summaryrefslogtreecommitdiff
path: root/src/backend/access/transam/xlog.c
AgeCommit message (Expand)Author
2007-10-12When telling the bgwriter that we need a checkpoint because too much xlogTom Lane
2007-09-30Adjust recovery PS display as agreed with Simon: 'waiting for XXX'Tom Lane
2007-09-29Make recovery show the current input WAL segment name in the startupTom Lane
2007-09-29Make archive recovery always start a new timeline, rather than only when aTom Lane
2007-09-26Minor improvements in backup and recovery:Tom Lane
2007-09-08Replace the former method of determining snapshot xmax --- to wit, callingTom Lane
2007-09-05Implement lazy XID allocation: transactions that do not modify any databaseTom Lane
2007-08-28Add a debug logging message when a resource manager rejects an attemptedTom Lane
2007-08-13Fix two bugs induced in VACUUM FULL by async-commit patch.Tom Lane
2007-08-04Switch over to using the src/timezone functions for formatting timestampsTom Lane
2007-08-01Support an optional asynchronous commit mode, in which we don't flush WALTom Lane
2007-07-24Create a new dedicated Postgres process, "wal writer", which exists to writeTom Lane
2007-06-30Improve logging of checkpoints. Patch by Greg Smith, worked overTom Lane
2007-06-28Implement "distributed" checkpoints in which the checkpoint I/O is spreadTom Lane
2007-05-31Make some messages more consistentPeter Eisentraut
2007-05-31Downgrade some low-level startup messages to DEBUG1.Peter Eisentraut
2007-05-30Make large sequential scans and VACUUMs work in a limited-size "ring" ofTom Lane
2007-05-20To support external compression of archived WAL data, add a flag bit toTom Lane
2007-04-30Change the timestamps recorded in transaction commit/abort xlog recordsTom Lane
2007-04-03Remove the CheckpointStartLock in favor of having backends show whether theyTom Lane
2007-04-03Decouple the values of TOAST_TUPLE_THRESHOLD and TOAST_MAX_CHUNK_SIZE.Tom Lane
2007-03-03Remove undo information from pg_controldata --- never used.Bruce Momjian
2007-02-14Move fsync method macro defines into /include/access/xlogdefs.h so theyBruce Momjian
2007-02-08Normalize fgets() calls to use sizeof() for calculating the buffer sizePeter Eisentraut
2007-02-07Remove the xlog-centric "database system is ready" message and replace it withTom Lane
2007-02-01Wording cleanup for error messages. Also change can't -> cannot.Bruce Momjian
2007-01-05Update CVS HEAD for 2007 copyright. Back branches are typically notBruce Momjian
2006-12-08Remove the logId/logSeg fields from pg_control, because they are not neededTom Lane
2006-11-30Minor adjustments to make failures in startup/shutdown behave more cleanly.Tom Lane
2006-11-21On systems that have setsid(2) (which should be just about everything exceptTom Lane
2006-11-16String fixPeter Eisentraut
2006-11-10Clean up some misleading references to %p being a full path, per Simon.Tom Lane
2006-11-08Change Windows rename and unlink substitutes so that they time out afterTom Lane
2006-11-05Fix recently-understood problems with handling of XID freezing, particularlyTom Lane
2006-10-18Add some code to CREATE DATABASE to check for pre-existing subdirectoriesTom Lane
2006-10-06Message style improvementsPeter Eisentraut
2006-10-04pgindent run for 8.2.Bruce Momjian
2006-08-21Make the server track an 'XID epoch', that is, maintain higher-order bitsTom Lane
2006-08-17Implement archive_timeout feature to force xlog file switches to occur no moreTom Lane
2006-08-07Make recovery from WAL be restartable, by executing a checkpoint-likeTom Lane
2006-08-06Add support for forcing a switch to a new xlog file; cause such a switchTom Lane
2006-07-30Modify snapshot definition so that lazy vacuums are ignored by otherAlvaro Herrera
2006-07-14Remove 576 references of include files that were not needed.Bruce Momjian
2006-07-13Allow include files to compile own their own.Bruce Momjian
2006-06-27Put #ifdef NOT_USED around posix_fadvise call. We may want to resurrectTom Lane
2006-06-22pg_stop_backup was calling XLogArchiveNotify() twice for the newly createdTom Lane
2006-06-18Don't try to call posix_fadvise() unless <fcntl.h> supplies a declarationTom Lane
2006-06-16Test for POSIX_FADV_DONTNEED to use posix_fadvise().Bruce Momjian
2006-06-15Use posix_fadvise() to avoid kernel caching of WAL contents on WAL fileBruce Momjian
2006-04-20Ensure that we validate the page header of the first page of a WAL fileTom Lane