Re: [GENERAL] table locks

d enough to put adjectives in proper order ;) , but what I need is advisory [table] blocking transaction lock. Table-level is Ok for me while general form is probably better. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 6: explain analyze is your friend

Re: [GENERAL] table locks

e locks usage (briefly I need this thing to provide transaction isolation). Story is long and sad indeed, though I can retell it if it seems to be useful. Thanks a lot for your response. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP

Re: [GENERAL] table locks

u want to lock at all? It's long and sad story ;( -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq

[GENERAL] table locks

ntrib module? Thanks a lot. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 1: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through t

Re: [GENERAL] Discover temporary INDEX/TABLE name

Hello! >How could I find out if a temporary table >(or index on a temporary table) was created >by current session? May be the better question to ask is how one can find out the temporary schema name associated with the session. -- Best regards Ilja Golshtein --

[GENERAL] Discover temporary INDEX/TABLE name

to make a distinguish between temporary things belong to current session and others. I really do appreciate any help. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ?

[GENERAL] Performance: fsync vs. commit delay

ay PG server crashes are harmful as well. Is my understanding correct? What I want is to do write always but do fsync, say, every 100 ms. Is it possible to control fsyncs and writes separately, e.g. have write for every commit and fsync as often as specified time range? -- Best regards Ilja Gols

Re: [GENERAL] COPY FROM STDIN instead of INSERT

t. Since it would >allow transactions that insert data to finish faster, it should >actually work better under high concurrency. Do hope that's right. Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 3: H

Re: [GENERAL] COPY FROM STDIN instead of INSERT

Correct interpretation? ;) >Binary may be slightly faster because the datum parsing can be >partially skipped, but that's hardly much benefit over a text copy. I know example where it is up to three times faster. It depends on data. -- Best regards Ilja Golshtein -

Re: [GENERAL] COPY FROM STDIN instead of INSERT

que for some maintenance operations. My current goal is to increase performance in normal [almost ;)] OLTP mode of my application, so removing indexes for some time is not an option here. And my question remains. Is it Ok to use COPY BINARY FROM STDIN instead of multiple INSERTs? -- Bes

Re: [GENERAL] COPY FROM STDIN instead of INSERT

p too. I am working with existing database, though I am interested what "other things" you mean. Many thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster

[GENERAL] COPY FROM STDIN instead of INSERT

are highly appreciated. The only limitaion mentioned in Manual is about Rules and I don't care about this since I don't use Rules. Am I going to come across with any other problems (concurrency, reliability, compatibility, whatever) on this way? Many thanks. -- Best re

[GENERAL] apostrophes and psql variables

I was able to make up but no any luck. Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 1: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your messa

[GENERAL] xmin

, reliability, compartibility, whatever, point of view)? Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq

[GENERAL] pgstat.c: send/EINTR issue

erous case? -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 6: explain analyze is your friend

[GENERAL] indexes for ILIKE

Hello! Is there any way to use an index to pick up text field start from a substring? I need to ignore case of characters and encoding I use is UTF-8. I am aware of text_pattern_ops and stuff, though FAQ says ILIKE does not use indexes. Thanks. -- Best regards Ilja Golshtein

Re: [GENERAL] table is not a table

Hi! >On Wed, Jan 25, 2006 at 11:26:39AM -0500, Tom Lane wrote: >> "Ilja Golshtein" <[EMAIL PROTECTED]> writes: >> > postgres=# create table ddd(f1 int4); >> > CREATE TABLE >> > postgres=# drop table ddd; >> > ERROR: "ddd"

[GENERAL] table is not a table

c about 'ddd' - same thing for every table. I can ALTER this 'ddd', I can do any DML, but cannot drop! Any ideas what's wrong with permissions/roles or something? Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)

[GENERAL] Hang investigation

anual could be used in my case? I can produce log or all PG queries, but [as far as I understand] it is not enough to replay at least because PG does not put in log values of bound parameters. Many thanks. -- Best regards Ilja Golshtein ---(end of broa

Re: [GENERAL] SELECT FOR SHARE

>"Ilja Golshtein" <[EMAIL PROTECTED]> writes: >> I'm interested in SELECT ... FOR SHARE 8.1 feature. I'd tried to >> measure performance degradation and got something about 30-60% for >> queries retrieve a lot of rows. > >Degradation rel

[GENERAL] SELECT FOR SHARE and FOR UPDATE

Hello! I want to select data from two tables obtaining exclusive lock for records of the first table and nonexclusive lock for records of the second one. In other words, I need something like select a.f, b.f from a,b for update of a for share of b. Any hints? Thanks. -- Best regards Ilja

[GENERAL] SELECT FOR SHARE

, whatever) significantly? Is any sort of lock escalation [and other unpreditable things] involved here? Last question. What about ISOLATION LEVEL REPEATABLE READ syntax since non-exclusive locking for all SELECTs means [as far as I understand] nothing but REPEATABLE READ? Thanks. -- Best regards Ilja

Re: [GENERAL] CREATE TEMP TABLE AS SELECT/ GET DIAGNOSTICS ROW_COUNT again

Hi! >This change relates only to SPI, meaning plpsql. It does not help libpq. Thanks for response. I see. Any hope it would be possible to calculate number of rows inserted in newly created table via CREATE .. AS SELECT? -- Best regards Ilja Golshtein ---(end

[GENERAL] CREATE TEMP TABLE AS SELECT/ GET DIAGNOSTICS ROW_COUNT again

. Thanks a lot. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org

[GENERAL] COPY BINARY

column's data type (typically these functions are found in the src/backend/utils/adt/ directory of the source distribution). == I use libpq. And question number 0: how big perfomance difference between TEXT and BINARY formats ? Thanks. -- Best regards Ilja Gols

Re: [GENERAL] ctid access is slow

in analyze output. I played with fresh database. server version is 8.0.3, enable_tidscan is on, looks like hints have no effect. Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 6: explain analyze is your friend

Re: [GENERAL] ctid access is slow

Hello! >On Tue, Aug 23, 2005 at 09:15:42AM -0400, Robert Treat wrote: >> On Tuesday 23 August 2005 08:39, Ilja Golshtein wrote: >> > >> > select ctid from aaa where ctid in (select ctid from aaa limit 10); >Aside from that, ctid is of type tid, and its equality ope

[GENERAL] ctid access is slow

delete and use this info in DELETE statement (and something similar with UPDATE), but performance is absolutely unacceptable. Help is very very appeciated! -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 1: if posting/reading through U

Re: [GENERAL] Old question - failed to find conversion function from

ontexts where a >datatype can be assigned to it immediately. Honestly I cannot say this info changes too much for me (other DBMSs I use as backend handle this untyped NULLs), though the situation is clarified. Thanks a lot. -- Best regards Ilja Golshtein

Re: [GENERAL] Old question - failed to find conversion function from "unknown"

ing error. As far as I understand, it is not about comparision only, but about any operation. I really believe NULLs are special here. Of course I am a complete stranger and unaware of PostgreSQL internals so what I am saying may contradict with some basical concepts. Looks like it does since acco

Re: [GENERAL] Old question - failed to find conversion function from "unknown"

ed with anything with predictable result and no additional information about types is necessary. Is it correct vision? I agree it's hard to proceed your query with UNION and some sort of error is reasonable here. -- Best regards Ilja Golshtein ---(end of broadcast)-

[GENERAL] Old question - failed to find conversion function from "unknown"

stgreSQL should have all information required. Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 6: explain analyze is your friend

Re: [GENERAL] failed to find conversion function from "unknown" to ...

between this situations and don't have clear picture where (and what is the reason behind this) I should explicitly cast literals to text or varchar and where it is not required. Is this problem of string literals only or some other types are involved? Thanks. -- Best regards Ilja Golsht

[GENERAL] (2) failed to find conversion function from "unknown" to ...

et ERROR: failed to find conversion function from "unknown" to integer. Is it desired behavior or subject to change in future versions of PG? Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 7: don't forget to

[GENERAL] failed to find conversion function from "unknown" to ...

abase, IOW type is known. Is what I am describing desired behavior? Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 7: don't forget to increase your free space map settings

Re: [GENERAL] Hungry postmaster

his item in postgresql.conf a value, it will kill >long running queries for you. Thanks. Not perfect solution, but much better then nothing. To be honest, I don't quite understand why PG does 'send' again and again, while EPIPE seems to be unrecoverable. Isn't it better

Re: [GENERAL] Hungry postmaster

and killing such zombi-sessions? I mean, may be script or something to do this job exist in nature? Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org

Re: [GENERAL] DATATYPE for HEX

data type specifies a default strategy for columns of that data type". How one can find out these default strategies? Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster

[GENERAL] Hungry postmaster

ssue and/or prevent such thing in the future? Pg - 8.0.0 Kernel - 2.4.20 -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PRO

[GENERAL] DATATYPE for HEX

? Thanks a lot. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster

[GENERAL] slow parsing of queries with joins

some related configuration parameters. Setting join_collapse_limit as 1 doubles speed of INNER JOIN processing, though it is still incomparable with, say, MS SQL. This issue is crucial for for me since I use a tool that generates queries with many INNER JOIN clauses and it c

Re: [GENERAL] CREATE TEMP TABLE AS SELECT/ GET DIAGNOSTICS ROW_COUNT

Hi! >Done. Here is the patch (against CVS tip, but it should apply with >some fuzz in 8.0 or 7.4). Is this patch about CREATE TEMP TABLE AS SELECT only, or about SELECT INTO TEMP TABLE as well? -- Best regards Ilja Golshtein ---(end of bro

[GENERAL] CHAR LIKE VARCHAR

ay be wrong. Is there a clear concept behind the difference between "LIKE" and "="? Thanks. -- Best regards Ilja Golshtein ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate sub