/*
* Report the current WAL flush location (same format as pg_start_backup etc)
*
* This function is mostly for debugging purposes.
*/
Datum
pg_current_xlog_flush_location(PG_FUNCTION_ARGS)
{
XLogRecPtr current_recptr;
if (RecoveryInProgress())
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("recovery is in progress"),
errhint("WAL control functions cannot be executed during recovery.")));
current_recptr = GetFlushRecPtr();
PG_RETURN_LSN(current_recptr);
}
/*
* read_page callback for logical decoding contexts.
*
* Public because it would likely be very helpful for someone writing another
* output method outside walsender, e.g. in a bgworker.
*
* TODO: The walsender has it's own version of this, but it relies on the
* walsender's latch being set whenever WAL is flushed. No such infrastructure
* exists for normal backends, so we have to do a check/sleep/repeat style of
* loop for now.
*/
int
logical_read_local_xlog_page(XLogReaderState *state, XLogRecPtr targetPagePtr,
int reqLen, XLogRecPtr targetRecPtr, char *cur_page, TimeLineID *pageTLI)
{
XLogRecPtr flushptr,
loc;
int count;
loc = targetPagePtr + reqLen;
while (1)
{
/*
* TODO: we're going to have to do something more intelligent about
* timelines on standbys. Use readTimeLineHistory() and
* tliOfPointInHistory() to get the proper LSN? For now we'll catch
* that case earlier, but the code and TODO is left in here for when
* that changes.
*/
if (!RecoveryInProgress())
{
*pageTLI = ThisTimeLineID;
flushptr = GetFlushRecPtr();
}
else
flushptr = GetXLogReplayRecPtr(pageTLI);
if (loc <= flushptr)
break;
CHECK_FOR_INTERRUPTS();
pg_usleep(1000L);
}
/* more than one block available */
if (targetPagePtr + XLOG_BLCKSZ <= flushptr)
count = XLOG_BLCKSZ;
/* not enough data there */
else if (targetPagePtr + reqLen > flushptr)
return -1;
/* part of the page available */
else
count = flushptr - targetPagePtr;
XLogRead(cur_page, *pageTLI, targetPagePtr, XLOG_BLCKSZ);
return count;
}
/*
* Read up to MAX_SEND_SIZE bytes of WAL that's been flushed to disk,
* but not yet sent to the client, and buffer it in the libpq output
* buffer.
*
* msgbuf is a work area in which the output message is constructed. It's
* passed in just so we can avoid re-palloc'ing the buffer on each cycle.
* It must be of size 1 + sizeof(WalDataMessageHeader) + MAX_SEND_SIZE.
*
* If there is no unsent WAL remaining, *caughtup is set to true, otherwise
* *caughtup is set to false.
*/
static void
XLogSend(char *msgbuf, bool *caughtup)
{
XLogRecPtr SendRqstPtr;
XLogRecPtr startptr;
XLogRecPtr endptr;
Size nbytes;
WalDataMessageHeader msghdr;
/*
* Attempt to send all data that's already been written out and fsync'd to
* disk. We cannot go further than what's been written out given the
* current implementation of XLogRead(). And in any case it's unsafe to
* send WAL that is not securely down to disk on the master: if the master
* subsequently crashes and restarts, slaves must not have applied any WAL
* that gets lost on the master.
*/
SendRqstPtr = am_cascading_walsender ? GetStandbyFlushRecPtr() : GetFlushRecPtr();
/* Quick exit if nothing to do */
if (XLByteLE(SendRqstPtr, sentPtr))
{
*caughtup = true;
return;
}
/*
* Figure out how much to send in one message. If there's no more than
* MAX_SEND_SIZE bytes to send, send everything. Otherwise send
* MAX_SEND_SIZE bytes, but round back to logfile or page boundary.
*
* The rounding is not only for performance reasons. Walreceiver relies on
* the fact that we never split a WAL record across two messages. Since a
* long WAL record is split at page boundary into continuation records,
* page boundary is always a safe cut-off point. We also assume that
* SendRqstPtr never points to the middle of a WAL record.
*/
startptr = sentPtr;
if (startptr.xrecoff >= XLogFileSize)
{
/*
* crossing a logid boundary, skip the non-existent last log segment
* in previous logical log file.
*/
startptr.xlogid += 1;
startptr.xrecoff = 0;
}
endptr = startptr;
XLByteAdvance(endptr, MAX_SEND_SIZE);
if (endptr.xlogid != startptr.xlogid)
{
/* Don't cross a logfile boundary within one message */
Assert(endptr.xlogid == startptr.xlogid + 1);
endptr.xlogid = startptr.xlogid;
endptr.xrecoff = XLogFileSize;
}
/* if we went beyond SendRqstPtr, back off */
if (XLByteLE(SendRqstPtr, endptr))
{
endptr = SendRqstPtr;
*caughtup = true;
}
else
{
/* round down to page boundary. */
endptr.xrecoff -= (endptr.xrecoff % XLOG_BLCKSZ);
*caughtup = false;
}
nbytes = endptr.xrecoff - startptr.xrecoff;
Assert(nbytes <= MAX_SEND_SIZE);
/*
* OK to read and send the slice.
*/
msgbuf[0] = 'w';
/*
* Read the log directly into the output buffer to avoid extra memcpy
* calls.
*/
XLogRead(msgbuf + 1 + sizeof(WalDataMessageHeader), startptr, nbytes);
/*
* We fill the message header last so that the send timestamp is taken as
* late as possible.
*/
msghdr.dataStart = startptr;
msghdr.walEnd = SendRqstPtr;
msghdr.sendTime = GetCurrentTimestamp();
memcpy(msgbuf + 1, &msghdr, sizeof(WalDataMessageHeader));
pq_putmessage_noblock('d', msgbuf, 1 + sizeof(WalDataMessageHeader) + nbytes);
sentPtr = endptr;
/* Update shared memory status */
{
/* use volatile pointer to prevent code rearrangement */
volatile WalSnd *walsnd = MyWalSnd;
SpinLockAcquire(&walsnd->mutex);
walsnd->sentPtr = sentPtr;
SpinLockRelease(&walsnd->mutex);
}
/* Report progress of XLOG streaming in PS display */
if (update_process_title)
{
char activitymsg[50];
snprintf(activitymsg, sizeof(activitymsg), "streaming %X/%X",
sentPtr.xlogid, sentPtr.xrecoff);
set_ps_display(activitymsg, false);
}
return;
}
/*
* Helper function for the various SQL callable logical decoding functions.
*/
static Datum
pg_logical_slot_get_changes_guts(FunctionCallInfo fcinfo, bool confirm, bool binary)
{
Name name;
XLogRecPtr upto_lsn;
int32 upto_nchanges;
ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
MemoryContext per_query_ctx;
MemoryContext oldcontext;
XLogRecPtr end_of_wal;
XLogRecPtr startptr;
LogicalDecodingContext *ctx;
ResourceOwner old_resowner = CurrentResourceOwner;
ArrayType *arr;
Size ndim;
List *options = NIL;
DecodingOutputState *p;
check_permissions();
CheckLogicalDecodingRequirements();
if (PG_ARGISNULL(0))
ereport(ERROR,
(errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
errmsg("slot name must not be null")));
name = PG_GETARG_NAME(0);
if (PG_ARGISNULL(1))
upto_lsn = InvalidXLogRecPtr;
else
upto_lsn = PG_GETARG_LSN(1);
if (PG_ARGISNULL(2))
upto_nchanges = InvalidXLogRecPtr;
else
upto_nchanges = PG_GETARG_INT32(2);
if (PG_ARGISNULL(3))
ereport(ERROR,
(errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
errmsg("options array must not be null")));
arr = PG_GETARG_ARRAYTYPE_P(3);
/* check to see if caller supports us returning a tuplestore */
if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("set-valued function called in context that cannot accept a set")));
if (!(rsinfo->allowedModes & SFRM_Materialize))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("materialize mode required, but it is not allowed in this context")));
/* state to write output to */
p = palloc0(sizeof(DecodingOutputState));
p->binary_output = binary;
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &p->tupdesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
per_query_ctx = rsinfo->econtext->ecxt_per_query_memory;
oldcontext = MemoryContextSwitchTo(per_query_ctx);
/* Deconstruct options array */
ndim = ARR_NDIM(arr);
if (ndim > 1)
{
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("array must be one-dimensional")));
}
else if (array_contains_nulls(arr))
{
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("array must not contain nulls")));
}
else if (ndim == 1)
{
int nelems;
Datum *datum_opts;
int i;
Assert(ARR_ELEMTYPE(arr) == TEXTOID);
deconstruct_array(arr, TEXTOID, -1, false, 'i',
&datum_opts, NULL, &nelems);
if (nelems % 2 != 0)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("array must have even number of elements")));
for (i = 0; i < nelems; i += 2)
{
char *name = TextDatumGetCString(datum_opts[i]);
char *opt = TextDatumGetCString(datum_opts[i + 1]);
options = lappend(options, makeDefElem(name, (Node *) makeString(opt), -1));
}
}
p->tupstore = tuplestore_begin_heap(true, false, work_mem);
rsinfo->returnMode = SFRM_Materialize;
rsinfo->setResult = p->tupstore;
rsinfo->setDesc = p->tupdesc;
/*
* Compute the current end-of-wal and maintain ThisTimeLineID.
* RecoveryInProgress() will update ThisTimeLineID on promotion.
*/
if (!RecoveryInProgress())
end_of_wal = GetFlushRecPtr();
else
end_of_wal = GetXLogReplayRecPtr(&ThisTimeLineID);
ReplicationSlotAcquire(NameStr(*name), true);
PG_TRY();
{
/* restart at slot's confirmed_flush */
ctx = CreateDecodingContext(InvalidXLogRecPtr,
options,
false,
logical_read_local_xlog_page,
LogicalOutputPrepareWrite,
LogicalOutputWrite, NULL);
MemoryContextSwitchTo(oldcontext);
/*
* Check whether the output plugin writes textual output if that's
* what we need.
*/
if (!binary &&
ctx->options.output_type !=OUTPUT_PLUGIN_TEXTUAL_OUTPUT)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("logical decoding output plugin \"%s\" produces binary output, but function \"%s\" expects textual data",
NameStr(MyReplicationSlot->data.plugin),
format_procedure(fcinfo->flinfo->fn_oid))));
ctx->output_writer_private = p;
/*
* Decoding of WAL must start at restart_lsn so that the entirety of
* xacts that committed after the slot's confirmed_flush can be
* accumulated into reorder buffers.
*/
startptr = MyReplicationSlot->data.restart_lsn;
CurrentResourceOwner = ResourceOwnerCreate(CurrentResourceOwner, "logical decoding");
/* invalidate non-timetravel entries */
InvalidateSystemCaches();
/* Decode until we run out of records */
while ((startptr != InvalidXLogRecPtr && startptr < end_of_wal) ||
(ctx->reader->EndRecPtr != InvalidXLogRecPtr && ctx->reader->EndRecPtr < end_of_wal))
{
XLogRecord *record;
char *errm = NULL;
record = XLogReadRecord(ctx->reader, startptr, &errm);
if (errm)
elog(ERROR, "%s", errm);
/*
* Now that we've set up the xlog reader state, subsequent calls
* pass InvalidXLogRecPtr to say "continue from last record"
*/
startptr = InvalidXLogRecPtr;
/*
* The {begin_txn,change,commit_txn}_wrapper callbacks above will
* store the description into our tuplestore.
*/
if (record != NULL)
LogicalDecodingProcessRecord(ctx, ctx->reader);
/* check limits */
if (upto_lsn != InvalidXLogRecPtr &&
upto_lsn <= ctx->reader->EndRecPtr)
break;
if (upto_nchanges != 0 &&
upto_nchanges <= p->returned_rows)
break;
CHECK_FOR_INTERRUPTS();
}
tuplestore_donestoring(tupstore);
CurrentResourceOwner = old_resowner;
/*
* Next time, start where we left off. (Hunting things, the family
* business..)
*/
if (ctx->reader->EndRecPtr != InvalidXLogRecPtr && confirm)
{
LogicalConfirmReceivedLocation(ctx->reader->EndRecPtr);
/*
* If only the confirmed_flush_lsn has changed the slot won't get
* marked as dirty by the above. Callers on the walsender
* interface are expected to keep track of their own progress and
* don't need it written out. But SQL-interface users cannot
* specify their own start positions and it's harder for them to
* keep track of their progress, so we should make more of an
* effort to save it for them.
*
* Dirty the slot so it's written out at the next checkpoint.
* We'll still lose its position on crash, as documented, but it's
* better than always losing the position even on clean restart.
*/
ReplicationSlotMarkDirty();
}
/* free context, call shutdown callback */
FreeDecodingContext(ctx);
ReplicationSlotRelease();
InvalidateSystemCaches();
}
PG_CATCH();
{
/* clear all timetravel entries */
InvalidateSystemCaches();
PG_RE_THROW();
}
PG_END_TRY();
return (Datum) 0;
}
/*
* Helper function for the various SQL callable logical decoding functions.
*/
static Datum
pg_logical_slot_get_changes_guts(FunctionCallInfo fcinfo, bool confirm, bool binary)
{
Name name = PG_GETARG_NAME(0);
XLogRecPtr upto_lsn;
int32 upto_nchanges;
ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
MemoryContext per_query_ctx;
MemoryContext oldcontext;
XLogRecPtr end_of_wal;
XLogRecPtr startptr;
LogicalDecodingContext *ctx;
ResourceOwner old_resowner = CurrentResourceOwner;
ArrayType *arr;
Size ndim;
List *options = NIL;
DecodingOutputState *p;
if (PG_ARGISNULL(1))
upto_lsn = InvalidXLogRecPtr;
else
upto_lsn = PG_GETARG_LSN(1);
if (PG_ARGISNULL(2))
upto_nchanges = InvalidXLogRecPtr;
else
upto_nchanges = PG_GETARG_INT32(2);
/* check to see if caller supports us returning a tuplestore */
if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("set-valued function called in context that cannot accept a set")));
if (!(rsinfo->allowedModes & SFRM_Materialize))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("materialize mode required, but it is not allowed in this context")));
/* state to write output to */
p = palloc0(sizeof(DecodingOutputState));
p->binary_output = binary;
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &p->tupdesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
check_permissions();
CheckLogicalDecodingRequirements();
arr = PG_GETARG_ARRAYTYPE_P(3);
ndim = ARR_NDIM(arr);
per_query_ctx = rsinfo->econtext->ecxt_per_query_memory;
oldcontext = MemoryContextSwitchTo(per_query_ctx);
if (ndim > 1)
{
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("array must be one-dimensional")));
}
else if (array_contains_nulls(arr))
{
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("array must not contain nulls")));
}
else if (ndim == 1)
{
int nelems;
Datum *datum_opts;
int i;
Assert(ARR_ELEMTYPE(arr) == TEXTOID);
deconstruct_array(arr, TEXTOID, -1, false, 'i',
&datum_opts, NULL, &nelems);
if (nelems % 2 != 0)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("array must have even number of elements")));
for (i = 0; i < nelems; i += 2)
{
char *name = TextDatumGetCString(datum_opts[i]);
char *opt = TextDatumGetCString(datum_opts[i + 1]);
options = lappend(options, makeDefElem(name, (Node *) makeString(opt)));
}
}
p->tupstore = tuplestore_begin_heap(true, false, work_mem);
rsinfo->returnMode = SFRM_Materialize;
rsinfo->setResult = p->tupstore;
rsinfo->setDesc = p->tupdesc;
/* compute the current end-of-wal */
if (!RecoveryInProgress())
end_of_wal = GetFlushRecPtr();
else
end_of_wal = GetXLogReplayRecPtr(NULL);
CheckLogicalDecodingRequirements();
ReplicationSlotAcquire(NameStr(*name));
PG_TRY();
{
ctx = CreateDecodingContext(InvalidXLogRecPtr,
options,
logical_read_local_xlog_page,
LogicalOutputPrepareWrite,
LogicalOutputWrite);
MemoryContextSwitchTo(oldcontext);
/*
* Check whether the output pluggin writes textual output if that's
* what we need.
*/
if (!binary &&
ctx->options.output_type != OUTPUT_PLUGIN_TEXTUAL_OUTPUT)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("output plugin cannot produce binary output")));
ctx->output_writer_private = p;
startptr = MyReplicationSlot->data.restart_lsn;
CurrentResourceOwner = ResourceOwnerCreate(CurrentResourceOwner, "logical decoding");
/* invalidate non-timetravel entries */
InvalidateSystemCaches();
while ((startptr != InvalidXLogRecPtr && startptr < end_of_wal) ||
(ctx->reader->EndRecPtr && ctx->reader->EndRecPtr < end_of_wal))
{
XLogRecord *record;
char *errm = NULL;
record = XLogReadRecord(ctx->reader, startptr, &errm);
if (errm)
elog(ERROR, "%s", errm);
startptr = InvalidXLogRecPtr;
/*
* The {begin_txn,change,commit_txn}_wrapper callbacks above will
* store the description into our tuplestore.
*/
if (record != NULL)
LogicalDecodingProcessRecord(ctx, record);
/* check limits */
if (upto_lsn != InvalidXLogRecPtr &&
upto_lsn <= ctx->reader->EndRecPtr)
break;
if (upto_nchanges != 0 &&
upto_nchanges <= p->returned_rows)
break;
}
}
PG_CATCH();
{
/* clear all timetravel entries */
InvalidateSystemCaches();
PG_RE_THROW();
}
PG_END_TRY();
tuplestore_donestoring(tupstore);
CurrentResourceOwner = old_resowner;
/*
* Next time, start where we left off. (Hunting things, the family
* business..)
*/
if (ctx->reader->EndRecPtr != InvalidXLogRecPtr && confirm)
LogicalConfirmReceivedLocation(ctx->reader->EndRecPtr);
/* free context, call shutdown callback */
FreeDecodingContext(ctx);
ReplicationSlotRelease();
InvalidateSystemCaches();
return (Datum) 0;
}
