/*
* open xlog file
*/
static void
openXlogEnd(XLogRecPtr *endLocation)
{
char path[MAXPGPATH];
uint32 logid;
uint32 seg;
char *xlogDir = NULL;
XLByteToSeg(*endLocation, logid, seg);
XLogFileName(xlogfilename, ThisTimeLineID, logid, seg);
elog((Debug_print_qd_mirroring ? LOG : DEBUG5), "QDSYNC: opening logid %d seg %d file %s",
logid, seg, xlogfilename);
xlogDir = makeRelativeToTxnFilespace(XLOGDIR);
/* the first time or file is changed */
if (snprintf(path, MAXPGPATH, "%s/%s", xlogDir, xlogfilename) >= MAXPGPATH)
{
ereport(ERROR,
(errcode_for_file_access(),
errmsg("QDSYNC: could not create xlog file \"%s/%s\"",
xlogDir, xlogfilename)));
}
if (xlogfilefd >= 0)
{
close(xlogfilefd);
xlogfilefd = -1;
xlogfileoffset = -1;
}
xlogfilefd = open(path, O_RDWR, 0);
if (xlogfilefd < 0)
{
elog((Debug_print_qd_mirroring ? LOG : DEBUG5), "QDSYNC: creating xlog file %s", xlogfilename);
xlogfilefd = open(path, O_RDWR | O_CREAT | O_EXCL, S_IRUSR | S_IWUSR);
if (xlogfilefd < 0)
{
ereport(ERROR,
(errcode_for_file_access(),
errmsg("QDSYNC: could not create xlog file \"%s\"",
path)));
}
}
xlogid = logid;
xseg = seg;
xlogfileoffset = endLocation->xrecoff % XLogSegSize;
elog((Debug_print_qd_mirroring ? LOG : DEBUG5),"QDSYNC: opened '%s' offset 0x%X",
xlogfilename, xlogfileoffset);
pfree(xlogDir);
}
Datum
pg_control_checkpoint(PG_FUNCTION_ARGS)
{
Datum values[19];
bool nulls[19];
TupleDesc tupdesc;
HeapTuple htup;
ControlFileData *ControlFile;
XLogSegNo segno;
char xlogfilename[MAXFNAMELEN];
/*
* Construct a tuple descriptor for the result row. This must match this
* function's pg_proc entry!
*/
tupdesc = CreateTemplateTupleDesc(19, false);
TupleDescInitEntry(tupdesc, (AttrNumber) 1, "checkpoint_location",
LSNOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 2, "prior_location",
LSNOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 3, "redo_location",
LSNOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 4, "redo_wal_file",
TEXTOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 5, "timeline_id",
INT4OID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 6, "prev_timeline_id",
INT4OID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 7, "full_page_writes",
BOOLOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 8, "next_xid",
TEXTOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 9, "next_oid",
OIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 10, "next_multixact_id",
XIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 11, "next_multi_offset",
XIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 12, "oldest_xid",
XIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 13, "oldest_xid_dbid",
OIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 14, "oldest_active_xid",
XIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 15, "oldest_multi_xid",
XIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 16, "oldest_multi_dbid",
OIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 17, "oldest_commit_ts_xid",
XIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 18, "newest_commit_ts_xid",
XIDOID, -1, 0);
TupleDescInitEntry(tupdesc, (AttrNumber) 19, "checkpoint_time",
TIMESTAMPTZOID, -1, 0);
tupdesc = BlessTupleDesc(tupdesc);
/* Read the control file. */
ControlFile = get_controlfile(DataDir, NULL);
/*
* Calculate name of the WAL file containing the latest checkpoint's REDO
* start point.
*/
XLByteToSeg(ControlFile->checkPointCopy.redo, segno);
XLogFileName(xlogfilename, ControlFile->checkPointCopy.ThisTimeLineID, segno);
/* Populate the values and null arrays */
values[0] = LSNGetDatum(ControlFile->checkPoint);
nulls[0] = false;
values[1] = LSNGetDatum(ControlFile->prevCheckPoint);
nulls[1] = false;
values[2] = LSNGetDatum(ControlFile->checkPointCopy.redo);
nulls[2] = false;
values[3] = CStringGetTextDatum(xlogfilename);
nulls[3] = false;
values[4] = Int32GetDatum(ControlFile->checkPointCopy.ThisTimeLineID);
nulls[4] = false;
values[5] = Int32GetDatum(ControlFile->checkPointCopy.PrevTimeLineID);
nulls[5] = false;
values[6] = BoolGetDatum(ControlFile->checkPointCopy.fullPageWrites);
nulls[6] = false;
values[7] = CStringGetTextDatum(psprintf("%u:%u",
ControlFile->checkPointCopy.nextXidEpoch,
ControlFile->checkPointCopy.nextXid));
nulls[7] = false;
values[8] = ObjectIdGetDatum(ControlFile->checkPointCopy.nextOid);
nulls[8] = false;
values[9] = TransactionIdGetDatum(ControlFile->checkPointCopy.nextMulti);
nulls[9] = false;
values[10] = TransactionIdGetDatum(ControlFile->checkPointCopy.nextMultiOffset);
nulls[10] = false;
values[11] = TransactionIdGetDatum(ControlFile->checkPointCopy.oldestXid);
nulls[11] = false;
values[12] = ObjectIdGetDatum(ControlFile->checkPointCopy.oldestXidDB);
nulls[12] = false;
values[13] = TransactionIdGetDatum(ControlFile->checkPointCopy.oldestActiveXid);
nulls[13] = false;
values[14] = TransactionIdGetDatum(ControlFile->checkPointCopy.oldestMulti);
nulls[14] = false;
values[15] = ObjectIdGetDatum(ControlFile->checkPointCopy.oldestMultiDB);
nulls[15] = false;
values[16] = TransactionIdGetDatum(ControlFile->checkPointCopy.oldestCommitTsXid);
nulls[16] = false;
values[17] = TransactionIdGetDatum(ControlFile->checkPointCopy.newestCommitTsXid);
nulls[17] = false;
values[18] = TimestampTzGetDatum(
time_t_to_timestamptz(ControlFile->checkPointCopy.time));
nulls[18] = false;
htup = heap_form_tuple(tupdesc, values, nulls);
PG_RETURN_DATUM(HeapTupleGetDatum(htup));
}
/*
* Read 'count' bytes from WAL into 'buf', starting at location 'startptr'
*
* XXX probably this should be improved to suck data directly from the
* WAL buffers when possible.
*
* Will open, and keep open, one WAL segment stored in the global file
* descriptor sendFile. This means if XLogRead is used once, there will
* always be one descriptor left open until the process ends, but never
* more than one.
*/
void
XLogRead(char *buf, XLogRecPtr startptr, Size count)
{
char *p;
XLogRecPtr recptr;
Size nbytes;
uint32 lastRemovedLog;
uint32 lastRemovedSeg;
uint32 log;
uint32 seg;
retry:
p = buf;
recptr = startptr;
nbytes = count;
while (nbytes > 0)
{
uint32 startoff;
int segbytes;
int readbytes;
startoff = recptr.xrecoff % XLogSegSize;
if (sendFile < 0 || !XLByteInSeg(recptr, sendId, sendSeg))
{
char path[MAXPGPATH];
/* Switch to another logfile segment */
if (sendFile >= 0)
close(sendFile);
XLByteToSeg(recptr, sendId, sendSeg);
XLogFilePath(path, ThisTimeLineID, sendId, sendSeg);
sendFile = BasicOpenFile(path, O_RDONLY | PG_BINARY, 0);
if (sendFile < 0)
{
/*
* If the file is not found, assume it's because the standby
* asked for a too old WAL segment that has already been
* removed or recycled.
*/
if (errno == ENOENT)
{
char filename[MAXFNAMELEN];
XLogFileName(filename, ThisTimeLineID, sendId, sendSeg);
ereport(ERROR,
(errcode_for_file_access(),
errmsg("requested WAL segment %s has already been removed",
filename)));
}
else
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open file \"%s\" (log file %u, segment %u): %m",
path, sendId, sendSeg)));
}
sendOff = 0;
}
/* Need to seek in the file? */
if (sendOff != startoff)
{
if (lseek(sendFile, (off_t) startoff, SEEK_SET) < 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not seek in log file %u, segment %u to offset %u: %m",
sendId, sendSeg, startoff)));
sendOff = startoff;
}
/* How many bytes are within this segment? */
if (nbytes > (XLogSegSize - startoff))
segbytes = XLogSegSize - startoff;
else
segbytes = nbytes;
readbytes = read(sendFile, p, segbytes);
if (readbytes <= 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not read from log file %u, segment %u, offset %u, "
"length %lu: %m",
sendId, sendSeg, sendOff, (unsigned long) segbytes)));
/* Update state for read */
XLByteAdvance(recptr, readbytes);
sendOff += readbytes;
nbytes -= readbytes;
p += readbytes;
}
/*
* After reading into the buffer, check that what we read was valid. We do
* this after reading, because even though the segment was present when we
* opened it, it might get recycled or removed while we read it. The
* read() succeeds in that case, but the data we tried to read might
* already have been overwritten with new WAL records.
*/
XLogGetLastRemoved(&lastRemovedLog, &lastRemovedSeg);
XLByteToSeg(startptr, log, seg);
if (log < lastRemovedLog ||
(log == lastRemovedLog && seg <= lastRemovedSeg))
{
char filename[MAXFNAMELEN];
XLogFileName(filename, ThisTimeLineID, log, seg);
ereport(ERROR,
(errcode_for_file_access(),
errmsg("requested WAL segment %s has already been removed",
filename)));
}
/*
* During recovery, the currently-open WAL file might be replaced with
* the file of the same name retrieved from archive. So we always need
* to check what we read was valid after reading into the buffer. If it's
* invalid, we try to open and read the file again.
*/
if (am_cascading_walsender)
{
/* use volatile pointer to prevent code rearrangement */
volatile WalSnd *walsnd = MyWalSnd;
bool reload;
SpinLockAcquire(&walsnd->mutex);
reload = walsnd->needreload;
walsnd->needreload = false;
SpinLockRelease(&walsnd->mutex);
if (reload && sendFile >= 0)
{
close(sendFile);
sendFile = -1;
goto retry;
}
}
}
/*
* Reserve WAL for the currently active slot.
*
* Compute and set restart_lsn in a manner that's appropriate for the type of
* the slot and concurrency safe.
*/
void
ReplicationSlotReserveWal(void)
{
ReplicationSlot *slot = MyReplicationSlot;
Assert(slot != NULL);
Assert(slot->data.restart_lsn == InvalidXLogRecPtr);
/*
* The replication slot mechanism is used to prevent removal of required
* WAL. As there is no interlock between this routine and checkpoints, WAL
* segments could concurrently be removed when a now stale return value of
* ReplicationSlotsComputeRequiredLSN() is used. In the unlikely case that
* this happens we'll just retry.
*/
while (true)
{
XLogSegNo segno;
/*
* For logical slots log a standby snapshot and start logical decoding
* at exactly that position. That allows the slot to start up more
* quickly.
*
* That's not needed (or indeed helpful) for physical slots as they'll
* start replay at the last logged checkpoint anyway. Instead return
* the location of the last redo LSN. While that slightly increases
* the chance that we have to retry, it's where a base backup has to
* start replay at.
*/
if (!RecoveryInProgress() && SlotIsLogical(slot))
{
XLogRecPtr flushptr;
/* start at current insert position */
slot->data.restart_lsn = GetXLogInsertRecPtr();
/* make sure we have enough information to start */
flushptr = LogStandbySnapshot();
/* and make sure it's fsynced to disk */
XLogFlush(flushptr);
}
else
{
slot->data.restart_lsn = GetRedoRecPtr();
}
/* prevent WAL removal as fast as possible */
ReplicationSlotsComputeRequiredLSN();
/*
* If all required WAL is still there, great, otherwise retry. The
* slot should prevent further removal of WAL, unless there's a
* concurrent ReplicationSlotsComputeRequiredLSN() after we've written
* the new restart_lsn above, so normally we should never need to loop
* more than twice.
*/
XLByteToSeg(slot->data.restart_lsn, segno, wal_segment_size);
if (XLogGetLastRemovedSegno() < segno)
break;
}
}
/*
* Read 'nbytes' bytes from WAL into 'buf', starting at location 'recptr'
*
* XXX probably this should be improved to suck data directly from the
* WAL buffers when possible.
*
* Will open, and keep open, one WAL segment stored in the global file
* descriptor sendFile. This means if XLogRead is used once, there will
* always be one descriptor left open until the process ends, but never
* more than one.
*/
void
XLogRead(char *buf, XLogRecPtr recptr, Size nbytes)
{
XLogRecPtr startRecPtr = recptr;
char path[MAXPGPATH];
uint32 lastRemovedLog;
uint32 lastRemovedSeg;
uint32 log;
uint32 seg;
while (nbytes > 0)
{
uint32 startoff;
int segbytes;
int readbytes;
startoff = recptr.xrecoff % XLogSegSize;
if (sendFile < 0 || !XLByteInSeg(recptr, sendId, sendSeg))
{
/* Switch to another logfile segment */
if (sendFile >= 0)
close(sendFile);
XLByteToSeg(recptr, sendId, sendSeg);
XLogFilePath(path, ThisTimeLineID, sendId, sendSeg);
sendFile = BasicOpenFile(path, O_RDONLY | PG_BINARY, 0);
if (sendFile < 0)
{
/*
* If the file is not found, assume it's because the standby
* asked for a too old WAL segment that has already been
* removed or recycled.
*/
if (errno == ENOENT)
{
char filename[MAXFNAMELEN];
XLogFileName(filename, ThisTimeLineID, sendId, sendSeg);
ereport(ERROR,
(errcode_for_file_access(),
errmsg("requested WAL segment %s has already been removed",
filename)));
}
else
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open file \"%s\" (log file %u, segment %u): %m",
path, sendId, sendSeg)));
}
sendOff = 0;
}
/* Need to seek in the file? */
if (sendOff != startoff)
{
if (lseek(sendFile, (off_t) startoff, SEEK_SET) < 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not seek in log file %u, segment %u to offset %u: %m",
sendId, sendSeg, startoff)));
sendOff = startoff;
}
/* How many bytes are within this segment? */
if (nbytes > (XLogSegSize - startoff))
segbytes = XLogSegSize - startoff;
else
segbytes = nbytes;
readbytes = read(sendFile, buf, segbytes);
if (readbytes <= 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not read from log file %u, segment %u, offset %u, "
"length %lu: %m",
sendId, sendSeg, sendOff, (unsigned long) segbytes)));
/* Update state for read */
XLByteAdvance(recptr, readbytes);
sendOff += readbytes;
nbytes -= readbytes;
buf += readbytes;
}
/*
* After reading into the buffer, check that what we read was valid. We do
* this after reading, because even though the segment was present when we
* opened it, it might get recycled or removed while we read it. The
* read() succeeds in that case, but the data we tried to read might
* already have been overwritten with new WAL records.
*/
XLogGetLastRemoved(&lastRemovedLog, &lastRemovedSeg);
XLByteToSeg(startRecPtr, log, seg);
if (log < lastRemovedLog ||
(log == lastRemovedLog && seg <= lastRemovedSeg))
{
char filename[MAXFNAMELEN];
XLogFileName(filename, ThisTimeLineID, log, seg);
ereport(ERROR,
(errcode_for_file_access(),
errmsg("requested WAL segment %s has already been removed",
filename)));
}
}
static void
WalSendServerDoRequest(WalSendRequest *walSendRequest)
{
bool successful;
struct timeval standbyTimeout;
WalSendServerGetStandbyTimeout(&standbyTimeout);
switch (walSendRequest->command)
{
case PositionToEnd:
elog((Debug_print_qd_mirroring ? LOG : DEBUG5), "PositionToEnd");
successful = write_position_to_end(&originalEndLocation,
NULL, &walsend_shutdown_requested);
if (successful)
elog(LOG,"Standby master returned transaction log end location %s",
XLogLocationToString(&originalEndLocation));
else
{
disableQDMirroring_ConnectionError(
"Unable to connect to standby master and determine transaction log end location",
GetStandbyErrorString());
disconnectMirrorQD_SendClose();
}
break;
case Catchup:
elog((Debug_print_qd_mirroring ? LOG : DEBUG5), "Catchup");
if (isQDMirroringCatchingUp())
{
bool tooFarBehind = false;
elog(LOG,"Current master transaction log is flushed through location %s",
XLogLocationToString(&walSendRequest->flushedLocation));
if (XLByteLT(originalEndLocation, walSendRequest->flushedLocation))
{
/*
* Standby master is behind the primary. Send catchup WAL.
*/
/*
* Use a TRY block to catch errors from our attempt to read
* the primary's WAL. Errors from sending to the standby
* come up as a boolean return (successful).
*/
PG_TRY();
{
successful = XLogCatchupQDMirror(
&originalEndLocation,
&walSendRequest->flushedLocation,
&standbyTimeout,
&walsend_shutdown_requested);
}
PG_CATCH();
{
/*
* Report the error related to reading the primary's WAL
* to the server log
*/
/*
* But first demote the error to something much less
* scary.
*/
if (!elog_demote(WARNING))
{
elog(LOG,"unable to demote error");
PG_RE_THROW();
}
EmitErrorReport();
FlushErrorState();
successful = false;
tooFarBehind = true;
}
PG_END_TRY();
if (successful)
{
elog((Debug_print_qd_mirroring ? LOG : DEBUG5),
"catchup send from standby end %s through primary flushed location %s",
XLogLocationToString(&originalEndLocation),
XLogLocationToString2(&walSendRequest->flushedLocation));
}
}
else if (XLByteEQ(originalEndLocation, walSendRequest->flushedLocation))
{
elog((Debug_print_qd_mirroring ? LOG : DEBUG5),"Mirror was already caught up");
successful = true;
}
else
{
elog(WARNING,"Standby master transaction log location %s is beyond the current master end location %s",
XLogLocationToString(&originalEndLocation),
XLogLocationToString2(&walSendRequest->flushedLocation));
successful = false;
}
if (successful)
{
char detail[200];
int count;
count = snprintf(
detail, sizeof(detail),
"Transaction log copied from locations %s through %s to the standby master",
XLogLocationToString(&originalEndLocation),
XLogLocationToString2(&walSendRequest->flushedLocation));
if (count >= sizeof(detail))
{
ereport(ERROR,
(errcode(ERRCODE_INTERNAL_ERROR),
errmsg("format command string failure")));
}
enableQDMirroring("Master mirroring is now synchronized", detail);
currentEndLocation = walSendRequest->flushedLocation;
periodicLen = 0;
periodicLocation = currentEndLocation;
}
else
{
if (tooFarBehind)
{
disableQDMirroring_TooFarBehind(
"The current master was unable to synchronize the standby master "
"because the transaction logs on the current master were recycled. "
"A gpinitstandby (at an appropriate time) will be necessary to copy "
"over the whole master database to the standby master so it may be synchronized");
}
else
{
disableQDMirroring_ConnectionError(
"Connection to the standby master was lost during transaction log catchup",
GetStandbyErrorString());
}
disconnectMirrorQD_SendClose();
}
}
else if (isQDMirroringDisabled())
{
elog((Debug_print_qd_mirroring ? LOG : DEBUG5), "Master Mirror Send: Master mirroring not catching-up (state is disabled)");
}
else
{
elog(ERROR,"unexpected master mirroring state %s",
QDMirroringStateString());
}
break;
case WriteWalPages:
if (Debug_print_qd_mirroring)
elog(LOG, "WriteWalPages");
if (isQDMirroringEnabled())
{
char *from;
Size nbytes;
bool more= false;
/*
* For now, save copy of data until flush. This could be
* optimized.
*/
if (saveBuffer == NULL)
{
uint32 totalBufferLen = XLOGbuffers * XLOG_BLCKSZ;
saveBuffer = malloc(totalBufferLen);
if (saveBuffer == NULL)
elog(ERROR,"Could not allocate buffer for xlog data (%d bytes)",
totalBufferLen);
saveBufferLen = 0;
}
XLogGetBuffer(walSendRequest->startidx, walSendRequest->npages,
&from, &nbytes);
if (saveBufferLen == 0)
{
more = false;
writeLogId = walSendRequest->logId;
writeLogSeg = walSendRequest->logSeg;
writeLogOff = walSendRequest->logOff;
memcpy(saveBuffer, from, nbytes);
saveBufferLen = nbytes;
}
else
{
more = true;
memcpy(&saveBuffer[saveBufferLen], from, nbytes);
saveBufferLen += nbytes;
}
if (Debug_print_qd_mirroring)
elog(LOG,
"Master Mirror Send: WriteWalPages (%s) startidx %d, npages %d, timeLineID %d, logId %u, logSeg %u, logOff 0x%X, nbytes 0x%X",
(more ? "more" : "new"),
walSendRequest->startidx,
walSendRequest->npages,
walSendRequest->timeLineID,
walSendRequest->logId,
walSendRequest->logSeg,
walSendRequest->logOff,
(int)nbytes);
}
case FlushWalPages:
if (Debug_print_qd_mirroring)
elog(LOG, "FlushWalPages");
if (isQDMirroringEnabled())
{
char cmd[MAXFNAMELEN + 50];
if (saveBufferLen == 0)
successful = true;
else
{
if (snprintf(cmd, sizeof(cmd),"xlog %d %d %d %d",
writeLogId, writeLogSeg, writeLogOff,
(int)saveBufferLen) >= sizeof(cmd))
elog(ERROR,"could not create cmd for qd mirror logid %d seg %d",
writeLogId, writeLogSeg);
successful = write_qd_sync(cmd, saveBuffer, saveBufferLen,
&standbyTimeout,
&walsend_shutdown_requested);
if (successful)
{
XLogRecPtr oldEndLocation;
oldEndLocation = currentEndLocation;
currentEndLocation.xlogid = writeLogId;
currentEndLocation.xrecoff = writeLogSeg * XLogSegSize + writeLogOff;
if (currentEndLocation.xrecoff >= XLogFileSize)
{
(currentEndLocation.xlogid)++;
currentEndLocation.xrecoff = 0;
}
if (XLByteLT(oldEndLocation,currentEndLocation))
{
periodicLen += saveBufferLen;
if (periodicLen > periodicReportLen)
{
elog(LOG,
"Master mirroring periodic report: %d bytes successfully send to standby master for locations %s through %s",
periodicLen,
XLogLocationToString(&periodicLocation),
XLogLocationToString2(¤tEndLocation));
periodicLen = 0;
periodicLocation = currentEndLocation;
}
}
else
{
if (Debug_print_qd_mirroring)
elog(LOG,
"Send to Master mirror successful. New end location %s (old %s)",
XLogLocationToString(¤tEndLocation),
XLogLocationToString2(&oldEndLocation));
}
}
else
{
disableQDMirroring_ConnectionError(
"Connection to the standby master was lost attempting to send new transaction log",
GetStandbyErrorString());
disconnectMirrorQD_SendClose();
}
/*
* Reset so WriteWalPages can fill the buffer again.
*/
saveBufferLen = 0;
writeLogId = 0;
writeLogSeg = 0;
writeLogOff = 0;
}
if (successful && walSendRequest->haveNewCheckpointLocation)
{
uint32 logid;
uint32 seg;
uint32 offset;
elog((Debug_print_qd_mirroring ? LOG : DEBUG5),"New previous checkpoint location %s",
XLogLocationToString(&walSendRequest->newCheckpointLocation));
XLByteToSeg(walSendRequest->newCheckpointLocation, logid, seg);
offset = walSendRequest->newCheckpointLocation.xrecoff % XLogSegSize;
if (snprintf(cmd, sizeof(cmd),"new_checkpoint_location %d %d %d",
logid, seg, offset) >= sizeof(cmd))
elog(ERROR,"could not create cmd for qd mirror logid %d seg %d offset %d",
logid, seg, offset);
successful = write_qd_sync(cmd, NULL, 0,
NULL, &walsend_shutdown_requested);
if (successful)
{
elog((Debug_print_qd_mirroring ? LOG : DEBUG5),"Send of new checkpoint location to master mirror successful");
}
else
{
disableQDMirroring_ConnectionError(
"Connection to the standby master was lost attempting to send new checkpoint location",
GetStandbyErrorString());
disconnectMirrorQD_SendClose();
}
}
}
else if (isQDMirroringDisabled())
{
elog((Debug_print_qd_mirroring ? LOG : DEBUG5), "Master Mirror Send: Master mirroring not enabled");
}
else
{
elog(ERROR,"unexpected master mirroring state %s",
QDMirroringStateString());
}
break;
case CloseForShutdown:
if (Debug_print_qd_mirroring)
elog(LOG, "CloseForShutdown");
/*
* Do the work we would normally do when signaled to stop.
*/
WalSendServer_ServiceShutdown();
break;
default:
elog(ERROR, "Unknown WalSendRequestCommand %d", walSendRequest->command);
}
}
/*
* Attempt to retrieve the specified file from off-line archival storage.
* If successful, fill "path" with its complete path (note that this will be
* a temp file name that doesn't follow the normal naming convention), and
* return TRUE.
*
* If not successful, fill "path" with the name of the normal on-line file
* (which may or may not actually exist, but we'll try to use it), and return
* FALSE.
*
* For fixed-size files, the caller may pass the expected size as an
* additional crosscheck on successful recovery. If the file size is not
* known, set expectedSize = 0.
*
* When 'cleanupEnabled' is false, refrain from deleting any old WAL segments
* in the archive. This is used when fetching the initial checkpoint record,
* when we are not yet sure how far back we need the WAL.
*/
bool
RestoreArchivedFile(char *path, const char *xlogfname,
const char *recovername, off_t expectedSize,
bool cleanupEnabled)
{
char xlogpath[MAXPGPATH];
char xlogRestoreCmd[MAXPGPATH];
char lastRestartPointFname[MAXPGPATH];
char *dp;
char *endp;
const char *sp;
int rc;
bool signaled;
struct stat stat_buf;
XLogSegNo restartSegNo;
XLogRecPtr restartRedoPtr;
TimeLineID restartTli;
/* In standby mode, restore_command might not be supplied */
if (recoveryRestoreCommand == NULL)
goto not_available;
/*
* When doing archive recovery, we always prefer an archived log file even
* if a file of the same name exists in XLOGDIR. The reason is that the
* file in XLOGDIR could be an old, un-filled or partly-filled version
* that was copied and restored as part of backing up $PGDATA.
*
* We could try to optimize this slightly by checking the local copy
* lastchange timestamp against the archived copy, but we have no API to
* do this, nor can we guarantee that the lastchange timestamp was
* preserved correctly when we copied to archive. Our aim is robustness,
* so we elect not to do this.
*
* If we cannot obtain the log file from the archive, however, we will try
* to use the XLOGDIR file if it exists. This is so that we can make use
* of log segments that weren't yet transferred to the archive.
*
* Notice that we don't actually overwrite any files when we copy back
* from archive because the restore_command may inadvertently restore
* inappropriate xlogs, or they may be corrupt, so we may wish to fallback
* to the segments remaining in current XLOGDIR later. The
* copy-from-archive filename is always the same, ensuring that we don't
* run out of disk space on long recoveries.
*/
snprintf(xlogpath, MAXPGPATH, XLOGDIR "/%s", recovername);
/*
* Make sure there is no existing file named recovername.
*/
if (stat(xlogpath, &stat_buf) != 0)
{
if (errno != ENOENT)
ereport(FATAL,
(errcode_for_file_access(),
errmsg("could not stat file \"%s\": %m",
xlogpath)));
}
else
{
if (unlink(xlogpath) != 0)
ereport(FATAL,
(errcode_for_file_access(),
errmsg("could not remove file \"%s\": %m",
xlogpath)));
}
/*
* Calculate the archive file cutoff point for use during log shipping
* replication. All files earlier than this point can be deleted from the
* archive, though there is no requirement to do so.
*
* If cleanup is not enabled, initialise this with the filename of
* InvalidXLogRecPtr, which will prevent the deletion of any WAL files
* from the archive because of the alphabetic sorting property of WAL
* filenames.
*
* Once we have successfully located the redo pointer of the checkpoint
* from which we start recovery we never request a file prior to the redo
* pointer of the last restartpoint. When redo begins we know that we have
* successfully located it, so there is no need for additional status
* flags to signify the point when we can begin deleting WAL files from
* the archive.
*/
if (cleanupEnabled)
{
GetOldestRestartPoint(&restartRedoPtr, &restartTli);
XLByteToSeg(restartRedoPtr, restartSegNo);
XLogFileName(lastRestartPointFname, restartTli, restartSegNo);
/* we shouldn't need anything earlier than last restart point */
Assert(strcmp(lastRestartPointFname, xlogfname) <= 0);
}
else
XLogFileName(lastRestartPointFname, 0, 0L);
/*
* construct the command to be executed
*/
dp = xlogRestoreCmd;
endp = xlogRestoreCmd + MAXPGPATH - 1;
*endp = '\0';
for (sp = recoveryRestoreCommand; *sp; sp++)
{
if (*sp == '%')
{
switch (sp[1])
{
case 'p':
/* %p: relative path of target file */
sp++;
StrNCpy(dp, xlogpath, endp - dp);
make_native_path(dp);
dp += strlen(dp);
break;
case 'f':
/* %f: filename of desired file */
sp++;
StrNCpy(dp, xlogfname, endp - dp);
dp += strlen(dp);
break;
case 'r':
/* %r: filename of last restartpoint */
sp++;
StrNCpy(dp, lastRestartPointFname, endp - dp);
dp += strlen(dp);
break;
case '%':
/* convert %% to a single % */
sp++;
if (dp < endp)
*dp++ = *sp;
break;
default:
/* otherwise treat the % as not special */
if (dp < endp)
*dp++ = *sp;
break;
}
}
else
{
if (dp < endp)
*dp++ = *sp;
}
}
*dp = '\0';
ereport(DEBUG3,
(errmsg_internal("executing restore command \"%s\"",
xlogRestoreCmd)));
/*
* Check signals before restore command and reset afterwards.
*/
PreRestoreCommand();
/*
* Copy xlog from archival storage to XLOGDIR
*/
rc = system(xlogRestoreCmd);
PostRestoreCommand();
if (rc == 0)
{
/*
* command apparently succeeded, but let's make sure the file is
* really there now and has the correct size.
*/
if (stat(xlogpath, &stat_buf) == 0)
{
if (expectedSize > 0 && stat_buf.st_size != expectedSize)
{
int elevel;
/*
* If we find a partial file in standby mode, we assume it's
* because it's just being copied to the archive, and keep
* trying.
*
* Otherwise treat a wrong-sized file as FATAL to ensure the
* DBA would notice it, but is that too strong? We could try
* to plow ahead with a local copy of the file ... but the
* problem is that there probably isn't one, and we'd
* incorrectly conclude we've reached the end of WAL and we're
* done recovering ...
*/
if (StandbyMode && stat_buf.st_size < expectedSize)
elevel = DEBUG1;
else
elevel = FATAL;
ereport(elevel,
(errmsg("archive file \"%s\" has wrong size: %lu instead of %lu",
xlogfname,
(unsigned long) stat_buf.st_size,
(unsigned long) expectedSize)));
return false;
}
else
{
ereport(LOG,
(errmsg("restored log file \"%s\" from archive",
xlogfname)));
strcpy(path, xlogpath);
return true;
}
}
else
{
/* stat failed */
if (errno != ENOENT)
ereport(FATAL,
(errcode_for_file_access(),
errmsg("could not stat file \"%s\": %m",
xlogpath)));
}
}
/*
* Remember, we rollforward UNTIL the restore fails so failure here is
* just part of the process... that makes it difficult to determine
* whether the restore failed because there isn't an archive to restore,
* or because the administrator has specified the restore program
* incorrectly. We have to assume the former.
*
* However, if the failure was due to any sort of signal, it's best to
* punt and abort recovery. (If we "return false" here, upper levels will
* assume that recovery is complete and start up the database!) It's
* essential to abort on child SIGINT and SIGQUIT, because per spec
* system() ignores SIGINT and SIGQUIT while waiting; if we see one of
* those it's a good bet we should have gotten it too.
*
* On SIGTERM, assume we have received a fast shutdown request, and exit
* cleanly. It's pure chance whether we receive the SIGTERM first, or the
* child process. If we receive it first, the signal handler will call
* proc_exit, otherwise we do it here. If we or the child process received
* SIGTERM for any other reason than a fast shutdown request, postmaster
* will perform an immediate shutdown when it sees us exiting
* unexpectedly.
*
* Per the Single Unix Spec, shells report exit status > 128 when a called
* command died on a signal. Also, 126 and 127 are used to report
* problems such as an unfindable command; treat those as fatal errors
* too.
*/
if (WIFSIGNALED(rc) && WTERMSIG(rc) == SIGTERM)
proc_exit(1);
signaled = WIFSIGNALED(rc) || WEXITSTATUS(rc) > 125;
ereport(signaled ? FATAL : DEBUG2,
(errmsg("could not restore file \"%s\" from archive: %s",
xlogfname, wait_result_to_str(rc))));
not_available:
/*
* if an archived file is not available, there might still be a version of
* this file in XLOGDIR, so return that as the filename to open.
*
* In many recovery scenarios we expect this to fail also, but if so that
* just means we've reached the end of WAL.
*/
snprintf(path, MAXPGPATH, XLOGDIR "/%s", xlogfname);
return false;
}
/*
* Attempt to execute an external shell command during recovery.
*
* 'command' is the shell command to be executed, 'commandName' is a
* human-readable name describing the command emitted in the logs. If
* 'failOnSignal' is true and the command is killed by a signal, a FATAL
* error is thrown. Otherwise a WARNING is emitted.
*
* This is currently used for recovery_end_command and archive_cleanup_command.
*/
void
ExecuteRecoveryCommand(char *command, char *commandName, bool failOnSignal)
{
char xlogRecoveryCmd[MAXPGPATH];
char lastRestartPointFname[MAXPGPATH];
char *dp;
char *endp;
const char *sp;
int rc;
bool signaled;
XLogSegNo restartSegNo;
XLogRecPtr restartRedoPtr;
TimeLineID restartTli;
Assert(command && commandName);
/*
* Calculate the archive file cutoff point for use during log shipping
* replication. All files earlier than this point can be deleted from the
* archive, though there is no requirement to do so.
*/
GetOldestRestartPoint(&restartRedoPtr, &restartTli);
XLByteToSeg(restartRedoPtr, restartSegNo);
XLogFileName(lastRestartPointFname, restartTli, restartSegNo);
/*
* construct the command to be executed
*/
dp = xlogRecoveryCmd;
endp = xlogRecoveryCmd + MAXPGPATH - 1;
*endp = '\0';
for (sp = command; *sp; sp++)
{
if (*sp == '%')
{
switch (sp[1])
{
case 'r':
/* %r: filename of last restartpoint */
sp++;
StrNCpy(dp, lastRestartPointFname, endp - dp);
dp += strlen(dp);
break;
case '%':
/* convert %% to a single % */
sp++;
if (dp < endp)
*dp++ = *sp;
break;
default:
/* otherwise treat the % as not special */
if (dp < endp)
*dp++ = *sp;
break;
}
}
else
{
if (dp < endp)
*dp++ = *sp;
}
}
*dp = '\0';
ereport(DEBUG3,
(errmsg_internal("executing %s \"%s\"", commandName, command)));
/*
* execute the constructed command
*/
rc = system(xlogRecoveryCmd);
if (rc != 0)
{
/*
* If the failure was due to any sort of signal, it's best to punt and
* abort recovery. See also detailed comments on signals in
* RestoreArchivedFile().
*/
signaled = WIFSIGNALED(rc) || WEXITSTATUS(rc) > 125;
ereport((signaled && failOnSignal) ? FATAL : WARNING,
/*------
translator: First %s represents a recovery.conf parameter name like
"recovery_end_command", the 2nd is the value of that parameter, the
third an already translated error message. */
(errmsg("%s \"%s\": %s", commandName,
command, wait_result_to_str(rc))));
}
}
/*
* Write XLOG data to disk.
*/
static void
XLogWalRcvWrite(char *buf, Size nbytes, XLogRecPtr recptr)
{
int startoff;
int byteswritten;
while (nbytes > 0)
{
int segbytes;
if (recvFile < 0 || !XLByteInSeg(recptr, recvId, recvSeg))
{
bool use_existent;
/*
* fsync() and close current file before we switch to next one. We
* would otherwise have to reopen this file to fsync it later
*/
if (recvFile >= 0)
{
XLogWalRcvFlush();
/*
* XLOG segment files will be re-read by recovery in startup
* process soon, so we don't advise the OS to release cache
* pages associated with the file like XLogFileClose() does.
*/
if (close(recvFile) != 0)
ereport(PANIC,
(errcode_for_file_access(),
errmsg("could not close log file %u, segment %u: %m",
recvId, recvSeg)));
}
recvFile = -1;
/* Create/use new log file */
XLByteToSeg(recptr, recvId, recvSeg);
use_existent = true;
recvFile = XLogFileInit(recvId, recvSeg, &use_existent, true);
recvOff = 0;
}
/* Calculate the start offset of the received logs */
startoff = recptr.xrecoff % XLogSegSize;
if (startoff + nbytes > XLogSegSize)
segbytes = XLogSegSize - startoff;
else
segbytes = nbytes;
/* Need to seek in the file? */
if (recvOff != startoff)
{
if (lseek(recvFile, (off_t) startoff, SEEK_SET) < 0)
ereport(PANIC,
(errcode_for_file_access(),
errmsg("could not seek in log file %u, "
"segment %u to offset %u: %m",
recvId, recvSeg, startoff)));
recvOff = startoff;
}
/* OK to write the logs */
errno = 0;
byteswritten = write(recvFile, buf, segbytes);
if (byteswritten <= 0)
{
/* if write didn't set errno, assume no disk space */
if (errno == 0)
errno = ENOSPC;
ereport(PANIC,
(errcode_for_file_access(),
errmsg("could not write to log file %u, segment %u "
"at offset %u, length %lu: %m",
recvId, recvSeg,
recvOff, (unsigned long) segbytes)));
}
/* Update state for write */
XLByteAdvance(recptr, byteswritten);
recvOff += byteswritten;
nbytes -= byteswritten;
buf += byteswritten;
LogstreamResult.Write = recptr;
}
}
int
do_delete(pgBackupRange *range)
{
int i;
int ret;
parray *backup_list;
bool do_delete = false;
XLogRecPtr oldest_lsn = InvalidXLogRecPtr;
TimeLineID oldest_tli;
/* DATE are always required */
if (!pgBackupRangeIsValid(range))
elog(ERROR, "required delete range option not specified: delete DATE");
/* Lock backup catalog */
ret = catalog_lock();
if (ret == -1)
elog(ERROR, "can't lock backup catalog.");
else if (ret == 1)
elog(ERROR,
"another pg_arman is running, stop delete.");
/* Get complete list of backups */
backup_list = catalog_get_backup_list(NULL);
if (!backup_list)
elog(ERROR, "No backup list found, can't process any more.");
/* Find backups to be deleted */
for (i = 0; i < parray_num(backup_list); i++)
{
pgBackup *backup = (pgBackup *) parray_get(backup_list, i);
/* delete backup and update status to DELETED */
if (do_delete)
{
/* check for interrupt */
if (interrupted)
elog(ERROR, "interrupted during delete backup");
pgBackupDeleteFiles(backup);
continue;
}
/* Found the latest full backup */
if (backup->backup_mode >= BACKUP_MODE_FULL &&
backup->status == BACKUP_STATUS_OK &&
backup->start_time <= range->begin)
{
do_delete = true;
oldest_lsn = backup->start_lsn;
oldest_tli = backup->tli;
}
}
/* release catalog lock */
catalog_unlock();
/* cleanup */
parray_walk(backup_list, pgBackupFree);
parray_free(backup_list);
/*
* Delete in archive WAL segments that are not needed anymore. The oldest
* segment to be kept is the first segment that the oldest full backup
* found around needs to keep.
*/
if (!XLogRecPtrIsInvalid(oldest_lsn))
{
XLogSegNo targetSegNo;
char oldestSegmentNeeded[MAXFNAMELEN];
DIR *arcdir;
struct dirent *arcde;
char wal_file[MAXPGPATH];
int rc;
XLByteToSeg(oldest_lsn, targetSegNo);
XLogFileName(oldestSegmentNeeded, oldest_tli, targetSegNo);
elog(LOG, "Removing segments older than %s", oldestSegmentNeeded);
/*
* Now is time to do the actual work and to remove all the segments
* not needed anymore.
*/
if ((arcdir = opendir(arclog_path)) != NULL)
{
while (errno = 0, (arcde = readdir(arcdir)) != NULL)
{
/*
* We ignore the timeline part of the XLOG segment identifiers in
* deciding whether a segment is still needed. This ensures that
* we won't prematurely remove a segment from a parent timeline.
* We could probably be a little more proactive about removing
* segments of non-parent timelines, but that would be a whole lot
* more complicated.
*
* We use the alphanumeric sorting property of the filenames to
* decide which ones are earlier than the exclusiveCleanupFileName
* file. Note that this means files are not removed in the order
* they were originally written, in case this worries you.
*/
if ((IsXLogFileName(arcde->d_name) ||
IsPartialXLogFileName(arcde->d_name)) &&
strcmp(arcde->d_name + 8, oldestSegmentNeeded + 8) < 0)
{
/*
* Use the original file name again now, including any
* extension that might have been chopped off before testing
* the sequence.
*/
snprintf(wal_file, MAXPGPATH, "%s/%s",
arclog_path, arcde->d_name);
rc = unlink(wal_file);
if (rc != 0)
{
elog(WARNING, "could not remove file \"%s\": %s",
wal_file, strerror(errno));
break;
}
elog(LOG, "removed WAL segment \"%s\"", wal_file);
}
}
if (errno)
elog(WARNING, "could not read archive location \"%s\": %s",
arclog_path, strerror(errno));
if (closedir(arcdir))
elog(WARNING, "could not close archive location \"%s\": %s",
arclog_path, strerror(errno));
}
else
elog(WARNING, "could not open archive location \"%s\": %s",
arclog_path, strerror(errno));
}
return 0;
}
/*
* TODO: This is duplicate code with pg_xlogdump, similar to walsender.c, but
* we currently don't have the infrastructure (elog!) to share it.
*/
static void
XLogRead(char *buf, TimeLineID tli, XLogRecPtr startptr, Size count)
{
char *p;
XLogRecPtr recptr;
Size nbytes;
static int sendFile = -1;
static XLogSegNo sendSegNo = 0;
static uint32 sendOff = 0;
p = buf;
recptr = startptr;
nbytes = count;
while (nbytes > 0)
{
uint32 startoff;
int segbytes;
int readbytes;
startoff = recptr % XLogSegSize;
if (sendFile < 0 || !XLByteInSeg(recptr, sendSegNo))
{
char path[MAXPGPATH];
/* Switch to another logfile segment */
if (sendFile >= 0)
close(sendFile);
XLByteToSeg(recptr, sendSegNo);
XLogFilePath(path, tli, sendSegNo);
sendFile = BasicOpenFile(path, O_RDONLY | PG_BINARY, 0);
if (sendFile < 0)
{
if (errno == ENOENT)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("requested WAL segment %s has already been removed",
path)));
else
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open file \"%s\": %m",
path)));
}
sendOff = 0;
}
/* Need to seek in the file? */
if (sendOff != startoff)
{
if (lseek(sendFile, (off_t) startoff, SEEK_SET) < 0)
{
char path[MAXPGPATH];
XLogFilePath(path, tli, sendSegNo);
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not seek in log segment %s to offset %u: %m",
path, startoff)));
}
sendOff = startoff;
}
/* How many bytes are within this segment? */
if (nbytes > (XLogSegSize - startoff))
segbytes = XLogSegSize - startoff;
else
segbytes = nbytes;
readbytes = read(sendFile, p, segbytes);
if (readbytes <= 0)
{
char path[MAXPGPATH];
XLogFilePath(path, tli, sendSegNo);
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not read from log segment %s, offset %u, length %lu: %m",
path, sendOff, (unsigned long) segbytes)));
}
/* Update state for read */
recptr += readbytes;
sendOff += readbytes;
nbytes -= readbytes;
p += readbytes;
}
}
void
cdb_perform_redo(XLogRecPtr *redoCheckPointLoc, CheckPoint *redoCheckPoint, XLogRecPtr *newCheckpointLoc)
{
CheckPoint oldRedoCheckpoint;
uint32 logid;
uint32 seg;
int nsegsremoved;
if (redoCheckPointLoc->xlogid == 0 && redoCheckPointLoc->xrecoff == 0)
{
XLogGetRecoveryStart("QDSYNC", "for redo apply", redoCheckPointLoc, redoCheckPoint);
}
XLogStandbyRecoverRange(redoCheckPointLoc, redoCheckPoint, newCheckpointLoc);
/*
* Sample the recovery start location now to see if appling redo
* processed checkpoint records and moved the restart location forward.
*/
oldRedoCheckpoint = *redoCheckPoint;
XLogGetRecoveryStart("QDSYNC", "for redo progress check", redoCheckPointLoc, redoCheckPoint);
if (XLByteLT(oldRedoCheckpoint.redo,redoCheckPoint->redo))
{
ereport(LOG,
(errmsg("QDSYNC: transaction redo moved the restart location from %s to %s",
XLogLocationToString(&oldRedoCheckpoint.redo),
XLogLocationToString2(&redoCheckPoint->redo))));
}
else
{
Assert(XLByteEQ(oldRedoCheckpoint.redo,redoCheckPoint->redo));
ereport(LOG,
(errmsg("QDSYNC: transaction redo did not move the restart location %s forward this pass",
XLogLocationToString(&oldRedoCheckpoint.redo))));
return;
}
XLByteToSeg(redoCheckPoint->redo, logid, seg);
/*
* Delete offline log files (those no longer needed even for previous
* checkpoint).
*/
elog((Debug_print_qd_mirroring ? LOG : DEBUG5),
"QDSYNC: keep log files as far back as (logid %d, seg %d)",
logid, seg);
if (logid || seg)
{
PrevLogSeg(logid, seg);
elog((Debug_print_qd_mirroring ? LOG : DEBUG5),
"QDSYNC: delete offline log files up to (logid %d, seg %d)",
logid, seg);
XLogRemoveStandbyLogs(logid, seg, &nsegsremoved);
if (nsegsremoved > 0)
{
// Throw in extra new line to make log more readable.
ereport(LOG,
(errmsg("QDSYNC: %d logs removed through logid %d, seg %d\n",
nsegsremoved,
logid, seg)));
}
}
// Throw in extra new line to make log more readable.
elog(LOG,"--------------------------");
}
