---------------------------------------------------------------------------
Attached is a set of patches for a couple of bugs dealing with
timestamps in JDBC.
Bug#1) Incorrect timestamp stored in DB if client timezone different
than DB.
timestamps in JDBC.
Bug#1) Incorrect timestamp stored in DB if client timezone different
than DB.
The buggy implementation of setTimestamp() in PreparedStatement simply
used the toString() method of the java.sql.Timestamp object to convert
to a string to send to the database. The format of this is yyyy-MM-dd
hh:mm:ss.SSS which doesn't include any timezone information. Therefore
the DB assumes its timezone since none is specified. That is OK if the
timezone of the client and server are the same, however if they are
different the wrong timestamp is received by the server. For example if
the client is running in timezone GMT and wants to send the timestamp
for noon to a server running in PST (GMT-8 hours), then the server will
receive 2000-01-12 12:00:00.0 and interprete it as 2000-01-12
12:00:00-08 which is 2000-01-12 04:00:00 in GMT. The fix is to send a
format to the server that includes the timezone offset. For simplicity
sake the fix uses a SimpleDateFormat object with its timezone set to GMT
so that '+00' can be used as the timezone for postgresql. This is done
as SimpleDateFormat doesn't support formating timezones in the way
postgresql expects.
Bug#2) Incorrect handling of partial seconds in getting timestamps from
the DB
When the SimpleDateFormat object parses a string with a format like
yyyy-MM-dd hh:mm:ss.SS it expects the fractional seconds to be three
decimal places (time precision in java is miliseconds = three decimal
places). This seems like a bug in java to me, but it is unlikely to be
fixed anytime soon, so the postgresql code needed modification to
support the java behaviour. So for example a string of '2000-01-12
12:00:00.12-08' coming from the database was being converted to a
timestamp object with a value of 2000-01-12 12:00:00.012GMT-08:00. The
fix was to check for a '.' in the string and if one is found append on
an extra zero to the fractional seconds part.
Bug#3) Performance problems
In fixing the above two bugs, I noticed some things that could be
improved. In PreparedStatement.setTimestamp(),
PreparedStatement.setDate(), ResultSet.getTimestamp(), and
ResultSet.getDate() these methods were creating a new SimpleDateFormat
object everytime they were called. To avoid this unnecessary object
creation overhead, I changed the code to use static variables for
keeping a single instance of the needed formating objects.
Also the code used the + operator for string concatenation. As everyone
should know this is very inefficient and the use of StringBuffers is
prefered.
I also did some cleanup in ResultSet.getTimestamp(). This method has
had multiple patches applied some of which resulted in code that was no
longer needed. For example the ISO timestamp format that postgresql
uses specifies the timezone as an offset like '-08'. Code was added at
one point to convert the postgresql format to the java one which is
GMT-08:00, however the old code was left around which did nothing. So
there was code that looked for yyyy-MM-dd hh:mm:sszzzzzzzzz and
yyyy-MM-dd hh:mm:sszzz. This second format would never be encountered
because zzz (i.e. -08) would be converted into the former (also note
that the SimpleDateFormat object treats zzzzzzzzz and zzz the same, the
number of z's does not matter).
There was another problem/fix mentioned on the email lists today by
mcannon@internet.com which is also fixed by this patch:
Bug#4) Fractional seconds lost when getting timestamp from the DB
A patch by Jan Thomea handled the case of yyyy-MM-dd hh:mm:sszzzzzzzzz
but not the fractional seconds version yyyy-MM-dd hh:mm:ss.SSzzzzzzzzz.
The code is fixed to handle this case as well.
Barry Lind
are now critical sections, so as to ensure die() won't interrupt us while
we are munging shared-memory data structures. Avoid insecure intermediate
states in some code that proc_exit will call, like palloc/pfree. Rename
START/END_CRIT_CODE to START/END_CRIT_SECTION, since that seems to be
what people tend to call them anyway, and make them be called with () like
a function call, in hopes of not confusing pg_indent.
I doubt that this is sufficient to make SIGTERM safe anywhere; there's
just too much code that could get invoked during proc_exit().
- Fix handling of --tables=* (multiple tables never worked properly, AFAICT)
- strdup() the current user in DB routines
- Check results of IO routines more carefully.
- Check results of PQ routines more carefully.
Have not fixed index output yet.
1. Support of variable size keys - new algorithm of insertion to tree
(GLI - gist layrered insertion). Previous algorithm was implemented
as described in paper by Joseph M. Hellerstein et.al
"Generalized Search Trees for Database Systems". This (old)
algorithm was not suitable for variable size keys and could be
not effective ( walking up-down ) in case of multiple levels split
Bug fixed:
1. fixed bug in gistPageAddItem - key values were written to disk
uncompressed. This caused failure if decompression function
does real job.
2. NULLs handling - we keep NULLs in tree. Right way is to remove them,
but we don't know how to inform vacuum about index statistics. This is
just cosmetic warning message (like in case with R-Tree),
but I'm not sure how to recognize real problem if we remove NULLs
and suppress this warning as Tom suggested.
3. various memory leaks
This work was done by Teodor Sigaev (teodor@stack.net) and
Oleg Bartunov (oleg@sai.msu.su).
entry:
----------------------------
revision 1.2
date: 2000/12/04 01:20:38; author: tgl; state: Exp; lines:
+18 -18
Eliminate some of the more blatant platform-dependencies ... it
builds here now, anyway ...
----------------------------
Which basically changes u_int*_t -> uint*_t, so now it does not
compile neither under Debian 2.2 nor under NetBSD 1.5 which
is platform independent<B8> all right. Also it replaces $KAME$
with $Id$ which is Bad Thing. PostgreSQL Id should be added as a
separate line so the file history could be seen.
So here is patch:
* changes uint*_t -> uint*. I guess that was the original
intention
* adds uint64 type to include/c.h because its needed
[somebody should check if I did it right]
* adds back KAME Id, because KAME is the master repository
* removes stupid c++ comments in pgcrypto.c
* removes <sys/types.h> from the code, its not needed
--
marko
Marko Kreen
- no more elog(STOP) in StartupXLOG();
- both checkpoint' undo & redo are used to define
oldest on-line log file.
2. Ability to pre-allocate a few log files at checkpoint time
(wal_files option). Off by default.
as both a GROUP BY item and an output expression, the top-level Group
node should just copy up the evaluated expression value from its input,
rather than re-evaluating the expression. Aside from any performance
benefit this might offer, this avoids a crash when there is a sub-SELECT
in said expression.
before calling RelationInvalidateHeapTuple(), which is bad because the
latter needs to look at the tuple data, which is in the shared disk
buffer. If another backend manages to recycle the buffer while this
is going on, we will compute the wrong hashindex for the tuple or
maybe even crash outright. Must hold buffer refcount until afterwards.
(This bug is not in 7.0.*; seems to be have introduced during WAL changes.)
and burn. Just for added luck, change reading of CONST nodes so that
we do not need to consult pg_type rows while reading them; this means
that no database access occurs during stringToNode. This requires
changing the order in which const-node fields are written, which means
an initdb is forced.
in per-entry sub-memory-context, where they were supposed to go, rather
than in CacheMemoryContext where the code was putting them. Must've
suffered a severe brain fade when I wrote this :-(
sequences. This is done by disabling multi-byte awareness when it's
not necessary. This is kind of a workaround, not a perfect solution.
However, there is no ideal way to parse broken multi-byte character
sequences. So I guess this is the best way what we could do right
now...
and revert documentation to describe the existing INHERITS clause
instead, per recent discussion in pghackers. Also fix implementation
of SQL_inheritance SET variable: it is not cool to look at this var
during the initial parsing phase, only during parse_analyze(). See
recent bug report concerning misinterpretation of date constants just
after a SET TIMEZONE command. gram.y really has to be an invariant
transformation of the query string to a raw parsetree; anything that
can vary with time must be done during parse analysis.
Previous result did not have correct month boundaries so anything near edge
cases was suspect (e.g. April was in Q1 and July, August were lumped into
Q2).
Thanks to Denis Osadchy <osadchy@turbo.nsk.su> for the report.
The leak is caused by the memory allocation in
src/interfaces/ecpg/lib/execute.c in line 669 which is never freed.
Adding a "free(array_query);" after PQexec in line 671 seems to fix the
leak.
Thorsten Knabe
starting a new hashtable search no longer clobbers any other search
active anywhere in the system. Fix RelationCacheInvalidate() so that
it will not crash or go into an infinite loop if invoked recursively,
as for example by a second SI Reset message arriving while we are still
processing a prior one.
for Alpha gcc case. For Alpha non-gcc case, replace use of
__INTERLOCKED_TESTBITSS_QUAD builtin with __LOCK_LONG_RETRY and
__UNLOCK_LONG. The former does not execute an MB instruction and
therefore was guaranteed not to work on multiprocessor machines.
The LOCK_LONG builtins produce code that is the same in all essential
details as the gcc assembler code.
In theory we should always get EEXIST if there's a key collision, but
if the kernel code tests error conditions in a weird order, perhaps
EACCES or EIDRM could occur too.
assume that TAS() will always succeed the first time, even if the lock
is known to be free. Also, make sure that code will eventually time out
and report a stuck spinlock, rather than looping forever. Small cleanups
in s_lock.h, too.
drivers.
The first fix fixes the PreparedStatement object to not allocate
unnecessary objects when converting native types to Stings. The old
code used the following format:
(new Integer(x)).toString()
whereas this can more efficiently be occompilshed by:
Integer.toString(x);
avoiding the unnecessary object creation.
The second fix is to release some resources on the close() of a
ResultSet. Currently the close() method on ResultSet is a noop. The
purpose of the close() method is to release resources when the ResultSet
is no longer needed. The fix is to free the tuples cached by the
ResultSet when it is closed (by clearing out the Vector object that
stores the tuples). This is important for my application, as I have a
cache of Statement objects that I reuse. Since the Statement object
maintains a reference to the ResultSet and the ResultSet kept references
to the old tuples, my cache was holding on to a lot of memory.
Barry Lind
1. Distinguish cases where a Datum representing a tuple datatype is an OID
from cases where it is a pointer to TupleTableSlot, and make sure we use
the right typlen in each case.
2. Make fetchatt() and related code support 8-byte by-value datatypes on
machines where Datum is 8 bytes. Centralize knowledge of the available
by-value datatype sizes in two macros in tupmacs.h, so that this will be
easier if we ever have to do it again.
table that inherits from a temp table. Make sure the right things happen
if one creates a temp table, creates another temp that inherits from it,
then renames the first one. (Previously, system would end up trying to
delete the temp tables in the wrong order.)
recommendation from Paul Vixie. Add a new abbrev() function to produce
abbreviated format as text. No forced initdb, but new function is not
available unless you do an initdb or add the pg_proc row manually.
If pghost == "" and pgport == "" then PQsetdbLogin() fails with a
error message:
Is the postmaster running locally
and accepting connections on Unix socket '/tmp/.s.PGSQL.0'?
I see many applications such as PHP fails due to this behavior.
Now if pgport == "", then it is assumed to be a DEF_PGPORT_STR. This
is the same behavior as the version prior 7.1.
added to support character set encodings. However I noticed that the
encoding that is used isn't obtained from the DB. Since Java uses
unicode UCS2 internally the character set encoding is used to translate
strings from/to the DB encoding. So it seems logical that the code
would get the encoding from the DB instead of the current method of
requiring the user pass it as a parameter.
Attached is a patch that gets the DB encoding from the DB in the same
manner as is done in libpq/fe-connect.c. The patch is created off of
the latest CVS sources (Connection.java version 1.10).
Barry Lind
level" locks. A session lock is not released at transaction commit (but it
is released on transaction abort, to ensure recovery after an elog(ERROR)).
In VACUUM, use a session lock to protect the master table while vacuuming a
TOAST table, so that the TOAST table can be done in an independent
transaction.
I also took this opportunity to do some cleanup and renaming in the lock
code. The previously noted bug in ProcLockWakeup, that it couldn't wake up
any waiters beyond the first non-wakeable waiter, is now fixed. Also found
a previously unknown bug of the same kind (failure to scan all members of
a lock queue in some cases) in DeadLockCheck. This might have led to failure
to detect a deadlock condition, resulting in indefinite waits, but it's
difficult to characterize the conditions required to trigger a failure.
applied to the duplicated subtree twice. Probably someday we should
fix the parser not to generate multiple links to the same subtree,
but for now a quick copyObject() is the path of least resistance.
observed by Inoue. Also, don't call ProcRemove() from postmaster if we
have detected a backend crash --- too risky if shared memory is corrupted.
It's not needed anyway, considering we are going to reinitialize shared
memory and semaphores as soon as the last child is dead.
>> xlog.c : special case for beos to avoid 'link' which does not work yet
>> beos/sem.c : implementation of new sem_ctl call (GETPID) and a new
>sem_op
>> flag (IPCNOWAIT)
>> dynloader/beos.c : add a verification of symbol validity (seem that
the
>> loader sometime return OK with an invalid symbol)
>> postmaster.c : add beos forking support for the new checkpoint
process
>> postgres.c : remove beos special case for getrusage
>> beos.h : Correction of a bas definition of AF_UNIX, misc defnitions
>>
>>
>> thanks
>>
>>
>> cyril
Cyril VELTER
might change it. Experimentation shows that the signal handler call
mechanism does not save/restore errno for you, at least not on Linux
or HPUX, so this is definitely a real risk.
to ensure that we have released buffer refcounts and so forth, rather than
putting ad-hoc operations before (some of the calls to) proc_exit. Add
commentary to discourage future hackers from repeating that mistake.
> Date: Thu, 14 Dec 2000 12:44:47 +0100 (CET)
> From: Kovacs Zoltan Sandor <tip@pc10.radnoti-szeged.sulinet.hu>
> To: pgsql-bugs@postgresql.org
> Subject: [BUGS] to_char() causes backend to close connection
>
> Hi, this query gives different strange results:
>
> select to_char(now()::abstime,'YYMMDDHH24MI');
>
> I get e.g. a "backend closed the channel unexpectedly..." error with
> successful or failed resetting attempt (indeterministic)
Again thanks Kovacs, you found really designing bug, that appear
if anyone write bad format template to "number" version of to_char()
(as you with 'DD').
Karel
comparison does not consider paths different when they differ only in
uninteresting aspects of sort order. (We had a special case of this
consideration for indexscans already, but generalize it to apply to
ordered join paths too.) Be stricter about what is a canonical pathkey
to allow faster pathkey comparison. Cache canonical pathkeys and
dispersion stats for left and right sides of a RestrictInfo's clause,
to avoid repeated computation. Total speedup will depend on number of
tables in a query, but I see about 4x speedup of planning phase for
a sample seven-table query.
OIDs rather than names. Aside from being simpler and faster, this way
doesn't blow up in the face of 'create temp table foo () inherits (foo)'.
Which is a rather odd thing to do, but it seems some people want to.
avoid repeated evaluations in cost_qual_eval(). This turns out to save
a useful fraction of planning time. No change to external representation
of RestrictInfo --- although that node type doesn't appear in stored
rules anyway.
all forms of foreign keys be exposed to SQLForeignKeys. This patch is in
addition to the ones I mailed yesterday (forget had I changed that as
well....)
Michael Fork - CCNA - MCP - A+
Network Support - Toledo Internet Access - Toledo Ohio
return foreign key information based on the pg_trigger system table. I
have tested the patch with (what I believe) is all possible
primary/foreign key combinations -- however I may have missed some, so if
anyone feels like taking the patch for a test drive, here are some useful
links:
Michael Fork
varlena type. (I did not force initdb, but you won't see the fix
unless you do one.) Also, make sure all index support operators and
functions are careful not to leak memory for toasted inputs; I had
missed some hash and rtree support ops on this point before.
value greater than one. The behavior this sought to disallow doesn't
seem any less confusing than the other behaviors of cached sequences.
Improve wording of some error messages, too.
Update documentation accordingly. Also add an explanation that
aborted transactions do not roll back their nextval() calls; this
seems to be a FAQ, so it ought to be mentioned here...
As I read it, the spec requires a non-null result in some cases where
one of the inputs is NULL: specifically, if the other endpoint of that
interval is between the endpoints of the other interval, then the result
is known TRUE despite the missing endpoint. The spec could've been a
lot simpler if they did not intend this behavior.
I did not force an initdb for this change, but if you don't do one you'll
still see the old strict-function behavior.
work where we can (given that the executor only handles it at top level)
and generate an error where we can't. Note that while the parser has
been allowing views to say SELECT FOR UPDATE for a few weeks now, that
hasn't actually worked until just now.
an error as we used to. In an OUTER JOIN scenario, retrieving a null
CTID from one of the input relations is entirely expected. We still
want to lock the input rows from the other relations, so just ignore
the null and keep going.
I believe this should fix the issue that Philip Warner
noticed about the check for unique constraints meeting the
referenced keys of a foreign key constraint allowing the
specification of a subset of a foreign key instead of
rejecting it. I also added tests for a base case of
this to the foreign key and alter table tests and patches
for expected output.
report from Joel Burton. Turns out that my simple idea of turning the
SELECT into a subquery does not interact well *at all* with the way the
rule rewriter works. Really what we need to make INSERT ... SELECT work
cleanly is to decouple targetlists from rangetables: an INSERT ... SELECT
wants to have two levels of targetlist but only one rangetable. No time
for that for 7.1, however, so I've inserted some ugly hacks to make the
rewriter know explicitly about the structure of INSERT ... SELECT queries.
Ugh :-(
Allow some operator-like tokens to be used as function names.
Flesh out support for time, timetz, and interval operators
and interactions.
Regression tests pass, but non-reference-platform horology test results
will need to be updated.
anymore. That won't teach us anything new for the rest of this release
cycle, so it seems better to keep the --assert environment more like the
non-assert environment for beta.
I'm going to leave CLOBBER_FREED_MEMORY and MEMORY_CONTEXT_CHECKING
turned on by --enable-cassert for now, however.
since those routines may do palloc's. We want to be fairly sure we can
send the error message to the client even under low-memory conditions.
That's what we stashed away 8K in ErrorContext for, after all ...
not-very-good handling of mid-size allocation requests. Do everything via
either the "small" case (chunk size rounded up to power of 2) or the "large"
case (pass it straight off to malloc()). Increase the number of freelists
a little to set the breakpoint between these behaviors at 8K.
by without them.
Don't check for preprocessor symbols from system header files in port
include files, since those header files aren't included at this point.
$(CC) $(CFLAGS) $(LDFLAGS) <object files> <extra-libraries> $(LIBS) -o $@
This form seemed to be the most portable, readable, and logical, but in any
case it's better than having a dozen different ones in the tree.
both MULTIBYTE and TOAST prevent char(n) from being truly fixed-size.
Simplify and speed up fastgetattr() and index_getattr() macros by
eliminating special cases for attnum=1. It's just as fast to handle
the first attribute by presetting its attcacheoff to zero; so do that
instead when loading the tupledesc in relcache.c.
included by everything that includes bufmgr.h --- it's supposed to be
internals, after all, not part of the API! This fixes the conflict
against FreeBSD headers reported by Rosenman, by making it unnecessary
for s_lock.h to be included by plperl.c.
postmaster, because it isn't updated after forking away from the terminal.
Apparently it's not used anyplace in the postmaster ... but seems best
to make it show the correct PID ...
socket file, in favor of having an ordinary lockfile beside the socket file.
Clean up a few robustness problems in the lockfile code. If postmaster is
going to reject a connection request based on database state, it will now
tell you so before authentication exchange not after. (Of course, a failure
after is still possible if conditions change meanwhile, but this makes life
easier for a yet-to-be-written pg_ping utility.)
IPC key assignment will now work correctly even when multiple postmasters
are using same logical port number (which is possible given -k switch).
There is only one shared-mem segment per postmaster now, not 3.
Rip out broken code for non-TAS case in bufmgr and xlog, substitute a
complete S_LOCK emulation using semaphores in spin.c. TAS and non-TAS
logic is now exactly the same.
When deadlock is detected, "Deadlock detected" is now the elog(ERROR)
message, rather than a NOTICE that comes out before an unhelpful ERROR.
--------------------------------------------------
Subject: Bug in unicode conversion ...
From: Jan Varga <varga@utcru.sk>
To: t-ishii@sra.co.jp
Date: Sat, 18 Nov 2000 17:41:20 +0100 (CET)
Hi,
I tried this new feature in PostgreSQL. I found one bug.
Script UCS_to_8859.pl skips input lines which
1. code <0x80 or
2. ucs <0x100
I think second one is not good idea because some codes in ISO8859-2
have ucs <0x100 (e.g. 0xE9 - 0x00E9)
--------------------------------------------------
for any other purpose than PGLC_localeconv()'s internal save/restore of
locale settings. Fix cash.c to call PGLC_localeconv() rather than
making a direct call to localeconv() --- the old way, if PGLC_localeconv()
had already cached a locale result, it would be overwritten by the first
cash_in or cash_out operation, leading to wrong-locale results later.
Probably no demonstrable bug today, since we only appear to be looking
at the LC_MONETARY results which should be the same anyway, but definitely
a gotcha waiting to strike.
re-adopt these settings at every postmaster or standalone-backend startup.
This should fix problems with indexes becoming corrupt due to failure to
provide consistent locale environment for postmaster at all times. Also,
refuse to start up a non-locale-enabled compilation in a database originally
initdb'd with a non-C locale. Suppress LIKE index optimization if locale
is not "C" or "POSIX" (are there any other locales where it's safe?).
Issue NOTICE during initdb if selected locale disables LIKE optimization.