rdelaage
/
postgresql
mirror of https://git.postgresql.org/git/postgresql.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Set relfrozenxid to oldest extant XID seen by VACUUM. 

When VACUUM set relfrozenxid before now, it set it to whatever value was
used to determine which tuples to freeze -- the FreezeLimit cutoff.
This approach was very naive.  The relfrozenxid invariant only requires
that new relfrozenxid values be <= the oldest extant XID remaining in
the table (at the point that the VACUUM operation ends), which in
general might be much more recent than FreezeLimit.

VACUUM now carefully tracks the oldest remaining XID/MultiXactId as it
goes (the oldest remaining values _after_ lazy_scan_prune processing).
The final values are set as the table's new relfrozenxid and new
relminmxid in pg_class at the end of each VACUUM.  The oldest XID might
come from a tuple's xmin, xmax, or xvac fields.  It might even come from
one of the table's remaining MultiXacts.

Final relfrozenxid values must still be >= FreezeLimit in an aggressive
VACUUM (FreezeLimit still acts as a lower bound on the final value that
aggressive VACUUM can set relfrozenxid to).  Since standard VACUUMs
still make no guarantees about advancing relfrozenxid, they might as
well set relfrozenxid to a value from well before FreezeLimit when the
opportunity presents itself.  In general standard VACUUMs may now set
relfrozenxid to any value > the original relfrozenxid and <= OldestXmin.

Credit for the general idea of using the oldest extant XID to set
pg_class.relfrozenxid at the end of VACUUM goes to Andres Freund.

Author: Peter Geoghegan <pg@bowt.ie>
Reviewed-By: Andres Freund <andres@anarazel.de>
Reviewed-By: Robert Haas <robertmhaas@gmail.com>
Discussion: https://postgr.es/m/CAH2-WzkymFbz6D_vL+jmqSn_5q1wsFvFrE+37yLgL_Rkfd6Gzg@mail.gmail.com
This commit is contained in:
Peter Geoghegan 2022-04-03 09:57:21 -07:00
parent 05023a237c
commit 0b018fabaa
13 changed files with 717 additions and 305 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
doc/src/sgml/maintenance.sgml
View File

@ -586,11 +586,11 @@
statistics in the system tables <structname>pg_class</structname> and
<structname>pg_database</structname>. In particular,
the <structfield>relfrozenxid</structfield> column of a table's
<structname>pg_class</structname> row contains the freeze cutoff XID that was used
by the last aggressive <command>VACUUM</command> for that table. All rows
inserted by transactions with XIDs older than this cutoff XID are
guaranteed to have been frozen. Similarly,
the <structfield>datfrozenxid</structfield> column of a database's
<structname>pg_class</structname> row contains the oldest remaining unfrozen
XID at the end of the most recent <command>VACUUM</command> that successfully
advanced <structfield>relfrozenxid</structfield> (typically the most recent
aggressive VACUUM). Similarly, the
<structfield>datfrozenxid</structfield> column of a database's
<structname>pg_database</structname> row is a lower bound on the unfrozen XIDs
appearing in that database &mdash; it is just the minimum of the
per-table <structfield>relfrozenxid</structfield> values within the database.
@ -638,7 +638,11 @@ SELECT datname, age(datfrozenxid) FROM pg_database;
set <literal>age(relfrozenxid)</literal> to a value just a little more than the
<varname>vacuum_freeze_min_age</varname> setting
that was used (more by the number of transactions started since the
<command>VACUUM</command> started). If no <structfield>relfrozenxid</structfield>-advancing
<command>VACUUM</command> started). <command>VACUUM</command>
will set <structfield>relfrozenxid</structfield> to the oldest XID
that remains in the table, so it's possible that the final value
will be much more recent than strictly required.
If no <structfield>relfrozenxid</structfield>-advancing
<command>VACUUM</command> is issued on the table until
<varname>autovacuum_freeze_max_age</varname> is reached, an autovacuum will soon
be forced for the table.

332
src/backend/access/heap/heapam.c
View File

@ -6079,10 +6079,12 @@ heap_inplace_update(Relation relation, HeapTuple tuple)
* Determine what to do during freezing when a tuple is marked by a
* MultiXactId.
*
* NB -- this might have the side-effect of creating a new MultiXactId!
*
* "flags" is an output value; it's used to tell caller what to do on return.
* Possible flags are:
*
* "mxid_oldest_xid_out" is an output value; it's used to track the oldest
* extant Xid within any Multixact that will remain after freezing executes.
*
* Possible values that we can set in "flags":
* FRM_NOOP
* don't do anything -- keep existing Xmax
* FRM_INVALIDATE_XMAX
@ -6094,12 +6096,17 @@ heap_inplace_update(Relation relation, HeapTuple tuple)
* FRM_RETURN_IS_MULTI
* The return value is a new MultiXactId to set as new Xmax.
* (caller must obtain proper infomask bits using GetMultiXactIdHintBits)
*
* "mxid_oldest_xid_out" is only set when "flags" contains either FRM_NOOP or
* FRM_RETURN_IS_MULTI, since we only leave behind a MultiXactId for these.
*
* NB: Creates a _new_ MultiXactId when FRM_RETURN_IS_MULTI is set in "flags".
*/
static TransactionId
FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
TransactionId relfrozenxid, TransactionId relminmxid,
TransactionId cutoff_xid, MultiXactId cutoff_multi,
uint16 *flags)
uint16 *flags, TransactionId *mxid_oldest_xid_out)
{
TransactionId xid = InvalidTransactionId;
int i;
@ -6111,6 +6118,7 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
bool has_lockers;
TransactionId update_xid;
bool update_committed;
TransactionId temp_xid_out;
*flags = 0;
@ -6147,7 +6155,7 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
if (HEAP_XMAX_IS_LOCKED_ONLY(t_infomask))
{
*flags |= FRM_INVALIDATE_XMAX;
xid = InvalidTransactionId; /* not strictly necessary */
xid = InvalidTransactionId;
}
else
{
@ -6174,7 +6182,7 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
(errcode(ERRCODE_DATA_CORRUPTED),
errmsg_internal("cannot freeze committed update xid %u", xid)));
*flags |= FRM_INVALIDATE_XMAX;
xid = InvalidTransactionId; /* not strictly necessary */
xid = InvalidTransactionId;
}
else
{
@ -6182,6 +6190,10 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
}
}
/*
* Don't push back mxid_oldest_xid_out using FRM_RETURN_IS_XID Xid, or
* when no Xids will remain
*/
return xid;
}
@ -6205,6 +6217,7 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
/* is there anything older than the cutoff? */
need_replace = false;
temp_xid_out = *mxid_oldest_xid_out; /* init for FRM_NOOP */
for (i = 0; i < nmembers; i++)
{
if (TransactionIdPrecedes(members[i].xid, cutoff_xid))
@ -6212,28 +6225,38 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
need_replace = true;
break;
}
if (TransactionIdPrecedes(members[i].xid, temp_xid_out))
temp_xid_out = members[i].xid;
}
/*
* In the simplest case, there is no member older than the cutoff; we can
* keep the existing MultiXactId as is.
* keep the existing MultiXactId as-is, avoiding a more expensive second
* pass over the multi
*/
if (!need_replace)
{
/*
* When mxid_oldest_xid_out gets pushed back here it's likely that the
* update Xid was the oldest member, but we don't rely on that
*/
*flags |= FRM_NOOP;
*mxid_oldest_xid_out = temp_xid_out;
pfree(members);
return InvalidTransactionId;
return multi;
}
/*
* If the multi needs to be updated, figure out which members do we need
* to keep.
* Do a more thorough second pass over the multi to figure out which
* member XIDs actually need to be kept. Checking the precise status of
* individual members might even show that we don't need to keep anything.
*/
nnewmembers = 0;
newmembers = palloc(sizeof(MultiXactMember) * nmembers);
has_lockers = false;
update_xid = InvalidTransactionId;
update_committed = false;
temp_xid_out = *mxid_oldest_xid_out; /* init for FRM_RETURN_IS_MULTI */
for (i = 0; i < nmembers; i++)
{
@ -6289,7 +6312,7 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
}
/*
* Since the tuple wasn't marked HEAPTUPLE_DEAD by vacuum, the
* Since the tuple wasn't totally removed when vacuum pruned, the
* update Xid cannot possibly be older than the xid cutoff. The
* presence of such a tuple would cause corruption, so be paranoid
* and check.
@ -6302,15 +6325,20 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
update_xid, cutoff_xid)));
/*
* If we determined that it's an Xid corresponding to an update
* that must be retained, additionally add it to the list of
* members of the new Multi, in case we end up using that. (We
* might still decide to use only an update Xid and not a multi,
* but it's easier to maintain the list as we walk the old members
* list.)
* We determined that this is an Xid corresponding to an update
* that must be retained -- add it to new members list for later.
*
* Also consider pushing back temp_xid_out, which is needed when
* we later conclude that a new multi is required (i.e. when we go
* on to set FRM_RETURN_IS_MULTI for our caller because we also
* need to retain a locker that's still running).
*/
if (TransactionIdIsValid(update_xid))
{
newmembers[nnewmembers++] = members[i];
if (TransactionIdPrecedes(members[i].xid, temp_xid_out))
temp_xid_out = members[i].xid;
}
}
else
{
@ -6318,8 +6346,18 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
if (TransactionIdIsCurrentTransactionId(members[i].xid) ||
TransactionIdIsInProgress(members[i].xid))
{
/* running locker cannot possibly be older than the cutoff */
/*
* Running locker cannot possibly be older than the cutoff.
*
* The cutoff is <= VACUUM's OldestXmin, which is also the
* initial value used for top-level relfrozenxid_out tracking
* state. A running locker cannot be older than VACUUM's
* OldestXmin, either, so we don't need a temp_xid_out step.
*/
Assert(TransactionIdIsNormal(members[i].xid));
Assert(!TransactionIdPrecedes(members[i].xid, cutoff_xid));
Assert(!TransactionIdPrecedes(members[i].xid,
*mxid_oldest_xid_out));
newmembers[nnewmembers++] = members[i];
has_lockers = true;
}
@ -6328,11 +6366,16 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
pfree(members);
/*
* Determine what to do with caller's multi based on information gathered
* during our second pass
*/
if (nnewmembers == 0)
{
/* nothing worth keeping!? Tell caller to remove the whole thing */
*flags |= FRM_INVALIDATE_XMAX;
xid = InvalidTransactionId;
/* Don't push back mxid_oldest_xid_out -- no Xids will remain */
}
else if (TransactionIdIsValid(update_xid) && !has_lockers)
{
@ -6348,15 +6391,18 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
if (update_committed)
*flags |= FRM_MARK_COMMITTED;
xid = update_xid;
/* Don't push back mxid_oldest_xid_out using FRM_RETURN_IS_XID Xid */
}
else
{
/*
* Create a new multixact with the surviving members of the previous
* one, to set as new Xmax in the tuple.
* one, to set as new Xmax in the tuple. The oldest surviving member
* might push back mxid_oldest_xid_out.
*/
xid = MultiXactIdCreateFromMembers(nnewmembers, newmembers);
*flags |= FRM_RETURN_IS_MULTI;
*mxid_oldest_xid_out = temp_xid_out;
}
pfree(newmembers);
@ -6375,31 +6421,41 @@ FreezeMultiXactId(MultiXactId multi, uint16 t_infomask,
* will be totally frozen after these operations are performed and false if
* more freezing will eventually be required.
*
* Caller is responsible for setting the offset field, if appropriate.
* Caller must set frz->offset itself, before heap_execute_freeze_tuple call.
*
* It is assumed that the caller has checked the tuple with
* HeapTupleSatisfiesVacuum() and determined that it is not HEAPTUPLE_DEAD
* (else we should be removing the tuple, not freezing it).
*
* NB: cutoff_xid *must* be <= the current global xmin, to ensure that any
* The *relfrozenxid_out and *relminmxid_out arguments are the current target
* relfrozenxid and relminmxid for VACUUM caller's heap rel. Any and all
* unfrozen XIDs or MXIDs that remain in caller's rel after VACUUM finishes
* _must_ have values >= the final relfrozenxid/relminmxid values in pg_class.
* This includes XIDs that remain as MultiXact members from any tuple's xmax.
* Each call here pushes back *relfrozenxid_out and/or *relminmxid_out as
* needed to avoid unsafe final values in rel's authoritative pg_class tuple.
*
* NB: cutoff_xid *must* be <= VACUUM's OldestXmin, to ensure that any
* XID older than it could neither be running nor seen as running by any
* open transaction. This ensures that the replacement will not change
* anyone's idea of the tuple state.
* Similarly, cutoff_multi must be less than or equal to the smallest
* MultiXactId used by any transaction currently open.
* Similarly, cutoff_multi must be <= VACUUM's OldestMxact.
*
* If the tuple is in a shared buffer, caller must hold an exclusive lock on
* that buffer.
* NB: This function has side effects: it might allocate a new MultiXactId.
* It will be set as tuple's new xmax when our *frz output is processed within
* heap_execute_freeze_tuple later on. If the tuple is in a shared buffer
* then caller had better have an exclusive lock on it already.
*
* NB: It is not enough to set hint bits to indicate something is
* committed/invalid -- they might not be set on a standby, or after crash
* recovery. We really need to remove old xids.
* NB: It is not enough to set hint bits to indicate an XID committed/aborted.
* The *frz WAL record we output completely removes all old XIDs during REDO.
*/
bool
heap_prepare_freeze_tuple(HeapTupleHeader tuple,
TransactionId relfrozenxid, TransactionId relminmxid,
TransactionId cutoff_xid, TransactionId cutoff_multi,
xl_heap_freeze_tuple *frz, bool *totally_frozen)
xl_heap_freeze_tuple *frz, bool *totally_frozen,
TransactionId *relfrozenxid_out,
MultiXactId *relminmxid_out)
{
bool changed = false;
bool xmax_already_frozen = false;
@ -6418,7 +6474,9 @@ heap_prepare_freeze_tuple(HeapTupleHeader tuple,
* already a permanent value), while in the block below it is set true to
* mean "xmin won't need freezing after what we do to it here" (false
* otherwise). In both cases we're allowed to set totally_frozen, as far
* as xmin is concerned.
* as xmin is concerned. Both cases also don't require relfrozenxid_out
* handling, since either way the tuple's xmin will be a permanent value
* once we're done with it.
*/
xid = HeapTupleHeaderGetXmin(tuple);
if (!TransactionIdIsNormal(xid))
@ -6443,6 +6501,12 @@ heap_prepare_freeze_tuple(HeapTupleHeader tuple,
frz->t_infomask |= HEAP_XMIN_FROZEN;
changed = true;
}
else
{
/* xmin to remain unfrozen. Could push back relfrozenxid_out. */
if (TransactionIdPrecedes(xid, *relfrozenxid_out))
*relfrozenxid_out = xid;
}
}
/*
@ -6452,7 +6516,7 @@ heap_prepare_freeze_tuple(HeapTupleHeader tuple,
* freezing, too. Also, if a multi needs freezing, we cannot simply take
* it out --- if there's a live updater Xid, it needs to be kept.
*
* Make sure to keep heap_tuple_needs_freeze in sync with this.
* Make sure to keep heap_tuple_would_freeze in sync with this.
*/
xid = HeapTupleHeaderGetRawXmax(tuple);
@ -6460,15 +6524,28 @@ heap_prepare_freeze_tuple(HeapTupleHeader tuple,
{
TransactionId newxmax;
uint16 flags;
TransactionId mxid_oldest_xid_out = *relfrozenxid_out;
newxmax = FreezeMultiXactId(xid, tuple->t_infomask,
relfrozenxid, relminmxid,
cutoff_xid, cutoff_multi, &flags);
cutoff_xid, cutoff_multi,
&flags, &mxid_oldest_xid_out);
freeze_xmax = (flags & FRM_INVALIDATE_XMAX);
if (flags & FRM_RETURN_IS_XID)
{
/*
* xmax will become an updater Xid (original MultiXact's updater
* member Xid will be carried forward as a simple Xid in Xmax).
* Might have to ratchet back relfrozenxid_out here, though never
* relminmxid_out.
*/
Assert(!freeze_xmax);
Assert(TransactionIdIsValid(newxmax));
if (TransactionIdPrecedes(newxmax, *relfrozenxid_out))
*relfrozenxid_out = newxmax;
/*
* NB -- some of these transformations are only valid because we
* know the return Xid is a tuple updater (i.e. not merely a
@ -6487,6 +6564,19 @@ heap_prepare_freeze_tuple(HeapTupleHeader tuple,
uint16 newbits;
uint16 newbits2;
/*
* xmax is an old MultiXactId that we have to replace with a new
* MultiXactId, to carry forward two or more original member XIDs.
* Might have to ratchet back relfrozenxid_out here, though never
* relminmxid_out.
*/
Assert(!freeze_xmax);
Assert(MultiXactIdIsValid(newxmax));
Assert(!MultiXactIdPrecedes(newxmax, *relminmxid_out));
Assert(TransactionIdPrecedesOrEquals(mxid_oldest_xid_out,
*relfrozenxid_out));
*relfrozenxid_out = mxid_oldest_xid_out;
/*
* We can't use GetMultiXactIdHintBits directly on the new multi
* here; that routine initializes the masks to all zeroes, which
@ -6503,6 +6593,30 @@ heap_prepare_freeze_tuple(HeapTupleHeader tuple,
changed = true;
}
else if (flags & FRM_NOOP)
{
/*
* xmax is a MultiXactId, and nothing about it changes for now.
* Might have to ratchet back relminmxid_out, relfrozenxid_out, or
* both together.
*/
Assert(!freeze_xmax);
Assert(MultiXactIdIsValid(newxmax) && xid == newxmax);
Assert(TransactionIdPrecedesOrEquals(mxid_oldest_xid_out,
*relfrozenxid_out));
if (MultiXactIdPrecedes(xid, *relminmxid_out))
*relminmxid_out = xid;
*relfrozenxid_out = mxid_oldest_xid_out;
}
else
{
/*
* Keeping nothing (neither an Xid nor a MultiXactId) in xmax.
* Won't have to ratchet back relminmxid_out or relfrozenxid_out.
*/
Assert(freeze_xmax);
Assert(!TransactionIdIsValid(newxmax));
}
}
else if (TransactionIdIsNormal(xid))
{
@ -6527,15 +6641,21 @@ heap_prepare_freeze_tuple(HeapTupleHeader tuple,
errmsg_internal("cannot freeze committed xmax %u",
xid)));
freeze_xmax = true;
/* No need for relfrozenxid_out handling, since we'll freeze xmax */
}
else
{
freeze_xmax = false;
if (TransactionIdPrecedes(xid, *relfrozenxid_out))
*relfrozenxid_out = xid;
}
}
else if ((tuple->t_infomask & HEAP_XMAX_INVALID) ||
!TransactionIdIsValid(HeapTupleHeaderGetRawXmax(tuple)))
{
freeze_xmax = false;
xmax_already_frozen = true;
/* No need for relfrozenxid_out handling for already-frozen xmax */
}
else
ereport(ERROR,
@ -6576,6 +6696,8 @@ heap_prepare_freeze_tuple(HeapTupleHeader tuple,
* was removed in PostgreSQL 9.0. Note that if we were to respect
* cutoff_xid here, we'd need to make surely to clear totally_frozen
* when we skipped freezing on that basis.
*
* No need for relfrozenxid_out handling, since we always freeze xvac.
*/
if (TransactionIdIsNormal(xid))
{
@ -6653,11 +6775,14 @@ heap_freeze_tuple(HeapTupleHeader tuple,
xl_heap_freeze_tuple frz;
bool do_freeze;
bool tuple_totally_frozen;
TransactionId relfrozenxid_out = cutoff_xid;
MultiXactId relminmxid_out = cutoff_multi;
do_freeze = heap_prepare_freeze_tuple(tuple,
relfrozenxid, relminmxid,
cutoff_xid, cutoff_multi,
&frz, &tuple_totally_frozen);
&frz, &tuple_totally_frozen,
&relfrozenxid_out, &relminmxid_out);
/*
* Note that because this is not a WAL-logged operation, we don't need to
@ -7036,9 +7161,7 @@ ConditionalMultiXactIdWait(MultiXactId multi, MultiXactStatus status,
* heap_tuple_needs_eventual_freeze
*
* Check to see whether any of the XID fields of a tuple (xmin, xmax, xvac)
* will eventually require freezing. Similar to heap_tuple_needs_freeze,
* but there's no cutoff, since we're trying to figure out whether freezing
* will ever be needed, not whether it's needed now.
* will eventually require freezing (if tuple isn't removed by pruning first).
*/
bool
heap_tuple_needs_eventual_freeze(HeapTupleHeader tuple)
@ -7082,87 +7205,106 @@ heap_tuple_needs_eventual_freeze(HeapTupleHeader tuple)
}
/*
* heap_tuple_needs_freeze
* heap_tuple_would_freeze
*
* Check to see whether any of the XID fields of a tuple (xmin, xmax, xvac)
* are older than the specified cutoff XID or MultiXactId. If so, return true.
* Return value indicates if heap_prepare_freeze_tuple sibling function would
* freeze any of the XID/XMID fields from the tuple, given the same cutoffs.
* We must also deal with dead tuples here, since (xmin, xmax, xvac) fields
* could be processed by pruning away the whole tuple instead of freezing.
*
* It doesn't matter whether the tuple is alive or dead, we are checking
* to see if a tuple needs to be removed or frozen to avoid wraparound.
*
* NB: Cannot rely on hint bits here, they might not be set after a crash or
* on a standby.
* The *relfrozenxid_out and *relminmxid_out input/output arguments work just
* like the heap_prepare_freeze_tuple arguments that they're based on. We
* never freeze here, which makes tracking the oldest extant XID/MXID simple.
*/
bool
heap_tuple_needs_freeze(HeapTupleHeader tuple, TransactionId cutoff_xid,
MultiXactId cutoff_multi)
heap_tuple_would_freeze(HeapTupleHeader tuple, TransactionId cutoff_xid,
MultiXactId cutoff_multi,
TransactionId *relfrozenxid_out,
MultiXactId *relminmxid_out)
{
TransactionId xid;
MultiXactId multi;
bool would_freeze = false;
/* First deal with xmin */
xid = HeapTupleHeaderGetXmin(tuple);
if (TransactionIdIsNormal(xid) &&
TransactionIdPrecedes(xid, cutoff_xid))
return true;
/*
* The considerations for multixacts are complicated; look at
* heap_prepare_freeze_tuple for justifications. This routine had better
* be in sync with that one!
*/
if (tuple->t_infomask & HEAP_XMAX_IS_MULTI)
if (TransactionIdIsNormal(xid))
{
MultiXactId multi;
if (TransactionIdPrecedes(xid, *relfrozenxid_out))
*relfrozenxid_out = xid;
if (TransactionIdPrecedes(xid, cutoff_xid))
would_freeze = true;
}
/* Now deal with xmax */
xid = InvalidTransactionId;
multi = InvalidMultiXactId;
if (tuple->t_infomask & HEAP_XMAX_IS_MULTI)
multi = HeapTupleHeaderGetRawXmax(tuple);
if (!MultiXactIdIsValid(multi))
{
/* no xmax set, ignore */
;
}
else if (HEAP_LOCKED_UPGRADED(tuple->t_infomask))
return true;
else if (MultiXactIdPrecedes(multi, cutoff_multi))
return true;
else
{
MultiXactMember *members;
int nmembers;
int i;
else
xid = HeapTupleHeaderGetRawXmax(tuple);
/* need to check whether any member of the mxact is too old */
nmembers = GetMultiXactIdMembers(multi, &members, false,
HEAP_XMAX_IS_LOCKED_ONLY(tuple->t_infomask));
for (i = 0; i < nmembers; i++)
{
if (TransactionIdPrecedes(members[i].xid, cutoff_xid))
{
pfree(members);
return true;
}
}
if (nmembers > 0)
pfree(members);
}
if (TransactionIdIsNormal(xid))
{
/* xmax is a non-permanent XID */
if (TransactionIdPrecedes(xid, *relfrozenxid_out))
*relfrozenxid_out = xid;
if (TransactionIdPrecedes(xid, cutoff_xid))
would_freeze = true;
}
else if (!MultiXactIdIsValid(multi))
{
/* xmax is a permanent XID or invalid MultiXactId/XID */
}
else if (HEAP_LOCKED_UPGRADED(tuple->t_infomask))
{
/* xmax is a pg_upgrade'd MultiXact, which can't have updater XID */
if (MultiXactIdPrecedes(multi, *relminmxid_out))
*relminmxid_out = multi;
/* heap_prepare_freeze_tuple always freezes pg_upgrade'd xmax */
would_freeze = true;
}
else
{
xid = HeapTupleHeaderGetRawXmax(tuple);
if (TransactionIdIsNormal(xid) &&
TransactionIdPrecedes(xid, cutoff_xid))
return true;
/* xmax is a MultiXactId that may have an updater XID */
MultiXactMember *members;
int nmembers;
if (MultiXactIdPrecedes(multi, *relminmxid_out))
*relminmxid_out = multi;
if (MultiXactIdPrecedes(multi, cutoff_multi))
would_freeze = true;
/* need to check whether any member of the mxact is old */
nmembers = GetMultiXactIdMembers(multi, &members, false,
HEAP_XMAX_IS_LOCKED_ONLY(tuple->t_infomask));
for (int i = 0; i < nmembers; i++)
{
xid = members[i].xid;
Assert(TransactionIdIsNormal(xid));
if (TransactionIdPrecedes(xid, *relfrozenxid_out))
*relfrozenxid_out = xid;
if (TransactionIdPrecedes(xid, cutoff_xid))
would_freeze = true;
}
if (nmembers > 0)
pfree(members);
}
if (tuple->t_infomask & HEAP_MOVED)
{
xid = HeapTupleHeaderGetXvac(tuple);
if (TransactionIdIsNormal(xid) &&
TransactionIdPrecedes(xid, cutoff_xid))
return true;
if (TransactionIdIsNormal(xid))
{
if (TransactionIdPrecedes(xid, *relfrozenxid_out))
*relfrozenxid_out = xid;
/* heap_prepare_freeze_tuple always freezes xvac */
would_freeze = true;
}
}
return false;
return would_freeze;
}
/*

156
src/backend/access/heap/vacuumlazy.c
View File

@ -144,7 +144,7 @@ typedef struct LVRelState
Relation *indrels;
int nindexes;
/* Aggressive VACUUM (scan all unfrozen pages)? */
/* Aggressive VACUUM? (must set relfrozenxid >= FreezeLimit) */
bool aggressive;
/* Use visibility map to skip? (disabled by DISABLE_PAGE_SKIPPING) */
bool skipwithvm;
@ -173,8 +173,9 @@ typedef struct LVRelState
/* VACUUM operation's target cutoffs for freezing XIDs and MultiXactIds */
TransactionId FreezeLimit;
MultiXactId MultiXactCutoff;
/* Are FreezeLimit/MultiXactCutoff still valid? */
bool freeze_cutoffs_valid;
/* Tracks oldest extant XID/MXID for setting relfrozenxid/relminmxid */
TransactionId NewRelfrozenXid;
MultiXactId NewRelminMxid;
/* Error reporting state */
char *relnamespace;
@ -313,7 +314,8 @@ heap_vacuum_rel(Relation rel, VacuumParams *params,
minmulti_updated;
TransactionId OldestXmin,
FreezeLimit;
MultiXactId MultiXactCutoff;
MultiXactId OldestMxact,
MultiXactCutoff;
BlockNumber orig_rel_pages,
new_rel_pages,
new_rel_allvisible;
@ -345,20 +347,17 @@ heap_vacuum_rel(Relation rel, VacuumParams *params,
/*
* Get OldestXmin cutoff, which is used to determine which deleted tuples
* are considered DEAD, not just RECENTLY_DEAD. Also get related cutoffs
* used to determine which XIDs/MultiXactIds will be frozen.
*
* If this is an aggressive VACUUM, then we're strictly required to freeze
* any and all XIDs from before FreezeLimit, so that we will be able to
* safely advance relfrozenxid up to FreezeLimit below (we must be able to
* advance relminmxid up to MultiXactCutoff, too).
* used to determine which XIDs/MultiXactIds will be frozen. If this is
* an aggressive VACUUM then lazy_scan_heap cannot leave behind unfrozen
* XIDs < FreezeLimit (all MXIDs < MultiXactCutoff also need to go away).
*/
aggressive = vacuum_set_xid_limits(rel,
params->freeze_min_age,
params->freeze_table_age,
params->multixact_freeze_min_age,
params->multixact_freeze_table_age,
&OldestXmin, &FreezeLimit,
&MultiXactCutoff);
&OldestXmin, &OldestMxact,
&FreezeLimit, &MultiXactCutoff);
skipwithvm = true;
if (params->options & VACOPT_DISABLE_PAGE_SKIPPING)
@ -505,10 +504,11 @@ heap_vacuum_rel(Relation rel, VacuumParams *params,
vacrel->vistest = GlobalVisTestFor(rel);
/* FreezeLimit controls XID freezing (always <= OldestXmin) */
vacrel->FreezeLimit = FreezeLimit;
/* MultiXactCutoff controls MXID freezing */
/* MultiXactCutoff controls MXID freezing (always <= OldestMxact) */
vacrel->MultiXactCutoff = MultiXactCutoff;
/* Track if cutoffs became invalid (possible in !aggressive case only) */
vacrel->freeze_cutoffs_valid = true;
/* Initialize state used to track oldest extant XID/XMID */
vacrel->NewRelfrozenXid = OldestXmin;
vacrel->NewRelminMxid = OldestMxact;
/*
* Call lazy_scan_heap to perform all required heap pruning, index
@ -542,13 +542,33 @@ heap_vacuum_rel(Relation rel, VacuumParams *params,
/*
* Prepare to update rel's pg_class entry.
*
* In principle new_live_tuples could be -1 indicating that we (still)
* don't know the tuple count. In practice that probably can't happen,
* since we'd surely have scanned some pages if the table is new and
* nonempty.
*
* Aggressive VACUUMs must always be able to advance relfrozenxid to a
* value >= FreezeLimit, and relminmxid to a value >= MultiXactCutoff.
* Non-aggressive VACUUMs may advance them by any amount, or not at all.
*/
Assert(vacrel->NewRelfrozenXid == OldestXmin ||
TransactionIdPrecedesOrEquals(aggressive ? FreezeLimit :
vacrel->relfrozenxid,
vacrel->NewRelfrozenXid));
Assert(vacrel->NewRelminMxid == OldestMxact ||
MultiXactIdPrecedesOrEquals(aggressive ? MultiXactCutoff :
vacrel->relminmxid,
vacrel->NewRelminMxid));
if (vacrel->scanned_pages + vacrel->frozenskipped_pages < orig_rel_pages)
{
/*
* Must keep original relfrozenxid in a non-aggressive VACUUM that
* had to skip an all-visible page. The state that tracks new
* values will have missed unfrozen XIDs from the pages we skipped.
*/
Assert(!aggressive);
vacrel->NewRelfrozenXid = InvalidTransactionId;
vacrel->NewRelminMxid = InvalidMultiXactId;
}
/*
* For safety, clamp relallvisible to be not more than what we're setting
* relpages to.
* pg_class.relpages to
*/
new_rel_pages = vacrel->rel_pages; /* After possible rel truncation */
visibilitymap_count(rel, &new_rel_allvisible, NULL);
@ -558,33 +578,14 @@ heap_vacuum_rel(Relation rel, VacuumParams *params,
/*
* Now actually update rel's pg_class entry.
*
* Aggressive VACUUM must reliably advance relfrozenxid (and relminmxid).
* We are able to advance relfrozenxid in a non-aggressive VACUUM too,
* provided we didn't skip any all-visible (not all-frozen) pages using
* the visibility map, and assuming that we didn't fail to get a cleanup
* lock that made it unsafe with respect to FreezeLimit (or perhaps our
* MultiXactCutoff) established for VACUUM operation.
* In principle new_live_tuples could be -1 indicating that we (still)
* don't know the tuple count. In practice that can't happen, since we
* scan every page that isn't skipped using the visibility map.
*/
if (vacrel->scanned_pages + vacrel->frozenskipped_pages < orig_rel_pages ||
!vacrel->freeze_cutoffs_valid)
{
/* Cannot advance relfrozenxid/relminmxid */
Assert(!aggressive);
frozenxid_updated = minmulti_updated = false;
vac_update_relstats(rel, new_rel_pages, vacrel->new_live_tuples,
new_rel_allvisible, vacrel->nindexes > 0,
InvalidTransactionId, InvalidMultiXactId,
NULL, NULL, false);
}
else
{
Assert(vacrel->scanned_pages + vacrel->frozenskipped_pages ==
orig_rel_pages);
vac_update_relstats(rel, new_rel_pages, vacrel->new_live_tuples,
new_rel_allvisible, vacrel->nindexes > 0,
FreezeLimit, MultiXactCutoff,
&frozenxid_updated, &minmulti_updated, false);
}
vac_update_relstats(rel, new_rel_pages, vacrel->new_live_tuples,
new_rel_allvisible, vacrel->nindexes > 0,
vacrel->NewRelfrozenXid, vacrel->NewRelminMxid,
&frozenxid_updated, &minmulti_updated, false);
/*
* Report results to the stats collector, too.
@ -692,17 +693,17 @@ heap_vacuum_rel(Relation rel, VacuumParams *params,
OldestXmin, diff);
if (frozenxid_updated)
{
diff = (int32) (FreezeLimit - vacrel->relfrozenxid);
diff = (int32) (vacrel->NewRelfrozenXid - vacrel->relfrozenxid);
appendStringInfo(&buf,
_("new relfrozenxid: %u, which is %d xids ahead of previous value\n"),
FreezeLimit, diff);
vacrel->NewRelfrozenXid, diff);
}
if (minmulti_updated)
{
diff = (int32) (MultiXactCutoff - vacrel->relminmxid);
diff = (int32) (vacrel->NewRelminMxid - vacrel->relminmxid);
appendStringInfo(&buf,
_("new relminmxid: %u, which is %d mxids ahead of previous value\n"),
MultiXactCutoff, diff);
vacrel->NewRelminMxid, diff);
}
if (orig_rel_pages > 0)
{
@ -1582,6 +1583,8 @@ lazy_scan_prune(LVRelState *vacrel,
recently_dead_tuples;
int nnewlpdead;
int nfrozen;
TransactionId NewRelfrozenXid;
MultiXactId NewRelminMxid;
OffsetNumber deadoffsets[MaxHeapTuplesPerPage];
xl_heap_freeze_tuple frozen[MaxHeapTuplesPerPage];
@ -1591,7 +1594,9 @@ lazy_scan_prune(LVRelState *vacrel,
retry:
/* Initialize (or reset) page-level counters */
/* Initialize (or reset) page-level state */
NewRelfrozenXid = vacrel->NewRelfrozenXid;
NewRelminMxid = vacrel->NewRelminMxid;
tuples_deleted = 0;
lpdead_items = 0;
live_tuples = 0;
@ -1798,8 +1803,8 @@ retry:
vacrel->relminmxid,
vacrel->FreezeLimit,
vacrel->MultiXactCutoff,
&frozen[nfrozen],
&tuple_totally_frozen))
&frozen[nfrozen], &tuple_totally_frozen,
&NewRelfrozenXid, &NewRelminMxid))
{
/* Will execute freeze below */
frozen[nfrozen++].offset = offnum;
@ -1813,13 +1818,16 @@ retry:
prunestate->all_frozen = false;
}
vacrel->offnum = InvalidOffsetNumber;
/*
* We have now divided every item on the page into either an LP_DEAD item
* that will need to be vacuumed in indexes later, or a LP_NORMAL tuple
* that remains and needs to be considered for freezing now (LP_UNUSED and
* LP_REDIRECT items also remain, but are of no further interest to us).
*/
vacrel->offnum = InvalidOffsetNumber;
vacrel->NewRelfrozenXid = NewRelfrozenXid;
vacrel->NewRelminMxid = NewRelminMxid;
/*
* Consider the need to freeze any items with tuple storage from the page
@ -1969,6 +1977,8 @@ lazy_scan_noprune(LVRelState *vacrel,
recently_dead_tuples,
missed_dead_tuples;
HeapTupleHeader tupleheader;
TransactionId NewRelfrozenXid = vacrel->NewRelfrozenXid;
MultiXactId NewRelminMxid = vacrel->NewRelminMxid;
OffsetNumber deadoffsets[MaxHeapTuplesPerPage];
Assert(BufferGetBlockNumber(buf) == blkno);
@ -2013,22 +2023,37 @@ lazy_scan_noprune(LVRelState *vacrel,
*hastup = true; /* page prevents rel truncation */
tupleheader = (HeapTupleHeader) PageGetItem(page, itemid);
if (heap_tuple_needs_freeze(tupleheader,
if (heap_tuple_would_freeze(tupleheader,
vacrel->FreezeLimit,
vacrel->MultiXactCutoff))
vacrel->MultiXactCutoff,
&NewRelfrozenXid, &NewRelminMxid))
{
/* Tuple with XID < FreezeLimit (or MXID < MultiXactCutoff) */
if (vacrel->aggressive)
{
/* Going to have to get cleanup lock for lazy_scan_prune */
/*
* Aggressive VACUUMs must always be able to advance rel's
* relfrozenxid to a value >= FreezeLimit (and be able to
* advance rel's relminmxid to a value >= MultiXactCutoff).
* The ongoing aggressive VACUUM won't be able to do that
* unless it can freeze an XID (or XMID) from this tuple now.
*
* The only safe option is to have caller perform processing
* of this page using lazy_scan_prune. Caller might have to
* wait a while for a cleanup lock, but it can't be helped.
*/
vacrel->offnum = InvalidOffsetNumber;
return false;
}
/*
* Current non-aggressive VACUUM operation definitely won't be
* able to advance relfrozenxid or relminmxid
* Non-aggressive VACUUMs are under no obligation to advance
* relfrozenxid (even by one XID). We can be much laxer here.
*
* Currently we always just accept an older final relfrozenxid
* and/or relminmxid value. We never make caller wait or work a
* little harder, even when it likely makes sense to do so.
*/
vacrel->freeze_cutoffs_valid = false;
}
ItemPointerSet(&(tuple.t_self), blkno, offnum);
@ -2078,9 +2103,14 @@ lazy_scan_noprune(LVRelState *vacrel,
vacrel->offnum = InvalidOffsetNumber;
/*
* Now save details of the LP_DEAD items from the page in vacrel (though
* only when VACUUM uses two-pass strategy)
* By here we know for sure that caller can put off freezing and pruning
* this particular page until the next VACUUM. Remember its details now.
* (lazy_scan_prune expects a clean slate, so we have to do this last.)
*/
vacrel->NewRelfrozenXid = NewRelfrozenXid;
vacrel->NewRelminMxid = NewRelminMxid;
/* Save any LP_DEAD items found on the page in dead_items array */
if (vacrel->nindexes == 0)
{
/* Using one-pass strategy (since table has no indexes) */

11
src/backend/commands/cluster.c
View File

@ -806,9 +806,10 @@ copy_table_data(Oid OIDNewHeap, Oid OIDOldHeap, Oid OIDOldIndex, bool verbose,
Form_pg_class relform;
TupleDesc oldTupDesc PG_USED_FOR_ASSERTS_ONLY;
TupleDesc newTupDesc PG_USED_FOR_ASSERTS_ONLY;
TransactionId OldestXmin;
TransactionId FreezeXid;
MultiXactId MultiXactCutoff;
TransactionId OldestXmin,
FreezeXid;
MultiXactId OldestMxact,
MultiXactCutoff;
bool use_sort;
double num_tuples = 0,
tups_vacuumed = 0,
@ -896,8 +897,8 @@ copy_table_data(Oid OIDNewHeap, Oid OIDOldHeap, Oid OIDOldIndex, bool verbose,
* Since we're going to rewrite the whole table anyway, there's no reason
* not to be aggressive about this.
*/
vacuum_set_xid_limits(OldHeap, 0, 0, 0, 0,
&OldestXmin, &FreezeXid, &MultiXactCutoff);
vacuum_set_xid_limits(OldHeap, 0, 0, 0, 0, &OldestXmin, &OldestMxact,
&FreezeXid, &MultiXactCutoff);
/*
* FreezeXid will become the table's new relfrozenxid, and that mustn't go

39
src/backend/commands/vacuum.c
View File

@ -945,14 +945,22 @@ get_all_vacuum_rels(int options)
* The output parameters are:
* - oldestXmin is the Xid below which tuples deleted by any xact (that
* committed) should be considered DEAD, not just RECENTLY_DEAD.
* - freezeLimit is the Xid below which all Xids are replaced by
* FrozenTransactionId during vacuum.
* - multiXactCutoff is the value below which all MultiXactIds are removed
* from Xmax.
* - oldestMxact is the Mxid below which MultiXacts are definitely not
* seen as visible by any running transaction.
* - freezeLimit is the Xid below which all Xids are definitely replaced by
* FrozenTransactionId during aggressive vacuums.
* - multiXactCutoff is the value below which all MultiXactIds are definitely
* removed from Xmax during aggressive vacuums.
*
* Return value indicates if vacuumlazy.c caller should make its VACUUM
* operation aggressive. An aggressive VACUUM must advance relfrozenxid up to
* FreezeLimit, and relminmxid up to multiXactCutoff.
* FreezeLimit (at a minimum), and relminmxid up to multiXactCutoff (at a
* minimum).
*
* oldestXmin and oldestMxact are the most recent values that can ever be
* passed to vac_update_relstats() as frozenxid and minmulti arguments by our
* vacuumlazy.c caller later on. These values should be passed when it turns
* out that VACUUM will leave no unfrozen XIDs/XMIDs behind in the table.
*/
bool
vacuum_set_xid_limits(Relation rel,
@ -961,6 +969,7 @@ vacuum_set_xid_limits(Relation rel,
int multixact_freeze_min_age,
int multixact_freeze_table_age,
TransactionId *oldestXmin,
MultiXactId *oldestMxact,
TransactionId *freezeLimit,
MultiXactId *multiXactCutoff)
{
@ -969,7 +978,6 @@ vacuum_set_xid_limits(Relation rel,
int effective_multixact_freeze_max_age;
TransactionId limit;
TransactionId safeLimit;
MultiXactId oldestMxact;
MultiXactId mxactLimit;
MultiXactId safeMxactLimit;
int freezetable;
@ -1065,9 +1073,11 @@ vacuum_set_xid_limits(Relation rel,
effective_multixact_freeze_max_age / 2);
Assert(mxid_freezemin >= 0);
/* Remember for caller */
*oldestMxact = GetOldestMultiXactId();
/* compute the cutoff multi, being careful to generate a valid value */
oldestMxact = GetOldestMultiXactId();
mxactLimit = oldestMxact - mxid_freezemin;
mxactLimit = *oldestMxact - mxid_freezemin;
if (mxactLimit < FirstMultiXactId)
mxactLimit = FirstMultiXactId;
@ -1082,8 +1092,8 @@ vacuum_set_xid_limits(Relation rel,
(errmsg("oldest multixact is far in the past"),
errhint("Close open transactions with multixacts soon to avoid wraparound problems.")));
/* Use the safe limit, unless an older mxact is still running */
if (MultiXactIdPrecedes(oldestMxact, safeMxactLimit))
mxactLimit = oldestMxact;
if (MultiXactIdPrecedes(*oldestMxact, safeMxactLimit))
mxactLimit = *oldestMxact;
else
mxactLimit = safeMxactLimit;
}
@ -1390,12 +1400,9 @@ vac_update_relstats(Relation relation,
* Update relfrozenxid, unless caller passed InvalidTransactionId
* indicating it has no new data.
*
* Ordinarily, we don't let relfrozenxid go backwards: if things are
* working correctly, the only way the new frozenxid could be older would
* be if a previous VACUUM was done with a tighter freeze_min_age, in
* which case we don't want to forget the work it already did. However,
* if the stored relfrozenxid is "in the future", then it must be corrupt
* and it seems best to overwrite it with the cutoff we used this time.
* Ordinarily, we don't let relfrozenxid go backwards. However, if the
* stored relfrozenxid is "in the future" then it seems best to assume
* it's corrupt, and overwrite with the oldest remaining XID in the table.
* This should match vac_update_datfrozenxid() concerning what we consider
* to be "in the future".
*/

6
src/include/access/heapam.h
View File

@ -167,8 +167,10 @@ extern void heap_inplace_update(Relation relation, HeapTuple tuple);
extern bool heap_freeze_tuple(HeapTupleHeader tuple,
TransactionId relfrozenxid, TransactionId relminmxid,
TransactionId cutoff_xid, TransactionId cutoff_multi);
extern bool heap_tuple_needs_freeze(HeapTupleHeader tuple, TransactionId cutoff_xid,
MultiXactId cutoff_multi);
extern bool heap_tuple_would_freeze(HeapTupleHeader tuple, TransactionId cutoff_xid,
MultiXactId cutoff_multi,
TransactionId *relfrozenxid_out,
MultiXactId *relminmxid_out);
extern bool heap_tuple_needs_eventual_freeze(HeapTupleHeader tuple);
extern void simple_heap_insert(Relation relation, HeapTuple tup);

4
src/include/access/heapam_xlog.h
View File

@ -410,7 +410,9 @@ extern bool heap_prepare_freeze_tuple(HeapTupleHeader tuple,
TransactionId cutoff_xid,
TransactionId cutoff_multi,
xl_heap_freeze_tuple *frz,
bool *totally_frozen);
bool *totally_frozen,
TransactionId *relfrozenxid_out,
MultiXactId *relminmxid_out);
extern void heap_execute_freeze_tuple(HeapTupleHeader tuple,
xl_heap_freeze_tuple *xlrec_tp);
extern XLogRecPtr log_heap_visible(RelFileNode rnode, Buffer heap_buffer,

1
src/include/commands/vacuum.h
View File

@ -291,6 +291,7 @@ extern bool vacuum_set_xid_limits(Relation rel,
int multixact_freeze_min_age,
int multixact_freeze_table_age,
TransactionId *oldestXmin,
MultiXactId *oldestMxact,
TransactionId *freezeLimit,
MultiXactId *multiXactCutoff);
extern bool vacuum_xid_failsafe_check(TransactionId relfrozenxid,

189
src/test/isolation/expected/vacuum-no-cleanup-lock.out Normal file
View File

@ -0,0 +1,189 @@
Parsed test spec with 4 sessions
starting permutation: vacuumer_pg_class_stats dml_insert vacuumer_nonaggressive_vacuum vacuumer_pg_class_stats
step vacuumer_pg_class_stats:
SELECT relpages, reltuples FROM pg_class WHERE oid = 'smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 20
(1 row)
step dml_insert:
INSERT INTO smalltbl SELECT max(id) + 1 FROM smalltbl;
step vacuumer_nonaggressive_vacuum:
VACUUM smalltbl;
step vacuumer_pg_class_stats:
SELECT relpages, reltuples FROM pg_class WHERE oid = 'smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 21
(1 row)
starting permutation: vacuumer_pg_class_stats dml_insert pinholder_cursor vacuumer_nonaggressive_vacuum vacuumer_pg_class_stats pinholder_commit
step vacuumer_pg_class_stats:
SELECT relpages, reltuples FROM pg_class WHERE oid = 'smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 20
(1 row)
step dml_insert:
INSERT INTO smalltbl SELECT max(id) + 1 FROM smalltbl;
step pinholder_cursor:
BEGIN;
DECLARE c1 CURSOR FOR SELECT 1 AS dummy FROM smalltbl;
FETCH NEXT FROM c1;
dummy
-----
1
(1 row)
step vacuumer_nonaggressive_vacuum:
VACUUM smalltbl;
step vacuumer_pg_class_stats:
SELECT relpages, reltuples FROM pg_class WHERE oid = 'smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 21
(1 row)
step pinholder_commit:
COMMIT;
starting permutation: vacuumer_pg_class_stats pinholder_cursor dml_insert dml_delete dml_insert vacuumer_nonaggressive_vacuum vacuumer_pg_class_stats pinholder_commit
step vacuumer_pg_class_stats:
SELECT relpages, reltuples FROM pg_class WHERE oid = 'smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 20
(1 row)
step pinholder_cursor:
BEGIN;
DECLARE c1 CURSOR FOR SELECT 1 AS dummy FROM smalltbl;
FETCH NEXT FROM c1;
dummy
-----
1
(1 row)
step dml_insert:
INSERT INTO smalltbl SELECT max(id) + 1 FROM smalltbl;
step dml_delete:
DELETE FROM smalltbl WHERE id = (SELECT min(id) FROM smalltbl);
step dml_insert:
INSERT INTO smalltbl SELECT max(id) + 1 FROM smalltbl;
step vacuumer_nonaggressive_vacuum:
VACUUM smalltbl;
step vacuumer_pg_class_stats:
SELECT relpages, reltuples FROM pg_class WHERE oid = 'smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 21
(1 row)
step pinholder_commit:
COMMIT;
starting permutation: vacuumer_pg_class_stats dml_insert dml_delete pinholder_cursor dml_insert vacuumer_nonaggressive_vacuum vacuumer_pg_class_stats pinholder_commit
step vacuumer_pg_class_stats:
SELECT relpages, reltuples FROM pg_class WHERE oid = 'smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 20
(1 row)
step dml_insert:
INSERT INTO smalltbl SELECT max(id) + 1 FROM smalltbl;
step dml_delete:
DELETE FROM smalltbl WHERE id = (SELECT min(id) FROM smalltbl);
step pinholder_cursor:
BEGIN;
DECLARE c1 CURSOR FOR SELECT 1 AS dummy FROM smalltbl;
FETCH NEXT FROM c1;
dummy
-----
1
(1 row)
step dml_insert:
INSERT INTO smalltbl SELECT max(id) + 1 FROM smalltbl;
step vacuumer_nonaggressive_vacuum:
VACUUM smalltbl;
step vacuumer_pg_class_stats:
SELECT relpages, reltuples FROM pg_class WHERE oid = 'smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 21
(1 row)
step pinholder_commit:
COMMIT;
starting permutation: dml_begin dml_other_begin dml_key_share dml_other_key_share vacuumer_nonaggressive_vacuum pinholder_cursor dml_other_update dml_commit dml_other_commit vacuumer_nonaggressive_vacuum pinholder_commit vacuumer_nonaggressive_vacuum
step dml_begin: BEGIN;
step dml_other_begin: BEGIN;
step dml_key_share: SELECT id FROM smalltbl WHERE id = 3 FOR KEY SHARE;
id
--
3
(1 row)
step dml_other_key_share: SELECT id FROM smalltbl WHERE id = 3 FOR KEY SHARE;
id
--
3
(1 row)
step vacuumer_nonaggressive_vacuum:
VACUUM smalltbl;
step pinholder_cursor:
BEGIN;
DECLARE c1 CURSOR FOR SELECT 1 AS dummy FROM smalltbl;
FETCH NEXT FROM c1;
dummy
-----
1
(1 row)
step dml_other_update: UPDATE smalltbl SET t = 'u' WHERE id = 3;
step dml_commit: COMMIT;
step dml_other_commit: COMMIT;
step vacuumer_nonaggressive_vacuum:
VACUUM smalltbl;
step pinholder_commit:
COMMIT;
step vacuumer_nonaggressive_vacuum:
VACUUM smalltbl;

67
src/test/isolation/expected/vacuum-reltuples.out
View File

@ -1,67 +0,0 @@
Parsed test spec with 2 sessions
starting permutation: modify vac stats
step modify:
insert into smalltbl select max(id)+1 from smalltbl;
step vac:
vacuum smalltbl;
step stats:
select relpages, reltuples from pg_class
where oid='smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 21
(1 row)
starting permutation: modify open fetch1 vac close stats
step modify:
insert into smalltbl select max(id)+1 from smalltbl;
step open:
begin;
declare c1 cursor for select 1 as dummy from smalltbl;
step fetch1:
fetch next from c1;
dummy
-----
1
(1 row)
step vac:
vacuum smalltbl;
step close:
commit;
step stats:
select relpages, reltuples from pg_class
where oid='smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 21
(1 row)
starting permutation: modify vac stats
step modify:
insert into smalltbl select max(id)+1 from smalltbl;
step vac:
vacuum smalltbl;
step stats:
select relpages, reltuples from pg_class
where oid='smalltbl'::regclass;
relpages|reltuples
--------+---------
1| 21
(1 row)

2
src/test/isolation/isolation_schedule
View File

@ -84,7 +84,7 @@ test: alter-table-4
test: create-trigger
test: sequence-ddl
test: async-notify
test: vacuum-reltuples
test: vacuum-no-cleanup-lock
test: timeouts
test: vacuum-concurrent-drop
test: vacuum-conflict

150
src/test/isolation/specs/vacuum-no-cleanup-lock.spec Normal file
View File

@ -0,0 +1,150 @@
# Test for vacuum's reduced processing of heap pages (used for any heap page
# where a cleanup lock isn't immediately available)
#
# Debugging tip: Change VACUUM to VACUUM VERBOSE to get feedback on what's
# really going on
# Use name type here to avoid TOAST table:
setup
{
CREATE TABLE smalltbl AS SELECT i AS id, 't'::name AS t FROM generate_series(1,20) i;
ALTER TABLE smalltbl SET (autovacuum_enabled = off);
ALTER TABLE smalltbl ADD PRIMARY KEY (id);
}
setup
{
VACUUM ANALYZE smalltbl;
}
teardown
{
DROP TABLE smalltbl;
}
# This session holds a pin on smalltbl's only heap page:
session pinholder
step pinholder_cursor
{
BEGIN;
DECLARE c1 CURSOR FOR SELECT 1 AS dummy FROM smalltbl;
FETCH NEXT FROM c1;
}
step pinholder_commit
{
COMMIT;
}
# This session inserts and deletes tuples, potentially affecting reltuples:
session dml
step dml_insert
{
INSERT INTO smalltbl SELECT max(id) + 1 FROM smalltbl;
}
step dml_delete
{
DELETE FROM smalltbl WHERE id = (SELECT min(id) FROM smalltbl);
}
step dml_begin { BEGIN; }
step dml_key_share { SELECT id FROM smalltbl WHERE id = 3 FOR KEY SHARE; }
step dml_commit { COMMIT; }
# Needed for Multixact test:
session dml_other
step dml_other_begin { BEGIN; }
step dml_other_key_share { SELECT id FROM smalltbl WHERE id = 3 FOR KEY SHARE; }
step dml_other_update { UPDATE smalltbl SET t = 'u' WHERE id = 3; }
step dml_other_commit { COMMIT; }
# This session runs non-aggressive VACUUM, but with maximally aggressive
# cutoffs for tuple freezing (e.g., FreezeLimit == OldestXmin):
session vacuumer
setup
{
SET vacuum_freeze_min_age = 0;
SET vacuum_multixact_freeze_min_age = 0;
}
step vacuumer_nonaggressive_vacuum
{
VACUUM smalltbl;
}
step vacuumer_pg_class_stats
{
SELECT relpages, reltuples FROM pg_class WHERE oid = 'smalltbl'::regclass;
}
# Test VACUUM's reltuples counting mechanism.
#
# Final pg_class.reltuples should never be affected by VACUUM's inability to
# get a cleanup lock on any page, except to the extent that any cleanup lock
# contention changes the number of tuples that remain ("missed dead" tuples
# are counted in reltuples, much like "recently dead" tuples).
# Easy case:
permutation
vacuumer_pg_class_stats # Start with 20 tuples
dml_insert
vacuumer_nonaggressive_vacuum
vacuumer_pg_class_stats # End with 21 tuples
# Harder case -- count 21 tuples at the end (like last time), but with cleanup
# lock contention this time:
permutation
vacuumer_pg_class_stats # Start with 20 tuples
dml_insert
pinholder_cursor
vacuumer_nonaggressive_vacuum
vacuumer_pg_class_stats # End with 21 tuples
pinholder_commit # order doesn't matter
# Same as "harder case", but vary the order, and delete an inserted row:
permutation
vacuumer_pg_class_stats # Start with 20 tuples
pinholder_cursor
dml_insert
dml_delete
dml_insert
vacuumer_nonaggressive_vacuum
# reltuples is 21 here again -- "recently dead" tuple won't be included in
# count here:
vacuumer_pg_class_stats
pinholder_commit # order doesn't matter
# Same as "harder case", but initial insert and delete before cursor:
permutation
vacuumer_pg_class_stats # Start with 20 tuples
dml_insert
dml_delete
pinholder_cursor
dml_insert
vacuumer_nonaggressive_vacuum
# reltuples is 21 here again -- "missed dead" tuple ("recently dead" when
# concurrent activity held back VACUUM's OldestXmin) won't be included in
# count here:
vacuumer_pg_class_stats
pinholder_commit # order doesn't matter
# Test VACUUM's mechanism for skipping MultiXact freezing.
#
# This provides test coverage for code paths that are only hit when we need to
# freeze, but inability to acquire a cleanup lock on a heap page makes
# freezing some XIDs/XMIDs < FreezeLimit/MultiXactCutoff impossible (without
# waiting for a cleanup lock, which non-aggressive VACUUM is unwilling to do).
permutation
dml_begin
dml_other_begin
dml_key_share
dml_other_key_share
# Will get cleanup lock, can't advance relminmxid yet:
# (though will usually advance relfrozenxid by ~2 XIDs)
vacuumer_nonaggressive_vacuum
pinholder_cursor
dml_other_update
dml_commit
dml_other_commit
# Can't cleanup lock, so still can't advance relminmxid here:
# (relfrozenxid held back by XIDs in MultiXact too)
vacuumer_nonaggressive_vacuum
pinholder_commit
# Pin was dropped, so will advance relminmxid, at long last:
# (ditto for relfrozenxid advancement)
vacuumer_nonaggressive_vacuum

49
src/test/isolation/specs/vacuum-reltuples.spec
View File

@ -1,49 +0,0 @@
# Test for vacuum's handling of reltuples when pages are skipped due
# to page pins. We absolutely need to avoid setting reltuples=0 in
# such cases, since that interferes badly with planning.
#
# Expected result for all three permutation is 21 tuples, including
# the second permutation. VACUUM is able to count the concurrently
# inserted tuple in its final reltuples, even when a cleanup lock
# cannot be acquired on the affected heap page.
setup {
create table smalltbl
as select i as id from generate_series(1,20) i;
alter table smalltbl set (autovacuum_enabled = off);
}
setup {
vacuum analyze smalltbl;
}
teardown {
drop table smalltbl;
}
session worker
step open {
begin;
declare c1 cursor for select 1 as dummy from smalltbl;
}
step fetch1 {
fetch next from c1;
}
step close {
commit;
}
step stats {
select relpages, reltuples from pg_class
where oid='smalltbl'::regclass;
}
session vacuumer
step vac {
vacuum smalltbl;
}
step modify {
insert into smalltbl select max(id)+1 from smalltbl;
}
permutation modify vac stats
permutation modify open fetch1 vac close stats
permutation modify vac stats