Tom Lane [Sat, 17 Mar 2001 20:54:13 +0000 (20:54 +0000)]
Rearrange XLogFileInit so that control-file spinlock is not held while filling
the new log file with zeroes, only while renaming it into place. This should
prevent problems with 'stuck spinlock' errors under heavy load.
Tom Lane [Fri, 16 Mar 2001 05:44:33 +0000 (05:44 +0000)]
Support syncing WAL log to disk using either fsync(), fdatasync(),
O_SYNC, or O_DSYNC (as available on a given platform). Add GUC parameter
to control sync method.
Also, add defense to XLogWrite to prevent it from going nuts if passed
a target write position that's past the end of the buffers so far filled
by XLogInsert.
Tom Lane [Fri, 16 Mar 2001 05:08:39 +0000 (05:08 +0000)]
Resetting the XLOG can't include rewinding it to logical position zero,
because we need page LSNs stored in the main database to be less than
the current XLOG position. Hence, generate the new XLOG segment at last
old segment number plus one.
Hiroshi Inoue [Fri, 16 Mar 2001 01:17:23 +0000 (01:17 +0000)]
Add multibyte support.
Provide an extenisible scheme of encoding conversion.
As the first step, SJIS and BIG5 are supported.
From now on multibyte people would be happy to use
this psqlodbc driver.
Tom Lane [Wed, 14 Mar 2001 21:47:50 +0000 (21:47 +0000)]
COPY should handle after-insert triggers the same as execMain.c does.
I'm not sure that it's really necessary to save insert events when there
are only after update or delete triggers, but certainly it's wrong for
COPY to behave differently from an INSERT query.
Tom Lane [Wed, 14 Mar 2001 20:12:10 +0000 (20:12 +0000)]
Remove obsolete PowerPC-specific hack for comparisons to DBL_MIN
(per recent discussion with Tatsuo). Hopefully the compilers with
that old bug are all long gone.
Tom Lane [Tue, 13 Mar 2001 20:32:37 +0000 (20:32 +0000)]
Change xlog page-header format to include StartUpID. Use the SUI to
detect case that next page in log came from an older run than the prior
page. This avoids the necessity to re-zero the log after recovery from
a crash, which is good because we need not risk destroying valuable log
information.
This forces another initdb since yesterday :-(. Need to get that log
reset utility done...
Hiroshi Inoue [Tue, 13 Mar 2001 09:03:24 +0000 (09:03 +0000)]
The attached file contains the registry settings required to install the
ODBC driver on Windows 9X/ME/NT/2K when using the later versions of the
driver that don't have the Installshield installation:
1) Install psqlodbc.dll in to C:\Windows\System or C:\Winnt\System32
2) Add the registry settings in the attached file using regedit.
Tom Lane [Tue, 13 Mar 2001 01:17:06 +0000 (01:17 +0000)]
XLOG (and related) changes:
* Store two past checkpoint locations, not just one, in pg_control.
On startup, we fall back to the older checkpoint if the newer one
is unreadable. Also, a physical copy of the newest checkpoint record
is kept in pg_control for possible use in disaster recovery (ie,
complete loss of pg_xlog). Also add a version number for pg_control
itself. Remove archdir from pg_control; it ought to be a GUC
parameter, not a special case (not that it's implemented yet anyway).
* Suppress successive checkpoint records when nothing has been entered
in the WAL log since the last one. This is not so much to avoid I/O
as to make it actually useful to keep track of the last two
checkpoints. If the things are right next to each other then there's
not a lot of redundancy gained...
* Change CRC scheme to a true 64-bit CRC, not a pair of 32-bit CRCs
on alternate bytes. Polynomial borrowed from ECMA DLT1 standard.
* Fix XLOG record length handling so that it will work at BLCKSZ = 32k.
* Change XID allocation to work more like OID allocation. (This is of
dubious necessity, but I think it's a good idea anyway.)
* Fix a number of minor bugs, such as off-by-one logic for XLOG file
wraparound at the 4 gig mark.
* Add documentation and clean up some coding infelicities; move file
format declarations out to include files where planned contrib
utilities can get at them.
* Checkpoint will now occur every CHECKPOINT_SEGMENTS log segments or
every CHECKPOINT_TIMEOUT seconds, whichever comes first. It is also
possible to force a checkpoint by sending SIGUSR1 to the postmaster
(undocumented feature...)
* Defend against kill -9 postmaster by storing shmem block's key and ID
in postmaster.pid lockfile, and checking at startup to ensure that no
processes are still connected to old shmem block (if it still exists).
* Switch backends to accept SIGQUIT rather than SIGUSR1 for emergency
stop, for symmetry with postmaster and xlog utilities. Clean up signal
handling in bootstrap.c so that xlog utilities launched by postmaster
will react to signals better.
* Standalone bootstrap now grabs lockfile in target directory, as added
insurance against running it in parallel with live postmaster.
Tom Lane [Mon, 12 Mar 2001 23:02:00 +0000 (23:02 +0000)]
Avoid O(N^2) behavior in deferredTriggerAddEvent() for large numbers of
tuples inserted/deleted/updated in a single transaction. On my machine,
this reduced the time to delete 80000 tuples in a foreign-key-referencing
table from ~15min to ~8sec.
Peter Eisentraut [Sun, 11 Mar 2001 11:07:01 +0000 (11:07 +0000)]
Add uninstall target to Java build.
Respect default port setting in JDBC driver.
Pick up version number from Makefile.global.
Change installation directory to share/java/.
Document.
Tom Lane [Sat, 10 Mar 2001 23:15:20 +0000 (23:15 +0000)]
Turns out the HPUX linker likes -Bsymbolic too. Without this, ODBC
driver does not work because its internal cross-references get bound
to similarly named functions in unixODBC shared library.
Tom Lane [Sat, 10 Mar 2001 04:21:51 +0000 (04:21 +0000)]
FATAL errors should cause exit with nonzero status if we are not running
under the postmaster --- specifically, if we are a standalone backend
running under the initdb script, this is critical!
Tom Lane [Thu, 8 Mar 2001 01:49:01 +0000 (01:49 +0000)]
A subplan invoked within an aggregate function's argument should
be allowed to receive ungrouped variables of the current query level.
Curious that no one reported this bug before...
Tom Lane [Thu, 8 Mar 2001 00:24:34 +0000 (00:24 +0000)]
Modify wchar conversion routines to not fetch the next byte past the end
of a counted input string. Marinos Yannikos' recent crash report turns
out to be due to applying pg_ascii2wchar_with_len to a TEXT object that
is smack up against the end of memory. This is the second just-barely-
reproducible bug report I have seen that traces to some bit of code
fetching one more byte than it is allowed to. Let's be more careful
out there, boys and girls.
While at it, I changed the code to not risk a similar crash when there
is a truncated multibyte character at the end of an input string. The
output in this case might not be the most reasonable output possible;
if anyone wants to improve it further, step right up...
Tom Lane [Wed, 7 Mar 2001 21:20:26 +0000 (21:20 +0000)]
Repair a number of places that didn't bother to check whether PageAddItem
succeeds or not. Revise rtree page split algorithm to take care about
making a feasible split --- ie, will the incoming tuple actually fit?
Failure to make a feasible split, combined with failure to notice the
failure, account for Jim Stone's recent bug report. I suspect that
hash and gist indices may have the same type of bug, but at least now
we'll get error messages rather than silent failures if so. Also clean
up rtree code to use Datum rather than char* where appropriate.
Bruce Momjian [Wed, 7 Mar 2001 16:18:08 +0000 (16:18 +0000)]
> Applied. Thanks.
One more :)) It's for improper function argumets for
PLTCL_UNKNOWN_SUPPORT code
I'm not an autoconf expert, but is it possible to enable unknown
support in pltcl with configure option ?
This support is really handy for real life usage of pl/tcl.
Peter Mount [Tue, 6 Mar 2001 12:04:46 +0000 (12:04 +0000)]
Tue Mar 06 12:05:00 GMT 2001 peter@retep.org.uk
- Removed org.postgresql.xa.Test from the JDBC EE driver as it's an old
test class and prevented it from compiling.
Philip Warner [Tue, 6 Mar 2001 04:08:04 +0000 (04:08 +0000)]
- Only disable triggers in DataOnly (or implied data-only) restores.
- Change -U option to -L to allow -U to specify username in future.\7f (pg_restore)
Bruce Momjian [Mon, 5 Mar 2001 18:42:57 +0000 (18:42 +0000)]
I'm attaching those diffs for the Reference Guide in a tar file, as
not all of them attached properly in the post I made a few minutes
ago. Please disregard those earlier files. The diffs in the tar file
replace them.