* Alfred Perlstein [EMAIL PROTECTED] [001205 04:00]:
* Larry Rosenman [EMAIL PROTECTED] [001128 20:44] wrote:
* Tom Lane [EMAIL PROTECTED] [001128 22:31]:
Larry Rosenman [EMAIL PROTECTED] writes:
The last batch of commits break on FreeBSD 4.2-STABLE.
On Mon, Dec 04, 2000 at 09:09:55AM +0100, [EMAIL PROTECTED] wrote:
i have wrote an application dealing with ean13 and ean8 type,how can i
submit it ??
Post a link to your patches here and see if it generates some
interest. Some description would be nice too, what you are
exactly trying to
It seems that R-trees become inefficient when the number of dimensions
increase. Has anyone thoght of a transparent way to use Peano codes (hhcode
in Oracle lingo), and use B-tree indexes instead?
Also, I've read that R-trees sometimes suffer a lot when an update overflows
a node in the index.
Here's the query that, given the primary key table, lists all foreign
keys, their tables, the RI type, and defereability.
Michael Fork - CCNA - MCP - A+
Network Support - Toledo Internet Access - Toledo Ohio
SELECT pg_trigger.tgargs,
pg_trigger.tgnargs,
pg_trigger.tgdeferrable,
I'm debugging some code here where I get problems related to
spinlocks, anyhow, while running through the files I noticed
that the UNLOCK code seems sort of broken.
What I mean is that on machines that have loosely ordered
memory models you can have problems because of data that's
supposed to be
On FreeBSD 4.1.1 and above there's a sysctl tunable called
kern.ipc.shm_use_phys, when set to 1 it's supposed to
make the kernel's handling of shared memory much more
effecient at the expense or making the shm segment unpageable.
I tried to use this option with 7.0.3 and FreeBSD 4.2 but
for some
* Tom Lane [EMAIL PROTECTED] [001205 07:14] wrote:
Alfred Perlstein [EMAIL PROTECTED] writes:
Anyhow, to address the problem I've removed struct mount from
userland visibility in both FreeBSD 5.x (current) and FreeBSD 4.x
(stable).
That might fix things on your box, but we can hardly
Alfred Perlstein [EMAIL PROTECTED] writes:
I'm pretty sure you guys need memory barrier ops.
On a machine that requires such a thing, the assembly code for UNLOCK
should include it. Want to provide a patch?
regards, tom lane
* Tom Lane [EMAIL PROTECTED] [001205 07:24] wrote:
Alfred Perlstein [EMAIL PROTECTED] writes:
I'm pretty sure you guys need memory barrier ops.
On a machine that requires such a thing, the assembly code for UNLOCK
should include it. Want to provide a patch?
My assembler is extremely
Actually, Alfred is a FreeBSD committer, and committed it
to the FreeBSD source tree.
It's for ALL at FreeBSD 4-STABLE as of today.
LER
-Original Message-
From: Tom Lane [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, December 05, 2000 9:14 AM
To: Alfred Perlstein
Cc: Larry Rosenman;
On Tue, Dec 05, 2000 at 10:07:37AM +0100, Zeugswetter Andreas SB wrote:
And using the following program for timing thread creation
and cleanup:
#include pthread.h
threadfn() { pthread_exit(0); }
I think you would mainly need to test how the system behaves, if
the threads and
Alfred Perlstein [EMAIL PROTECTED] writes:
Here's the log, the number in parens is the address of the lock,
on tas() the value printed to the right is the value in _ret,
for the others, it's the value before the lock count is set.
This looks to be the trace of a SpinAcquire()
(see
* Tom Lane [EMAIL PROTECTED] [001205 07:43] wrote:
Alfred Perlstein [EMAIL PROTECTED] writes:
Here's the log, the number in parens is the address of the lock,
on tas() the value printed to the right is the value in _ret,
for the others, it's the value before the lock count is set.
This
BTW, I just remembered that in 7.0.*, the SLocks that are managed by
SpinAcquire() all live in their own little shm segment. On a machine
where slock_t is char, it'd likely only amount to 128 bytes or so.
Maybe you are seeing some bug in FreeBSD's handling of tiny shm
segments?
On Tue, 5 Dec 2000, Tom Lane wrote:
Hmm. The rule will generate a query along these lines:
DELETE FROM ips_free
FROM ips_free ipsfree2
WHERE ips_free.block_id = ipsfree2.block_id
AND ips_free.ip = ipsfree2.ip
AND ipsfree2.ip = '10.10.10.10';
(I'm using ipsfree2 to
I have been watching this thread vs non-threaded discussion and am completely with the
process-only crew for a couple reasons, but lets look at a few things:
The process vs threads benchmark which showed 160us vs 120us, only did the process
creation, not the delayed hit of the "copy on write"
As far as I know (and have tested in excess) Informix IDS
does survive any power loss without leaving the db in a
corrupted state. The basic technology is, that it only relys
on writes to one "file" (raw device in that case), the txlog,
which is directly written. All writes to the txlog are
On Sunday 03 December 2000 04:00, Vadim Mikheev wrote:
There is risk here. It isn't so much in the fact that PostgreSQL, Inc
is doing a couple of modest closed-source things with the code. After
all, the PG community has long acknowleged that the BSD license would
allow others to co-op
On Sunday 03 December 2000 12:41, mlw wrote:
Thomas Lockhart wrote:
As soon as you find a business model which does not require income, let
me know. The .com'ers are trying it at the moment, and there seems to be
a few flaws... ;)
While I have not contributed anything to Postgres yet, I
On Sunday 03 December 2000 21:49, The Hermit Hacker wrote:
I've been trying to follow this thread, and seem to have missed where
someone arrived at the conclusion that we were proprietarizing(word?) this
I have missed that part as well.
... we do apologize that it didn't get out
On Tue, Dec 05, 2000 at 02:52:48PM -0500, Tom Lane wrote:
There aren't going to be all that many data pages needing the COW
treatment, because the postmaster uses very little data space of its
own. I think this would become an issue if we tried to have the
postmaster pre-cache catalog
Minor usability/debuggability suggestion...
RI violation error messages in 7.0.0 do not appear to identify the
offending value.
Example:
ERROR: fk_employee_currency referential integrity violation - key
referenced from employee not found in currency
Easier to debug would be:
ERROR:
I totaly missed your point here. How closing source of
ERserver is related to closing code of PostgreSQL DB server?
Let me clear things:
1. ERserver isn't based on WAL. It will work with any version = 6.5
2. WAL was partially sponsored by my employer, Sectorbase.com,
not
Greetings! 'copy from stdin' on 7.0.2 appears to simply hang if more
than ~ 1000 records follow in one shot. I couldn't see this behavior
documented anywhere. Is this a bug?
Take care,
--
Camm Maguire[EMAIL PROTECTED]
Greetings! I've noticed in the documentation that the sql standard
requires foreign keys to reference primary key/(or maybe just unique)
columns, but that postgresql does not enforce this. Is this a feature
that is intended to persist, or a temporary deviation from the sql
standard? The
Alfred,
do you have any numbers with and without your patch ?
I mean performance. You may use pg_check utility.
Oleg
On Tue, 5 Dec 2000, Alfred Perlstein wrote:
Date: Tue, 5 Dec 2000 13:04:45 -0800
From: Alfred Perlstein [EMAIL PROTECTED]
To: Tom Lane [EMAIL PROTECTED]
Cc: [EMAIL
On Tue, 5 Dec 2000, Martin A. Marques wrote:
On Sunday 03 December 2000 04:00, Vadim Mikheev wrote:
There is risk here. It isn't so much in the fact that PostgreSQL, Inc
is doing a couple of modest closed-source things with the code. After
all, the PG community has long acknowleged
The Hermit Hacker wrote:
its been brought up and rejected continuously ... in some of our opinions,
GPL is more harmful then helpful ... as has been said before many times,
and I'm sure will continue to be said "changing the license to GPL is a
non-discussable issue" ...
I've declined
I am running 7.0 and for columns that have type 'timestamp'
the values end up with the format of year-month-day HH:MM:SS-[0-n]
e.g.
2000-12-05 15:58:12-06
the trailing -n (e.g. -06) is killing the JDBC driver.
Is there a work around. No matter what I Insert, a trailing
-0n
Regardless of what license is best, could the license even be changed now? I
mean, some of the initial Berkeley code is still in there in some sense and
I would think that the original license (BSD I assume) of the initial source
code release would have to be somehow honored.. I'm just wondering
On Tuesday 05 December 2000 18:03, The Hermit Hacker wrote:
Has somebody thought about putting PG in the GPL licence instead of the
BSD?
its been brought up and rejected continuously ... in some of our opinions,
GPL is more harmful then helpful ... as has been said before many times,
Mitch Vincent wrote:
Regardless of what license is best, could the license even be changed now? I
mean, some of the initial Berkeley code is still in there in some sense and
I would think that the original license (BSD I assume) of the initial source
code release would have to be somehow
Lamar Owen [EMAIL PROTECTED] writes:
Mitch Vincent wrote:
Regardless of what license is best, could the license even be changed now? I
mean, some of the initial Berkeley code is still in there in some sense and
I would think that the original license (BSD I assume) of the initial
Short Description
foreign key check makes a big LOCK
Long Description
in: src/backend/utils/adt/ri_triggers.c
RI_FKey_check(), RI_FKey_noaction_upd(), RI_FKey_noaction_del(), etc..
checking the referential with SELECT FOR UPDATE.
After BEGIN TRANSACTION: the INSERT/DELETE/UPDATE
I see now the following message and couldn't start
postmaster.
FATAL 2: btree_insert_redo: uninitialized page
Is it a bug ?
Seems so. btree_insert_redo shouldn't see uninitialized pages
(only newroot and split ops add pages to index and they should
be redone before insert op).
Can you
I am running 7.0 and for columns that have type 'timestamp'
the values end up with the format of year-month-day HH:MM:SS-[0-n]
afaik there is a newer JDBC driver which copes with this.
- Thomas
Hello,
what this can be?
FATAL: s_lock(40015071) at spin.c:127, stuck spinlock. Aborting.
From other sources I can find out that there was real memory starvation. All
swap was eated out (that's not PostgreSQL problem).
--
Sincerely Yours,
Denis Perchine
--
Edmar Wiggers wrote:
It seems that R-trees become inefficient when the number of dimensions
increase. Has anyone thoght of a transparent way to use Peano codes (hhcode
in Oracle lingo), and use B-tree indexes instead?
Do you have a reference, or more information on what a Peano code is?
38 matches
Mail list logo
