/*
* TransactionIdPrecedesOrEquals --- is id1 logically <= id2?
*/
bool
TransactionIdPrecedesOrEquals(TransactionId id1, TransactionId id2)
{
int32 diff;
if (!TransactionIdIsNormal(id1) || !TransactionIdIsNormal(id2))
return (id1 <= id2);
diff = (int32) (id1 - id2);
return (diff <= 0);
}
/*
* TransactionIdFollowsOrEquals --- is id1 logically >= id2?
*/
bool
TransactionIdFollowsOrEquals(TransactionId id1, TransactionId id2)
{
int32 diff;
if (!TransactionIdIsNormal(id1) || !TransactionIdIsNormal(id2))
return (id1 >= id2);
diff = (int32) (id1 - id2);
return (diff >= 0);
}
/*
* TransactionIdPrecedes --- is id1 logically < id2?
*/
bool
TransactionIdPrecedes(TransactionId id1, TransactionId id2)
{
/*
* If either ID is a permanent XID then we can just do unsigned
* comparison. If both are normal, do a modulo-2^31 comparison.
*/
int32 diff;
if (!TransactionIdIsNormal(id1) || !TransactionIdIsNormal(id2))
return (id1 < id2);
diff = (int32) (id1 - id2);
return (diff < 0);
}
/*
* Interrogate the parent of a transaction in the subtrans log.
*/
TransactionId
SubTransGetParent(TransactionId xid)
{
int pageno = TransactionIdToPage(xid);
int entryno = TransactionIdToEntry(xid);
int slotno;
TransactionId *ptr;
TransactionId parent;
/* Can't ask about stuff that might not be around anymore */
Assert(TransactionIdFollowsOrEquals(xid, TransactionXmin));
/* Bootstrap and frozen XIDs have no parent */
if (!TransactionIdIsNormal(xid))
return InvalidTransactionId;
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(SubTransCtl, pageno, xid);
ptr = (TransactionId *) SubTransCtl->shared->page_buffer[slotno];
ptr += entryno;
parent = *ptr;
LWLockRelease(SubtransControlLock);
return parent;
}
/*
* do a TransactionId -> txid conversion for an XID near the given epoch
*/
static txid
convert_xid(TransactionId xid, const TxidEpoch *state)
{
#ifndef INT64_IS_BUSTED
uint64 epoch;
/* return special xid's as-is */
if (!TransactionIdIsNormal(xid))
return (txid) xid;
/* xid can be on either side when near wrap-around */
epoch = (uint64) state->epoch;
if (xid > state->last_xid &&
TransactionIdPrecedes(xid, state->last_xid))
epoch--;
else if (xid < state->last_xid &&
TransactionIdFollows(xid, state->last_xid))
epoch++;
return (epoch << 32) | xid;
#else /* INT64_IS_BUSTED */
/* we can't do anything with the epoch, so ignore it */
return (txid) xid & MAX_TXID;
#endif /* INT64_IS_BUSTED */
}
/*
* TransactionIdGetCommitLSN
*
* This function returns an LSN that is late enough to be able
* to guarantee that if we flush up to the LSN returned then we
* will have flushed the transaction's commit record to disk.
*
* The result is not necessarily the exact LSN of the transaction's
* commit record! For example, for long-past transactions (those whose
* clog pages already migrated to disk), we'll return InvalidXLogRecPtr.
* Also, because we group transactions on the same clog page to conserve
* storage, we might return the LSN of a later transaction that falls into
* the same group.
*/
XLogRecPtr
TransactionIdGetCommitLSN(TransactionId xid)
{
XLogRecPtr result;
/*
* Currently, all uses of this function are for xids that were just
* reported to be committed by TransactionLogFetch, so we expect that
* checking TransactionLogFetch's cache will usually succeed and avoid an
* extra trip to shared memory.
*/
if (TransactionIdEquals(xid, cachedFetchXid))
return cachedCommitLSN;
/* Special XIDs are always known committed */
if (!TransactionIdIsNormal(xid))
return InvalidXLogRecPtr;
/*
* Get the transaction status.
*/
(void) TransactionIdGetStatus(xid, &result);
return result;
}
/*
* GetOldestXmin -- returns oldest transaction that was running
* when any current transaction was started.
*
* If allDbs is TRUE then all backends are considered; if allDbs is FALSE
* then only backends running in my own database are considered.
*
* This is used by VACUUM to decide which deleted tuples must be preserved
* in a table. allDbs = TRUE is needed for shared relations, but allDbs =
* FALSE is sufficient for non-shared relations, since only backends in my
* own database could ever see the tuples in them.
*
* Note: we include the currently running xids in the set of considered xids.
* This ensures that if a just-started xact has not yet set its snapshot,
* when it does set the snapshot it cannot set xmin less than what we compute.
*/
TransactionId
GetOldestXmin(bool allDbs)
{
SISeg *segP = shmInvalBuffer;
ProcState *stateP = segP->procState;
TransactionId result;
int index;
result = GetCurrentTransactionId();
LWLockAcquire(SInvalLock, LW_SHARED);
for (index = 0; index < segP->lastBackend; index++)
{
SHMEM_OFFSET pOffset = stateP[index].procStruct;
if (pOffset != INVALID_OFFSET)
{
PGPROC *proc = (PGPROC *) MAKE_PTR(pOffset);
if (allDbs || proc->databaseId == MyDatabaseId)
{
/* Fetch xid just once - see GetNewTransactionId */
TransactionId xid = proc->xid;
if (TransactionIdIsNormal(xid))
{
if (TransactionIdPrecedes(xid, result))
result = xid;
xid = proc->xmin;
if (TransactionIdIsNormal(xid))
if (TransactionIdPrecedes(xid, result))
result = xid;
}
}
}
}
LWLockRelease(SInvalLock);
return result;
}
/* Record lowest soon-prunable XID */
static void
heap_prune_record_prunable(PruneState *prstate, TransactionId xid)
{
/*
* This should exactly match the PageSetPrunable macro. We can't store
* directly into the page header yet, so we update working state.
*/
Assert(TransactionIdIsNormal(xid));
if (!TransactionIdIsValid(prstate->new_prune_xid) ||
TransactionIdPrecedes(xid, prstate->new_prune_xid))
prstate->new_prune_xid = xid;
}
/*
* xid_age - compute age of an XID (relative to current xact)
*/
Datum
xid_age(PG_FUNCTION_ARGS)
{
TransactionId xid = PG_GETARG_TRANSACTIONID(0);
TransactionId now = GetTopTransactionId();
/* Permanent XIDs are always infinitely old */
if (!TransactionIdIsNormal(xid))
PG_RETURN_INT32(INT_MAX);
PG_RETURN_INT32((int32) (now - xid));
}
/*
* Sets the commit timestamp of a single transaction.
*
* Must be called with CommitTsControlLock held
*/
static void
TransactionIdSetCommitTs(TransactionId xid, TimestampTz ts,
RepOriginId nodeid, int slotno)
{
int entryno = TransactionIdToCTsEntry(xid);
CommitTimestampEntry entry;
Assert(TransactionIdIsNormal(xid));
entry.time = ts;
entry.nodeid = nodeid;
memcpy(CommitTsCtl->shared->page_buffer[slotno] +
SizeOfCommitTimestampEntry * entryno,
&entry, SizeOfCommitTimestampEntry);
}
/*
* Get oldest Xid visible by any active transaction (global or local)
* Take in account global Xmin received from DTMD
*/
static TransactionId DtmGetOldestXmin(Relation rel, bool ignoreVacuum)
{
TransactionId localXmin = PgGetOldestXmin(rel, ignoreVacuum);
TransactionId globalXmin = dtm->minXid;
XTM_INFO("XTM: DtmGetOldestXmin localXmin=%d, globalXmin=%d\n", localXmin, globalXmin);
if (TransactionIdIsValid(globalXmin))
{
globalXmin -= vacuum_defer_cleanup_age;
if (!TransactionIdIsNormal(globalXmin))
globalXmin = FirstNormalTransactionId;
if (TransactionIdPrecedes(globalXmin, localXmin))
localXmin = globalXmin;
XTM_INFO("XTM: DtmGetOldestXmin adjusted localXmin=%d, globalXmin=%d\n", localXmin, globalXmin);
}
return localXmin;
}
/*
* TransactionLogFetch --- fetch commit status of specified transaction id
*/
static XidStatus
TransactionLogFetch(TransactionId transactionId)
{
XidStatus xidstatus;
XLogRecPtr xidlsn;
/*
* Before going to the commit log manager, check our single item cache to
* see if we didn't just check the transaction status a moment ago.
*/
if (TransactionIdEquals(transactionId, cachedFetchXid))
return cachedFetchXidStatus;
/*
* Also, check to see if the transaction ID is a permanent one.
*/
if (!TransactionIdIsNormal(transactionId))
{
if (TransactionIdEquals(transactionId, BootstrapTransactionId))
return TRANSACTION_STATUS_COMMITTED;
if (TransactionIdEquals(transactionId, FrozenTransactionId))
return TRANSACTION_STATUS_COMMITTED;
return TRANSACTION_STATUS_ABORTED;
}
/*
* Get the transaction status.
*/
xidstatus = TransactionIdGetStatus(transactionId, &xidlsn);
/*
* Cache it, but DO NOT cache status for unfinished or sub-committed
* transactions! We only cache status that is guaranteed not to change.
*/
if (xidstatus != TRANSACTION_STATUS_IN_PROGRESS &&
xidstatus != TRANSACTION_STATUS_SUB_COMMITTED)
{
cachedFetchXid = transactionId;
cachedFetchXidStatus = xidstatus;
cachedCommitLSN = xidlsn;
}
return xidstatus;
}
static void
SubTransGetData(TransactionId xid, SubTransData* subData)
{
MIRRORED_LOCK_DECLARE;
int pageno = TransactionIdToPage(xid);
int entryno = TransactionIdToEntry(xid);
int slotno;
SubTransData *ptr;
/* Can't ask about stuff that might not be around anymore */
Assert(TransactionIdFollowsOrEquals(xid, TransactionXmin));
/* Bootstrap and frozen XIDs have no parent and itself as topMostParent */
if (!TransactionIdIsNormal(xid))
{
subData->parent = InvalidTransactionId;
subData->topMostParent = xid;
return;
}
MIRRORED_LOCK;
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(SubTransCtl, pageno, xid, NULL);
ptr = (SubTransData *) SubTransCtl->shared->page_buffer[slotno];
ptr += entryno;
subData->parent = ptr->parent;
subData->topMostParent = ptr->topMostParent;
if ( subData->topMostParent == InvalidTransactionId )
{
/* Here means parent is Main XID, hence set parent itself as topMostParent */
subData->topMostParent = xid;
}
LWLockRelease(SubtransControlLock);
MIRRORED_UNLOCK;
return;
}
/* adapted from Postgres' txid.c#convert_xid function */
uint64 convert_xid(TransactionId xid) {
TxidEpoch state;
uint64 epoch;
GetNextXidAndEpoch(&state.last_xid, &state.epoch);
/* return special xid's as-is */
if (!TransactionIdIsNormal(xid))
return (uint64) xid;
/* xid can be on either side when near wrap-around */
epoch = (uint64) state.epoch;
if (xid > state.last_xid && TransactionIdPrecedes(xid, state.last_xid))
epoch--;
else if (xid < state.last_xid && TransactionIdFollows(xid, state.last_xid))
epoch++;
return (epoch << 32) | xid;
}
/*
* do a TransactionId -> txid conversion for an XID near the given epoch
*/
static txid
convert_xid(TransactionId xid, const TxidEpoch *state)
{
uint64 epoch;
/* return special xid's as-is */
if (!TransactionIdIsNormal(xid))
return (txid) xid;
/* xid can be on either side when near wrap-around */
epoch = (uint64) state->epoch;
if (xid > state->last_xid &&
TransactionIdPrecedes(xid, state->last_xid))
epoch--;
else if (xid < state->last_xid &&
TransactionIdFollows(xid, state->last_xid))
epoch++;
return (epoch << 32) | xid;
}
Datum
pg_last_committed_xact(PG_FUNCTION_ARGS)
{
TransactionId xid;
TimestampTz ts;
Datum values[2];
bool nulls[2];
TupleDesc tupdesc;
HeapTuple htup;
/* and construct a tuple with our data */
xid = GetLatestCommitTsData(&ts, NULL);
/*
* Construct a tuple descriptor for the result row. This must match this
* function's pg_proc entry!
*/
tupdesc = CreateTemplateTupleDesc(2, false);
TupleDescInitEntry(tupdesc, (AttrNumber) 1, "xid",
XIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 2, "timestamp",
TIMESTAMPTZOID, -1, 0);
tupdesc = BlessTupleDesc(tupdesc);
if (!TransactionIdIsNormal(xid))
{
memset(nulls, true, sizeof(nulls));
}
else
{
values[0] = TransactionIdGetDatum(xid);
nulls[0] = false;
values[1] = TimestampTzGetDatum(ts);
nulls[1] = false;
}
htup = heap_form_tuple(tupdesc, values, nulls);
PG_RETURN_DATUM(HeapTupleGetDatum(htup));
}
/*
* Update local Recent*Xmin variables taken in account MinXmin received from DTMD
*/
static void DtmUpdateRecentXmin(Snapshot snapshot)
{
TransactionId xmin = dtm->minXid;
XTM_INFO("XTM: DtmUpdateRecentXmin global xmin=%d, snapshot xmin %d\n", dtm->minXid, DtmSnapshot.xmin);
if (TransactionIdIsValid(xmin))
{
xmin -= vacuum_defer_cleanup_age;
if (!TransactionIdIsNormal(xmin))
xmin = FirstNormalTransactionId;
if (TransactionIdFollows(RecentGlobalDataXmin, xmin))
RecentGlobalDataXmin = xmin;
if (TransactionIdFollows(RecentGlobalXmin, xmin))
RecentGlobalXmin = xmin;
}
if (TransactionIdFollows(RecentXmin, snapshot->xmin))
{
ProcArrayInstallImportedXmin(snapshot->xmin, GetCurrentTransactionId());
RecentXmin = snapshot->xmin;
}
}
/*----------
* GetSnapshotData -- returns information about running transactions.
*
* The returned snapshot includes xmin (lowest still-running xact ID),
* xmax (next xact ID to be assigned), and a list of running xact IDs
* in the range xmin <= xid < xmax. It is used as follows:
* All xact IDs < xmin are considered finished.
* All xact IDs >= xmax are considered still running.
* For an xact ID xmin <= xid < xmax, consult list to see whether
* it is considered running or not.
* This ensures that the set of transactions seen as "running" by the
* current xact will not change after it takes the snapshot.
*
* We also compute the current global xmin (oldest xmin across all running
* transactions) and save it in RecentGlobalXmin. This is the same
* computation done by GetOldestXmin(TRUE). The xmin value is also stored
* into RecentXmin.
*----------
*/
Snapshot
GetSnapshotData(Snapshot snapshot, bool serializable)
{
SISeg *segP = shmInvalBuffer;
ProcState *stateP = segP->procState;
TransactionId xmin;
TransactionId xmax;
TransactionId globalxmin;
int index;
int count = 0;
Assert(snapshot != NULL);
/*
* Allocating space for MaxBackends xids is usually overkill;
* lastBackend would be sufficient. But it seems better to do the
* malloc while not holding the lock, so we can't look at lastBackend.
*
* This does open a possibility for avoiding repeated malloc/free:
* since MaxBackends does not change at runtime, we can simply reuse
* the previous xip array if any. (This relies on the fact that all
* calls pass static SnapshotData structs.)
*/
if (snapshot->xip == NULL)
{
/*
* First call for this snapshot
*/
snapshot->xip = (TransactionId *)
malloc(MaxBackends * sizeof(TransactionId));
if (snapshot->xip == NULL)
ereport(ERROR,
(errcode(ERRCODE_OUT_OF_MEMORY),
errmsg("out of memory")));
}
globalxmin = xmin = GetCurrentTransactionId();
/*
* If we are going to set MyProc->xmin then we'd better get exclusive
* lock; if not, this is a read-only operation so it can be shared.
*/
LWLockAcquire(SInvalLock, serializable ? LW_EXCLUSIVE : LW_SHARED);
/*--------------------
* Unfortunately, we have to call ReadNewTransactionId() after acquiring
* SInvalLock above. It's not good because ReadNewTransactionId() does
* LWLockAcquire(XidGenLock), but *necessary*. We need to be sure that
* no transactions exit the set of currently-running transactions
* between the time we fetch xmax and the time we finish building our
* snapshot. Otherwise we could have a situation like this:
*
* 1. Tx Old is running (in Read Committed mode).
* 2. Tx S reads new transaction ID into xmax, then
* is swapped out before acquiring SInvalLock.
* 3. Tx New gets new transaction ID (>= S' xmax),
* makes changes and commits.
* 4. Tx Old changes some row R changed by Tx New and commits.
* 5. Tx S finishes getting its snapshot data. It sees Tx Old as
* done, but sees Tx New as still running (since New >= xmax).
*
* Now S will see R changed by both Tx Old and Tx New, *but* does not
* see other changes made by Tx New. If S is supposed to be in
* Serializable mode, this is wrong.
*
* By locking SInvalLock before we read xmax, we ensure that TX Old
* cannot exit the set of running transactions seen by Tx S. Therefore
* both Old and New will be seen as still running => no inconsistency.
*--------------------
*/
xmax = ReadNewTransactionId();
for (index = 0; index < segP->lastBackend; index++)
{
SHMEM_OFFSET pOffset = stateP[index].procStruct;
if (pOffset != INVALID_OFFSET)
{
PGPROC *proc = (PGPROC *) MAKE_PTR(pOffset);
/* Fetch xid just once - see GetNewTransactionId */
TransactionId xid = proc->xid;
/*
* Ignore my own proc (dealt with my xid above), procs not
* running a transaction, and xacts started since we read the
* next transaction ID. There's no need to store XIDs above
* what we got from ReadNewTransactionId, since we'll treat
* them as running anyway. We also assume that such xacts
* can't compute an xmin older than ours, so they needn't be
* considered in computing globalxmin.
*/
if (proc == MyProc ||
!TransactionIdIsNormal(xid) ||
TransactionIdFollowsOrEquals(xid, xmax))
continue;
if (TransactionIdPrecedes(xid, xmin))
xmin = xid;
snapshot->xip[count] = xid;
count++;
/* Update globalxmin to be the smallest valid xmin */
xid = proc->xmin;
if (TransactionIdIsNormal(xid))
if (TransactionIdPrecedes(xid, globalxmin))
globalxmin = xid;
}
}
if (serializable)
MyProc->xmin = xmin;
LWLockRelease(SInvalLock);
/* Serializable snapshot must be computed before any other... */
Assert(TransactionIdIsValid(MyProc->xmin));
/*
* Update globalxmin to include actual process xids. This is a
* slightly different way of computing it than GetOldestXmin uses, but
* should give the same result.
*/
if (TransactionIdPrecedes(xmin, globalxmin))
globalxmin = xmin;
/* Update globals for use by VACUUM */
RecentGlobalXmin = globalxmin;
RecentXmin = xmin;
snapshot->xmin = xmin;
snapshot->xmax = xmax;
snapshot->xcnt = count;
snapshot->curcid = GetCurrentCommandId();
return snapshot;
}
/*
* Interrogate the commit timestamp of a transaction.
*
* The return value indicates whether a commit timestamp record was found for
* the given xid. The timestamp value is returned in *ts (which may not be
* null), and the origin node for the Xid is returned in *nodeid, if it's not
* null.
*/
bool
TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
RepOriginId *nodeid)
{
int pageno = TransactionIdToCTsPage(xid);
int entryno = TransactionIdToCTsEntry(xid);
int slotno;
CommitTimestampEntry entry;
TransactionId oldestCommitTsXid;
TransactionId newestCommitTsXid;
if (!TransactionIdIsValid(xid))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("cannot retrieve commit timestamp for transaction %u", xid)));
else if (!TransactionIdIsNormal(xid))
{
/* frozen and bootstrap xids are always committed far in the past */
*ts = 0;
if (nodeid)
*nodeid = 0;
return false;
}
LWLockAcquire(CommitTsLock, LW_SHARED);
/* Error if module not enabled */
if (!commitTsShared->commitTsActive)
error_commit_ts_disabled();
/*
* If we're asked for the cached value, return that. Otherwise, fall
* through to read from SLRU.
*/
if (commitTsShared->xidLastCommit == xid)
{
*ts = commitTsShared->dataLastCommit.time;
if (nodeid)
*nodeid = commitTsShared->dataLastCommit.nodeid;
LWLockRelease(CommitTsLock);
return *ts != 0;
}
oldestCommitTsXid = ShmemVariableCache->oldestCommitTsXid;
newestCommitTsXid = ShmemVariableCache->newestCommitTsXid;
/* neither is invalid, or both are */
Assert(TransactionIdIsValid(oldestCommitTsXid) == TransactionIdIsValid(newestCommitTsXid));
LWLockRelease(CommitTsLock);
/*
* Return empty if the requested value is outside our valid range.
*/
if (!TransactionIdIsValid(oldestCommitTsXid) ||
TransactionIdPrecedes(xid, oldestCommitTsXid) ||
TransactionIdPrecedes(newestCommitTsXid, xid))
{
*ts = 0;
if (nodeid)
*nodeid = InvalidRepOriginId;
return false;
}
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
CommitTsCtl->shared->page_buffer[slotno] +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
*ts = entry.time;
if (nodeid)
*nodeid = entry.nodeid;
LWLockRelease(CommitTsControlLock);
return *ts != 0;
}
/*
* write_database_file: update the flat database file
*
* A side effect is to determine the oldest database's datfrozenxid
* so we can set or update the XID wrap limit.
*
* Also, if "startup" is true, we tell relcache.c to clear out the relcache
* init file in each database. That's a bit nonmodular, but scanning
* pg_database twice during system startup seems too high a price for keeping
* things better separated.
*/
static void
write_database_file(Relation drel, bool startup)
{
StringInfoData buffer;
HeapScanDesc scan;
HeapTuple tuple;
NameData oldest_datname;
TransactionId oldest_datfrozenxid = InvalidTransactionId;
MirroredFlatFileOpen mirroredOpen;
initStringInfo(&buffer);
MirroredFlatFile_Open(
&mirroredOpen,
"global",
"pg_database",
O_CREAT | O_TRUNC | O_WRONLY | PG_BINARY,
S_IRUSR | S_IWUSR,
/* suppressError */ false,
/* atomic operation */ true,
/*isMirrorRecovery */ false);
/*
* Read pg_database and write the file.
*/
scan = heap_beginscan(drel, SnapshotNow, 0, NULL);
while ((tuple = heap_getnext(scan, ForwardScanDirection)) != NULL)
{
Form_pg_database dbform = (Form_pg_database) GETSTRUCT(tuple);
char *datname;
Oid datoid;
Oid dattablespace;
TransactionId datfrozenxid;
datname = NameStr(dbform->datname);
datoid = HeapTupleGetOid(tuple);
dattablespace = dbform->dattablespace;
datfrozenxid = dbform->datfrozenxid;
/*
* Identify the oldest datfrozenxid. This must match
* the logic in vac_truncate_clog() in vacuum.c.
*
* MPP-20053: Skip databases that cannot be connected to in computing
* the oldest database.
*/
if (dbform->datallowconn && TransactionIdIsNormal(datfrozenxid))
{
if (oldest_datfrozenxid == InvalidTransactionId ||
TransactionIdPrecedes(datfrozenxid, oldest_datfrozenxid))
{
oldest_datfrozenxid = datfrozenxid;
namestrcpy(&oldest_datname, datname);
}
}
/*
* Check for illegal characters in the database name.
*/
if (!name_okay(datname))
{
ereport(LOG,
(errmsg("invalid database name \"%s\"", datname)));
continue;
}
/*
* The file format is: "dbname" oid tablespace frozenxid
*
* The xids are not needed for backend startup, but are of use to
* autovacuum, and might also be helpful for forensic purposes.
*/
sputs_quote(&buffer, datname);
appendStringInfo(&buffer, " %u %u %u\n",
datoid, dattablespace, datfrozenxid);
/*
* MPP-10111 - During database expansion we need to be able to bring a
* database up in order to correct the filespace locations in the
* catalog. At this point we will not be able to resolve database paths
* for databases not stored in "pg_default" or "pg_global".
*
* This is solved by passing a special guc to the startup during this
* phase of expand to bypass logic involving non-system tablespaces.
* Since we are bypassing the clearing of the relation cache on these
* databases we need to ensure that we don't try to use them at all
* elsewhere. This is done with a similar check in
* PersistentTablespace_GetPrimaryAndMirrorFilespaces().
*/
if (gp_before_filespace_setup && !IsBuiltinTablespace(dattablespace))
continue;
}
heap_endscan(scan);
MirroredFlatFile_Append(&mirroredOpen, buffer.data, buffer.len,
/* suppressError */ false);
MirroredFlatFile_Flush(&mirroredOpen, /* suppressError */ false);
MirroredFlatFile_Close(&mirroredOpen);
if (buffer.maxlen > 0)
pfree(buffer.data);
/*
* Set the transaction ID wrap limit using the oldest datfrozenxid
*/
if (oldest_datfrozenxid != InvalidTransactionId)
SetTransactionIdLimit(oldest_datfrozenxid, &oldest_datname);
}
/*
* Hot Standby feedback
*/
static void
ProcessStandbyHSFeedbackMessage(void)
{
StandbyHSFeedbackMessage reply;
TransactionId nextXid;
uint32 nextEpoch;
/* Decipher the reply message */
pq_copymsgbytes(&reply_message, (char *) &reply,
sizeof(StandbyHSFeedbackMessage));
elog(DEBUG2, "hot standby feedback xmin %u epoch %u",
reply.xmin,
reply.epoch);
/* Ignore invalid xmin (can't actually happen with current walreceiver) */
if (!TransactionIdIsNormal(reply.xmin))
return;
/*
* Check that the provided xmin/epoch are sane, that is, not in the future
* and not so far back as to be already wrapped around. Ignore if not.
*
* Epoch of nextXid should be same as standby, or if the counter has
* wrapped, then one greater than standby.
*/
GetNextXidAndEpoch(&nextXid, &nextEpoch);
if (reply.xmin <= nextXid)
{
if (reply.epoch != nextEpoch)
return;
}
else
{
if (reply.epoch + 1 != nextEpoch)
return;
}
if (!TransactionIdPrecedesOrEquals(reply.xmin, nextXid))
return; /* epoch OK, but it's wrapped around */
/*
* Set the WalSender's xmin equal to the standby's requested xmin, so that
* the xmin will be taken into account by GetOldestXmin. This will hold
* back the removal of dead rows and thereby prevent the generation of
* cleanup conflicts on the standby server.
*
* There is a small window for a race condition here: although we just
* checked that reply.xmin precedes nextXid, the nextXid could have gotten
* advanced between our fetching it and applying the xmin below, perhaps
* far enough to make reply.xmin wrap around. In that case the xmin we
* set here would be "in the future" and have no effect. No point in
* worrying about this since it's too late to save the desired data
* anyway. Assuming that the standby sends us an increasing sequence of
* xmins, this could only happen during the first reply cycle, else our
* own xmin would prevent nextXid from advancing so far.
*
* We don't bother taking the ProcArrayLock here. Setting the xmin field
* is assumed atomic, and there's no real need to prevent a concurrent
* GetOldestXmin. (If we're moving our xmin forward, this is obviously
* safe, and if we're moving it backwards, well, the data is at risk
* already since a VACUUM could have just finished calling GetOldestXmin.)
*/
MyPgXact->xmin = reply.xmin;
}
/*
* This is a copy of swap_relation_files in cluster.c, but it also swaps
* relfrozenxid.
*/
static void
swap_heap_or_index_files(Oid r1, Oid r2)
{
Relation relRelation;
HeapTuple reltup1,
reltup2;
Form_pg_class relform1,
relform2;
Oid swaptemp;
CatalogIndexState indstate;
/* We need writable copies of both pg_class tuples. */
relRelation = heap_open(RelationRelationId, RowExclusiveLock);
reltup1 = SearchSysCacheCopy(RELOID,
ObjectIdGetDatum(r1),
0, 0, 0);
if (!HeapTupleIsValid(reltup1))
elog(ERROR, "cache lookup failed for relation %u", r1);
relform1 = (Form_pg_class) GETSTRUCT(reltup1);
reltup2 = SearchSysCacheCopy(RELOID,
ObjectIdGetDatum(r2),
0, 0, 0);
if (!HeapTupleIsValid(reltup2))
elog(ERROR, "cache lookup failed for relation %u", r2);
relform2 = (Form_pg_class) GETSTRUCT(reltup2);
Assert(relform1->relkind == relform2->relkind);
/*
* Actually swap the fields in the two tuples
*/
swaptemp = relform1->relfilenode;
relform1->relfilenode = relform2->relfilenode;
relform2->relfilenode = swaptemp;
swaptemp = relform1->reltablespace;
relform1->reltablespace = relform2->reltablespace;
relform2->reltablespace = swaptemp;
swaptemp = relform1->reltoastrelid;
relform1->reltoastrelid = relform2->reltoastrelid;
relform2->reltoastrelid = swaptemp;
/* set rel1's frozen Xid to larger one */
if (TransactionIdIsNormal(relform1->relfrozenxid))
{
if (TransactionIdFollows(relform1->relfrozenxid,
relform2->relfrozenxid))
relform1->relfrozenxid = relform2->relfrozenxid;
else
relform2->relfrozenxid = relform1->relfrozenxid;
}
/* swap size statistics too, since new rel has freshly-updated stats */
{
#if PG_VERSION_NUM >= 90300
int32 swap_pages;
#else
int4 swap_pages;
#endif
float4 swap_tuples;
swap_pages = relform1->relpages;
relform1->relpages = relform2->relpages;
relform2->relpages = swap_pages;
swap_tuples = relform1->reltuples;
relform1->reltuples = relform2->reltuples;
relform2->reltuples = swap_tuples;
}
/* Update the tuples in pg_class */
simple_heap_update(relRelation, &reltup1->t_self, reltup1);
simple_heap_update(relRelation, &reltup2->t_self, reltup2);
/* Keep system catalogs current */
indstate = CatalogOpenIndexes(relRelation);
CatalogIndexInsert(indstate, reltup1);
CatalogIndexInsert(indstate, reltup2);
CatalogCloseIndexes(indstate);
/*
* If we have toast tables associated with the relations being swapped,
* change their dependency links to re-associate them with their new
* owning relations. Otherwise the wrong one will get dropped ...
*
* NOTE: it is possible that only one table has a toast table; this can
* happen in CLUSTER if there were dropped columns in the old table, and
* in ALTER TABLE when adding or changing type of columns.
*
* NOTE: at present, a TOAST table's only dependency is the one on its
* owning table. If more are ever created, we'd need to use something
* more selective than deleteDependencyRecordsFor() to get rid of only the
* link we want.
*/
if (relform1->reltoastrelid || relform2->reltoastrelid)
{
ObjectAddress baseobject,
toastobject;
long count;
/* Delete old dependencies */
if (relform1->reltoastrelid)
{
count = deleteDependencyRecordsFor(RelationRelationId,
relform1->reltoastrelid,
false);
if (count != 1)
elog(ERROR, "expected one dependency record for TOAST table, found %ld",
count);
}
if (relform2->reltoastrelid)
{
count = deleteDependencyRecordsFor(RelationRelationId,
relform2->reltoastrelid,
false);
if (count != 1)
elog(ERROR, "expected one dependency record for TOAST table, found %ld",
count);
}
/* Register new dependencies */
baseobject.classId = RelationRelationId;
baseobject.objectSubId = 0;
toastobject.classId = RelationRelationId;
toastobject.objectSubId = 0;
if (relform1->reltoastrelid)
{
baseobject.objectId = r1;
toastobject.objectId = relform1->reltoastrelid;
recordDependencyOn(&toastobject, &baseobject, DEPENDENCY_INTERNAL);
}
if (relform2->reltoastrelid)
{
baseobject.objectId = r2;
toastobject.objectId = relform2->reltoastrelid;
recordDependencyOn(&toastobject, &baseobject, DEPENDENCY_INTERNAL);
}
}
/*
* Blow away the old relcache entries now. We need this kluge because
* relcache.c keeps a link to the smgr relation for the physical file, and
* that will be out of date as soon as we do CommandCounterIncrement.
* Whichever of the rels is the second to be cleared during cache
* invalidation will have a dangling reference to an already-deleted smgr
* relation. Rather than trying to avoid this by ordering operations just
* so, it's easiest to not have the relcache entries there at all.
* (Fortunately, since one of the entries is local in our transaction,
* it's sufficient to clear out our own relcache this way; the problem
* cannot arise for other backends when they see our update on the
* non-local relation.)
*/
RelationForgetRelation(r1);
RelationForgetRelation(r2);
/* Clean up. */
heap_freetuple(reltup1);
heap_freetuple(reltup2);
heap_close(relRelation, RowExclusiveLock);
}
/*
* Interrogate the commit timestamp of a transaction.
*
* The return value indicates whether a commit timestamp record was found for
* the given xid. The timestamp value is returned in *ts (which may not be
* null), and the origin node for the Xid is returned in *nodeid, if it's not
* null.
*/
bool
TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
RepOriginId *nodeid)
{
int pageno = TransactionIdToCTsPage(xid);
int entryno = TransactionIdToCTsEntry(xid);
int slotno;
CommitTimestampEntry entry;
TransactionId oldestCommitTs;
TransactionId newestCommitTs;
/* error if the given Xid doesn't normally commit */
if (!TransactionIdIsNormal(xid))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("cannot retrieve commit timestamp for transaction %u", xid)));
LWLockAcquire(CommitTsLock, LW_SHARED);
/* Error if module not enabled */
if (!commitTsShared->commitTsActive)
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("could not get commit timestamp data"),
errhint("Make sure the configuration parameter \"%s\" is set.",
"track_commit_timestamp")));
/*
* If we're asked for the cached value, return that. Otherwise, fall
* through to read from SLRU.
*/
if (commitTsShared->xidLastCommit == xid)
{
*ts = commitTsShared->dataLastCommit.time;
if (nodeid)
*nodeid = commitTsShared->dataLastCommit.nodeid;
LWLockRelease(CommitTsLock);
return *ts != 0;
}
oldestCommitTs = ShmemVariableCache->oldestCommitTs;
newestCommitTs = ShmemVariableCache->newestCommitTs;
/* neither is invalid, or both are */
Assert(TransactionIdIsValid(oldestCommitTs) == TransactionIdIsValid(newestCommitTs));
LWLockRelease(CommitTsLock);
/*
* Return empty if the requested value is outside our valid range.
*/
if (!TransactionIdIsValid(oldestCommitTs) ||
TransactionIdPrecedes(xid, oldestCommitTs) ||
TransactionIdPrecedes(newestCommitTs, xid))
{
*ts = 0;
if (nodeid)
*nodeid = InvalidRepOriginId;
return false;
}
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
CommitTsCtl->shared->page_buffer[slotno] +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
*ts = entry.time;
if (nodeid)
*nodeid = entry.nodeid;
LWLockRelease(CommitTsControlLock);
return *ts != 0;
}
/*
* ImportSnapshot
* Import a previously exported snapshot. The argument should be a
* filename in SNAPSHOT_EXPORT_DIR. Load the snapshot from that file.
* This is called by "SET TRANSACTION SNAPSHOT 'foo'".
*/
void
ImportSnapshot(const char *idstr)
{
char path[MAXPGPATH];
FILE *f;
struct stat stat_buf;
char *filebuf;
int xcnt;
int i;
TransactionId src_xid;
Oid src_dbid;
int src_isolevel;
bool src_readonly;
SnapshotData snapshot;
/*
* Must be at top level of a fresh transaction. Note in particular that
* we check we haven't acquired an XID --- if we have, it's conceivable
* that the snapshot would show it as not running, making for very
* screwy behavior.
*/
if (FirstSnapshotSet ||
GetTopTransactionIdIfAny() != InvalidTransactionId ||
IsSubTransaction())
ereport(ERROR,
(errcode(ERRCODE_ACTIVE_SQL_TRANSACTION),
errmsg("SET TRANSACTION SNAPSHOT must be called before any query")));
/*
* If we are in read committed mode then the next query would execute
* with a new snapshot thus making this function call quite useless.
*/
if (!IsolationUsesXactSnapshot())
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("a snapshot-importing transaction must have isolation level SERIALIZABLE or REPEATABLE READ")));
/*
* Verify the identifier: only 0-9, A-F and hyphens are allowed. We do
* this mainly to prevent reading arbitrary files.
*/
if (strspn(idstr, "0123456789ABCDEF-") != strlen(idstr))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("invalid snapshot identifier \"%s\"", idstr)));
/* OK, read the file */
snprintf(path, MAXPGPATH, SNAPSHOT_EXPORT_DIR "/%s", idstr);
f = AllocateFile(path, PG_BINARY_R);
if (!f)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("invalid snapshot identifier \"%s\"", idstr)));
/* get the size of the file so that we know how much memory we need */
if (fstat(fileno(f), &stat_buf))
elog(ERROR, "could not stat file \"%s\": %m", path);
/* and read the file into a palloc'd string */
filebuf = (char *) palloc(stat_buf.st_size + 1);
if (fread(filebuf, stat_buf.st_size, 1, f) != 1)
elog(ERROR, "could not read file \"%s\": %m", path);
filebuf[stat_buf.st_size] = '\0';
FreeFile(f);
/*
* Construct a snapshot struct by parsing the file content.
*/
memset(&snapshot, 0, sizeof(snapshot));
src_xid = parseXidFromText("xid:", &filebuf, path);
/* we abuse parseXidFromText a bit here ... */
src_dbid = parseXidFromText("dbid:", &filebuf, path);
src_isolevel = parseIntFromText("iso:", &filebuf, path);
src_readonly = parseIntFromText("ro:", &filebuf, path);
snapshot.xmin = parseXidFromText("xmin:", &filebuf, path);
snapshot.xmax = parseXidFromText("xmax:", &filebuf, path);
snapshot.xcnt = xcnt = parseIntFromText("xcnt:", &filebuf, path);
/* sanity-check the xid count before palloc */
if (xcnt < 0 || xcnt > GetMaxSnapshotXidCount())
ereport(ERROR,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
errmsg("invalid snapshot data in file \"%s\"", path)));
snapshot.xip = (TransactionId *) palloc(xcnt * sizeof(TransactionId));
for (i = 0; i < xcnt; i++)
snapshot.xip[i] = parseXidFromText("xip:", &filebuf, path);
snapshot.suboverflowed = parseIntFromText("sof:", &filebuf, path);
if (!snapshot.suboverflowed)
{
snapshot.subxcnt = xcnt = parseIntFromText("sxcnt:", &filebuf, path);
/* sanity-check the xid count before palloc */
if (xcnt < 0 || xcnt > GetMaxSnapshotSubxidCount())
ereport(ERROR,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
errmsg("invalid snapshot data in file \"%s\"", path)));
snapshot.subxip = (TransactionId *) palloc(xcnt * sizeof(TransactionId));
for (i = 0; i < xcnt; i++)
snapshot.subxip[i] = parseXidFromText("sxp:", &filebuf, path);
}
else
{
snapshot.subxcnt = 0;
snapshot.subxip = NULL;
}
snapshot.takenDuringRecovery = parseIntFromText("rec:", &filebuf, path);
/*
* Do some additional sanity checking, just to protect ourselves. We
* don't trouble to check the array elements, just the most critical
* fields.
*/
if (!TransactionIdIsNormal(src_xid) ||
!OidIsValid(src_dbid) ||
!TransactionIdIsNormal(snapshot.xmin) ||
!TransactionIdIsNormal(snapshot.xmax))
ereport(ERROR,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
errmsg("invalid snapshot data in file \"%s\"", path)));
/*
* If we're serializable, the source transaction must be too, otherwise
* predicate.c has problems (SxactGlobalXmin could go backwards). Also,
* a non-read-only transaction can't adopt a snapshot from a read-only
* transaction, as predicate.c handles the cases very differently.
*/
if (IsolationIsSerializable())
{
if (src_isolevel != XACT_SERIALIZABLE)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("a serializable transaction cannot import a snapshot from a non-serializable transaction")));
if (src_readonly && !XactReadOnly)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("a non-read-only serializable transaction cannot import a snapshot from a read-only transaction")));
}
/*
* We cannot import a snapshot that was taken in a different database,
* because vacuum calculates OldestXmin on a per-database basis; so the
* source transaction's xmin doesn't protect us from data loss. This
* restriction could be removed if the source transaction were to mark
* its xmin as being globally applicable. But that would require some
* additional syntax, since that has to be known when the snapshot is
* initially taken. (See pgsql-hackers discussion of 2011-10-21.)
*/
if (src_dbid != MyDatabaseId)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("cannot import a snapshot from a different database")));
/* OK, install the snapshot */
SetTransactionSnapshot(&snapshot, src_xid);
}