database/postgres
database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-04-24 10:47:04 +03:00
Code

Reduce checkpoints and WAL traffic on low activity database server

Browse Source 
Previously, we skipped a checkpoint if no WAL had been written since
last checkpoint, though this does not appear in user documentation.
As of now, we skip a checkpoint until we have written at least one
enough WAL to switch the next WAL file. This greatly reduces the
level of activity and number of WAL messages generated by a very
low activity server. This is safe because the purpose of a checkpoint
is to act as a starting place for a recovery, in case of crash.
This patch maintains minimal WAL volume for replay in case of crash,
thus maintaining very low crash recovery time.
This commit is contained in:
Simon Riggs 2011-11-02 15:26:33 +00:00
parent 9aceb6ab3c
commit 18fb9d8d21
1 changed files with 15 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
src/backend/access/transam/xlog.c
View File

@ -7560,6 +7560,10 @@ CreateCheckPoint(int flags)
uint32 freespace;
uint32 _logId;
uint32 _logSeg;
uint32 redo_logId;
uint32 redo_logSeg;
uint32 insert_logId;
uint32 insert_logSeg;
TransactionId *inCommitXids;
int nInCommit;
@ -7636,8 +7640,8 @@ CreateCheckPoint(int flags)
LWLockAcquire(WALInsertLock, LW_EXCLUSIVE);
/*
* If this isn't a shutdown or forced checkpoint, and we have not inserted
* any XLOG records since the start of the last checkpoint, skip the
* If this isn't a shutdown or forced checkpoint, and we have not switched
* to the next WAL file since the start of the last checkpoint, skip the
* checkpoint. The idea here is to avoid inserting duplicate checkpoints
* when the system is idle. That wastes log space, and more importantly it
* exposes us to possible loss of both current and previous checkpoint
@ -7645,10 +7649,11 @@ CreateCheckPoint(int flags)
* (Perhaps it'd make even more sense to checkpoint only when the previous
* checkpoint record is in a different xlog page?)
*
* We have to make two tests to determine that nothing has happened since
* the start of the last checkpoint: current insertion point must match
* the end of the last checkpoint record, and its redo pointer must point
* to itself.
* While holding the WALInsertLock we find the current WAL insertion point
* and compare that with the starting point of the last checkpoint, which
* is the redo pointer. We use the redo pointer because the start and end
* points of a checkpoint can be hundreds of files apart on large systems
* when checkpoint writes are spread out over time.
*/
if ((flags & (CHECKPOINT_IS_SHUTDOWN | CHECKPOINT_END_OF_RECOVERY |
CHECKPOINT_FORCE)) == 0)
@ -7656,13 +7661,10 @@ CreateCheckPoint(int flags)
XLogRecPtr curInsert;
INSERT_RECPTR(curInsert, Insert, Insert->curridx);
if (curInsert.xlogid == ControlFile->checkPoint.xlogid &&
curInsert.xrecoff == ControlFile->checkPoint.xrecoff +
MAXALIGN(SizeOfXLogRecord + sizeof(CheckPoint)) &&
ControlFile->checkPoint.xlogid ==
ControlFile->checkPointCopy.redo.xlogid &&
ControlFile->checkPoint.xrecoff ==
ControlFile->checkPointCopy.redo.xrecoff)
XLByteToSeg(curInsert, insert_logId, insert_logSeg);
XLByteToSeg(ControlFile->checkPointCopy.redo, redo_logId, redo_logSeg);
if (insert_logId == redo_logId &&
insert_logSeg == redo_logSeg)
{
LWLockRelease(WALInsertLock);
LWLockRelease(CheckpointLock);