */
if (params->index_cleanup == VACOPT_TERNARY_DEFAULT)
{
if (onerel->rd_options == NULL ||
((StdRdOptions *) onerel->rd_options)->vacuum_index_cleanup)
params->index_cleanup = VACOPT_TERNARY_ENABLED;
else
params->index_cleanup = VACOPT_TERNARY_DISABLED;
}
/*
* Remember the relation's TOAST relation for later, if the caller asked
* us to process it. In VACUUM FULL, though, the toast table is
@ -1899,3 +1915,15 @@ vacuum_delay_point(void)
CHECK_FOR_INTERRUPTS();
}
}
/*
* A wrapper function of defGetBoolean().
*
* This function returns VACOPT_TERNARY_ENABLED and VACOPT_TERNARY_DISABLED
* instead of true and false.
*/
static VacOptTernaryValue
get_vacopt_ternary_value(DefElem *def)
{
return defGetBoolean(def) ? VACOPT_TERNARY_ENABLED : VACOPT_TERNARY_DISABLED;
}

View File

@ -2886,6 +2886,7 @@ table_recheck_autovac(Oid relid, HTAB *table_toast_map,
(dovacuum ? VACOPT_VACUUM : 0) |
(doanalyze ? VACOPT_ANALYZE : 0) |
(!wraparound ? VACOPT_SKIP_LOCKED : 0);
tab->at_params.index_cleanup = VACOPT_TERNARY_DEFAULT;
tab->at_params.freeze_min_age = freeze_min_age;
tab->at_params.freeze_table_age = freeze_table_age;
tab->at_params.multixact_freeze_min_age = multixact_freeze_min_age;

View File

@ -1039,6 +1039,7 @@ static const char *const table_storage_parameters[] = {
"toast.log_autovacuum_min_duration",
"toast_tuple_target",
"user_catalog_table",
"vacuum_index_cleanup",
NULL
};
@ -3443,8 +3444,9 @@ psql_completion(const char *text, int start, int end)
*/
if (ends_with(prev_wd, '(') || ends_with(prev_wd, ','))
COMPLETE_WITH("FULL", "FREEZE", "ANALYZE", "VERBOSE",
"DISABLE_PAGE_SKIPPING", "SKIP_LOCKED");
else if (TailMatches("FULL|FREEZE|ANALYZE|VERBOSE|DISABLE_PAGE_SKIPPING|SKIP_LOCKED"))
"DISABLE_PAGE_SKIPPING", "SKIP_LOCKED",
"INDEX_CLEANUP");
else if (TailMatches("FULL|FREEZE|ANALYZE|VERBOSE|DISABLE_PAGE_SKIPPING|SKIP_LOCKED|INDEX_CLEANUP"))
COMPLETE_WITH("ON", "OFF");
}
else if (HeadMatches("VACUUM") && TailMatches("("))

View File

@ -148,6 +148,19 @@ typedef enum VacuumOption
VACOPT_DISABLE_PAGE_SKIPPING = 1 << 7 /* don't skip any pages */
} VacuumOption;
/*
* A ternary value used by vacuum parameters.
*
* DEFAULT value is used to determine the value based on other
* configurations, e.g. reloptions.
*/
typedef enum VacOptTernaryValue
{
VACOPT_TERNARY_DEFAULT = 0,
VACOPT_TERNARY_DISABLED,
VACOPT_TERNARY_ENABLED,
} VacOptTernaryValue;
/*
* Parameters customizing behavior of VACUUM and ANALYZE.
*
@ -167,6 +180,8 @@ typedef struct VacuumParams
int log_min_duration; /* minimum execution threshold in ms at
* which verbose logs are activated, -1
* to use default */
VacOptTernaryValue index_cleanup; /* Do index vacuum and cleanup,
* default value depends on reloptions */
} VacuumParams;
/* GUC parameters */

View File

@ -266,6 +266,7 @@ typedef struct StdRdOptions
AutoVacOpts autovacuum; /* autovacuum-related options */
bool user_catalog_table; /* use as an additional catalog relation */
int parallel_workers; /* max number of parallel workers */
bool vacuum_index_cleanup; /* enables index vacuuming and cleanup */
} StdRdOptions;
#define HEAP_MIN_FILLFACTOR 10

View File

@ -80,6 +80,14 @@ CONTEXT: SQL function "do_analyze" statement 1
SQL function "wrap_do_analyze" statement 1
VACUUM FULL vactst;
VACUUM (DISABLE_PAGE_SKIPPING) vaccluster;
-- INDEX_CLEANUP option
CREATE TABLE no_index_cleanup (i INT PRIMARY KEY) WITH (vacuum_index_cleanup = false);
VACUUM (INDEX_CLEANUP FALSE) vaccluster;
VACUUM (INDEX_CLEANUP FALSE) vactst; -- index cleanup option is ignored if no indexes
VACUUM (INDEX_CLEANUP FALSE, FREEZE TRUE) vaccluster;
-- index cleanup option is ignored if VACUUM FULL
VACUUM (INDEX_CLEANUP TRUE, FULL TRUE) no_index_cleanup;
VACUUM (FULL TRUE) no_index_cleanup;
-- partitioned table
CREATE TABLE vacparted (a int, b char) PARTITION BY LIST (a);
CREATE TABLE vacparted1 PARTITION OF vacparted FOR VALUES IN (1);
@ -136,6 +144,7 @@ ANALYZE (SKIP_LOCKED) vactst;
DROP TABLE vaccluster;
DROP TABLE vactst;
DROP TABLE vacparted;
DROP TABLE no_index_cleanup;
-- relation ownership, WARNING logs generated as all are skipped.
CREATE TABLE vacowned (a int);
CREATE TABLE vacowned_parted (a int) PARTITION BY LIST (a);

View File

@ -62,6 +62,15 @@ VACUUM FULL vactst;
VACUUM (DISABLE_PAGE_SKIPPING) vaccluster;
-- INDEX_CLEANUP option
CREATE TABLE no_index_cleanup (i INT PRIMARY KEY) WITH (vacuum_index_cleanup = false);
VACUUM (INDEX_CLEANUP FALSE) vaccluster;
VACUUM (INDEX_CLEANUP FALSE) vactst; -- index cleanup option is ignored if no indexes
VACUUM (INDEX_CLEANUP FALSE, FREEZE TRUE) vaccluster;
-- index cleanup option is ignored if VACUUM FULL
VACUUM (INDEX_CLEANUP TRUE, FULL TRUE) no_index_cleanup;
VACUUM (FULL TRUE) no_index_cleanup;
-- partitioned table
CREATE TABLE vacparted (a int, b char) PARTITION BY LIST (a);
CREATE TABLE vacparted1 PARTITION OF vacparted FOR VALUES IN (1);
@ -107,6 +116,7 @@ ANALYZE (SKIP_LOCKED) vactst;
DROP TABLE vaccluster;
DROP TABLE vactst;
DROP TABLE vacparted;
DROP TABLE no_index_cleanup;
-- relation ownership, WARNING logs generated as all are skipped.
CREATE TABLE vacowned (a int);