Re: [HACKERS] max_prepared_transactions default ... why 5?
I'm writing up the new GUCs, and noticed that max_prepared_transactions defaults to 5. This is too many for most applications (which don't use them at all) and far too few for applications which use them regularly. I think the intention was to have enough so you could test 'em (in particular, run the regression tests) without eating resources for the majority of installations that aren't using them. Certainly an installation that *is* using 'em would want a higher setting. Can' we make the default 0, which is what the majority should want, and have the regression test explicitly set it up on the commandline? /Magnus ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] Why copy_relation_data only use wal when WALarchiving is enabled
You need to set $PGDATA before running the script. And psql,pg_ctl and pg_resetxlog need to be in $PATH. After running the script, restart postmaster and run SELECT * FROM t2. There should be one row in the table, but it's empty. I've tried this script, and superisingly found that T2 is not empty, just as it should be. Then I see how cluster is done, and found that ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Heikki Linnakangas [EMAIL PROTECTED] writes: I tend to agree that truncating the file, and extending the fsync request mechanism to actually delete it after the next checkpoint, is the most reasonable route to a fix. How about just allowing to use wal even WAL archiving is disabled? It seems that recovery of XLOG_HEAP_NEWPAGE record will do the right thing for us, look at heap_xlog_newpage: XLogReadBuffer with init=true will extend the block rightly and rebuild it rightly. Someone may say that it's not worth recording xlog for operations such as copy_relation_data, but these operations shouldn't happen frequently. ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] Why copy_relation_data only use wal when WALarchiving is enabled
Sorry, send the mail wrongly just now. You need to set $PGDATA before running the script. And psql,pg_ctl and pg_resetxlog need to be in $PATH. After running the script, restart postmaster and run SELECT * FROM t2. There should be one row in the table, but it's empty. I've tried this script on postgres (PostgreSQL) 8.3devel, and found that T2 is not empty after recovery(just as it should be)---but the latest version act just like what you said. Then I see how cluster is done, and found that: In postgres (PostgreSQL) 8.3devel, unlike AlterTableSetTablespace (which copys the whole relation block-by-block, and doesn't use wal under non-archiving mode), Cluster copys the relation row-by-row(simple_heap_insert), which always uses wal regardless of archiving mode. As wal exists, recovery will cope with everything rightly. The latest version acts differently probably because that it removes wal of cluser under non-archiving mode. So the conclusion is: we can replace wal mechanism with smgrimmedsync only if relfilenode is not allowed to be reused, but this's not true, so what we should keep wal. Is it right? ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] Why copy_relation_data only use wal when WALarchiving is enabled
You need to set $PGDATA before running the script. And psql,pg_ctl and pg_resetxlog need to be in $PATH. After running the script, restart postmaster and run SELECT * FROM t2. There should be one row in the table, but it's empty. I've tried this script on postgres (PostgreSQL) 8.3devel, and found that T2 is not empty after recovery(just as it should be)---but the latest version act just like what you said. Then I see how cluster is done, and found that: In postgres (PostgreSQL) 8.3devel, unlike AlterTableSetTablespace (which copys the whole relation block-by-block, and doesn't use wal under non-archiving mode), Cluster copys the relation row-by-row(simple_heap_insert), which always uses wal regardless of archiving mode. As wal exists, recovery will cope with everything rightly. The latest version acts differently probably because that it removes wal of cluser under non-archiving mode. So the conclusion is: we can replace wal mechanism with smgrimmedsync only if relfilenode is not allowed to be reused, but this's not true, so what we should keep wal. Is it right? ---(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 to the mailing list cleanly
Re: [HACKERS] Why copy_relation_data only use wal when WALarchiving is enabled
You need to set $PGDATA before running the script. And psql,pg_ctl and pg_resetxlog need to be in $PATH. After running the script, restart postmaster and run SELECT * FROM t2. There should be one row in the table, but it's empty. I've tried this script on postgres (PostgreSQL) 8.3devel, and found that T2 is not empty after recovery(just as it should be)---but the latest version act just like what you said. Then I see how cluster is done, and found that: In postgres (PostgreSQL) 8.3devel, unlike AlterTableSetTablespace (which copys the whole relation block-by-block, and doesn't use wal under non-archiving mode), Cluster copys the relation row-by-row(simple_heap_insert), which always uses wal regardless of archiving mode. As wal exists, recovery will cope with everything rightly. The latest version acts differently probably because that it removes wal of cluser under non-archiving mode. So the conclusion is: we can replace wal mechanism with smgrimmedsync only if relfilenode is not allowed to be reused, but this's not true, so what we should keep wal. Is it right? ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Jacky Leng wrote: I tend to agree that truncating the file, and extending the fsync request mechanism to actually delete it after the next checkpoint, is the most reasonable route to a fix. How about just allowing to use wal even WAL archiving is disabled? It seems that recovery of XLOG_HEAP_NEWPAGE record will do the right thing for us, look at heap_xlog_newpage: XLogReadBuffer with init=true will extend the block rightly and rebuild it rightly. Someone may say that it's not worth recording xlog for operations such as copy_relation_data, but these operations shouldn't happen frequently. Always using WAL would fix the problem, but it's a big performance hit. WAL-logging doubles the amount of write I/O required. -- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] ts_rewrite aggregate API seems mighty ugly
Tom Lane [EMAIL PROTECTED] writes: Since we're already committed to an initdb for beta2, it's not quite too late to reconsider the API here. My feeling at the moment is that we should just drop the aggregate form of ts_rewrite; it does nothing you can't do better with the two-argument form, and it is just ugly to boot. Also, if anyone does come up with a not-so-ugly design later, we can always add things in 8.4 or beyond; but once it's in core it's going to be a very hard sell to take it out. The two-argument form may not be actively broken but it sounds not very integrated. Passing a string which is then planned as an SQL query is not very SQL-ish. If you prepare the query does that string get prepared? When do you get syntax errors? If the plan's invalidated does it handle it correctly? It sounds to me like once we have a good aggregate interface there's not much point to the two-argument form. A good aggregate form would let you do the subquery or a join or any other form of query to fetch all the replacements you want. I would suggest putting both the two-argument form and the aggregate form into a contrib module -- effectively *back* into a contrib module until the interfaces can be fully integrated into SQL. So whereas previously you installed a contrib module for the whole tsearch2, now it's all integrated except for a few oddball functions which you can still get by supplementing the core tsearch with the contrib module. -- Gregory Stark EnterpriseDB http://www.enterprisedb.com ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Heikki Linnakangas wrote: Tom Lane wrote: I tend to agree that truncating the file, and extending the fsync request mechanism to actually delete it after the next checkpoint, is the most reasonable route to a fix. Ok, I'll write a patch to do that. There's a small problem with that: DROP TABLESPACE checks that the tablespace directory is empty, and fails if it sees one of those empty files. You also run into that problem if you 1. BEGIN; CREATE TABLE; -- no commit 2. crash+restart 3. DROP TABLESPACE because we leave behind the stale file created by CREATE TABLE. The best I can think of is to rename the obsolete file to relfilenode.stale, when it's scheduled for deletion at next checkpoint, and check for .stale-suffixed files in GetNewRelFileNode, and delete them immediately in DropTableSpace. That still won't fix the problem with files created by a crashed transaction. For that we had a plan a long time ago: after recovery, scan the data directory for any files don't have a live row in pg_class, and write a message to log for each so that the DBA can delete them (deleting them automatically was considered too dangerous). That's probably 8.4 material, though. Thoughts? -- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] Proposal: generate_iterator functions
On 10/18/07, Pavel Stehule [EMAIL PROTECTED] wrote: this function can help with array's iteration. create function generate_iterator(anyarray) returns setof integer as $$ select i from generate_series(array_lower($1,1), array_upper($1,1)) g(i) $$ language sql; There was a very similar proposal a little while back (google: array_to_set). I think I like those names better since you are returning a set, not an iterator :-). Also, this should be internal as you suggest (there is an undocumented builtin that already does this, _pg_expandarray). merlin ---(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 to the mailing list cleanly
Re: [HACKERS] Proposal: generate_iterator functions
2007/10/18, Merlin Moncure [EMAIL PROTECTED]: On 10/18/07, Pavel Stehule [EMAIL PROTECTED] wrote: this function can help with array's iteration. create function generate_iterator(anyarray) returns setof integer as $$ select i from generate_series(array_lower($1,1), array_upper($1,1)) g(i) $$ language sql; There was a very similar proposal a little while back (google: array_to_set). I think I like those names better since you are returning a set, not an iterator :-). Also, this should be internal as you suggest (there is an undocumented builtin that already does this, _pg_expandarray). I remember. There is only one important difference. What is behave of array_to_set with multidim. array? the name generate_iterator is my first idea (it retunrs set but setof indexes). I am sure so there are better names :) Pavel ---(end of broadcast)--- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] Proposal: generate_iterator functions
2007/10/18, Merlin Moncure [EMAIL PROTECTED]: On 10/18/07, Pavel Stehule [EMAIL PROTECTED] wrote: this function can help with array's iteration. create function generate_iterator(anyarray) returns setof integer as $$ select i from generate_series(array_lower($1,1), array_upper($1,1)) g(i) $$ language sql; There was a very similar proposal a little while back (google: array_to_set). I think I like those names better since you are returning a set, not an iterator :-). Also, this should be internal as you suggest (there is an undocumented builtin that already does this, _pg_expandarray). merlin one sample: create or replace function array_unpack2(anyarray) returns setof anyelement as $$ select $1[i][j] from generate_iterator($1,1) i, generate_iterator($1,2) j$$ language sql; postgres=# select array_unpack2(ARRAY[[10,11,12],[13,14,15]]); array_unpack2 --- 10 11 12 13 14 15 (6 rows) ---(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 to the mailing list cleanly
Re: [HACKERS] max_prepared_transactions default ... why 5?
* Magnus Hagander ([EMAIL PROTECTED]) wrote: Certainly an installation that *is* using 'em would want a higher setting. Can' we make the default 0, which is what the majority should want, and have the regression test explicitly set it up on the commandline? I'm with Magnus on this one. Having it set to just enough to test with is just plain terrible. A value of '5' seems high enough that it might be fine while on a development system and maybe for a short time on a production system which means it might not get noticed early enough to get picked up on and fixed by an admin before a user notices. I also like the on/off option, with the default being 'off' (where that really means *off*, not on, but with a small number- that would be even worse than the current situation). Thanks, Stephen signature.asc Description: Digital signature
Re: [HACKERS] max_prepared_transactions default ... why 5?
Magnus Hagander [EMAIL PROTECTED] writes: Can' we make the default 0, which is what the majority should want, and have the regression test explicitly set it up on the commandline? No. It's a postmaster-start-time-only option, which means that your proposal breaks make installcheck. regards, tom lane ---(end of broadcast)--- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] max_prepared_transactions default ... why 5?
Can' we make the default 0, which is what the majority should want, and have the regression test explicitly set it up on the commandline? No. It's a postmaster-start-time-only option, which means that your proposal breaks make installcheck. Bummer. There are lots of ways to break installcheck though - locale being one I get biten by all the time... But I guess this one would break the default on US systems... /Magnus ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] ts_rewrite aggregate API seems mighty ugly
Gregory Stark [EMAIL PROTECTED] writes: The two-argument form may not be actively broken but it sounds not very integrated. Passing a string which is then planned as an SQL query is not very SQL-ish. True. I'll bet you don't like ts_stat() either. regards, tom lane ---(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 to the mailing list cleanly
[HACKERS] upgrade from 8.0.3 to 8.1.10 crash
Hello. I am trying to upgrade from PostgreSQL 8.0.3 to PostgreSQL
8.1.10. I'm on WindowsXP and I'm compiling with Visual C++ 6.0.
I have a very simple routine that works fine with the 8.0.3 version:
int easy_connect()
{
exec sql connect to my_db as my_cnxtn;
printf (connection results:\n);
printf (code: %d\n, sqlca.sqlcode);
printf (state: %s\n, sqlca.sqlstate);
printf (warn: %s\n, sqlca.sqlwarn);
exec sql begin work;
exec sql commit work;
return 0;
}
When I use the 8.1.10 version, the connection works ( sqlca.sqlcode is
0, sqlca.sqlstate is and sqlca.sqlwarn is blank). But it then
crashes on the exec sql begin work line.
I get the following error in a window:
Runtime Error!
Program: C:\testSQL.exe
This application has requested the Runtime to terminate it in an unusual
way.
Please contact the application's support team for more information.
I removed the 8.0.3 version using remove program from the control
panel. I then went and removed my C:\Program Files\PostgreSQL\8.0
folder, just to make sure I was not including the wrong header files,
linking to the wrong libraries, or calling the wrong version of
ecpg.exe.
Thanks for any help.
-Cheryl
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Heikki Linnakangas [EMAIL PROTECTED] writes: The best I can think of is to rename the obsolete file to relfilenode.stale, when it's scheduled for deletion at next checkpoint, and check for .stale-suffixed files in GetNewRelFileNode, and delete them immediately in DropTableSpace. This is getting too Rube Goldbergian for my tastes. What if we just make DROP TABLESPACE force a checkpoint before proceeding? regards, tom lane ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] Proposal: generate_iterator functions
Merlin Moncure [EMAIL PROTECTED] writes: There was a very similar proposal a little while back (google: array_to_set). I think I like those names better since you are returning a set, not an iterator :-). I agree, this is a very poor choice of name. There should be some reference to arrays in it, for one thing. generate_array_subscripts() maybe? regards, tom lane ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] Proposal: generate_iterator functions
2007/10/18, Tom Lane [EMAIL PROTECTED]: Merlin Moncure [EMAIL PROTECTED] writes: There was a very similar proposal a little while back (google: array_to_set). I think I like those names better since you are returning a set, not an iterator :-). I agree, this is a very poor choice of name. There should be some reference to arrays in it, for one thing. generate_array_subscripts() maybe? why not? Pavel ---(end of broadcast)--- TIP 6: explain analyze is your friend
[HACKERS] dblink un-named connection doesn't get re-used
Is it intentional that dblink's unnamed connections don't get re-used?
stats=# select datname, usename from pg_stat_activity;
datname | usename
-+-
stats | decibel
(1 row)
stats=# select dblink_connect('dbname=stats');
dblink_connect
OK
(1 row)
stats=# select dblink_connect('dbname=postgres');
dblink_connect
OK
(1 row)
stats=# select datname, usename from pg_stat_activity;
datname | usename
--+--
stats| decibel
stats| postgres
postgres | postgres
(3 rows)
AFAIK there's no way I could possibly use or refer to the connection
to stats at this point; so why doesn't dblink close it when I issue
the second connect?
--
Decibel!, aka Jim C. Nasby, Database Architect [EMAIL PROTECTED]
Give your computer some brain candy! www.distributed.net Team #1828
---(end of broadcast)---
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Tom Lane wrote: Heikki Linnakangas [EMAIL PROTECTED] writes: The best I can think of is to rename the obsolete file to relfilenode.stale, when it's scheduled for deletion at next checkpoint, and check for .stale-suffixed files in GetNewRelFileNode, and delete them immediately in DropTableSpace. This is getting too Rube Goldbergian for my tastes. What if we just make DROP TABLESPACE force a checkpoint before proceeding? True, that would work. DROP TABLESPACE should be uncommon enough that the performance hit is ok. We only need to checkpoint if the directory isn't empty, though I think that's the case more often than not; you're most likely to drop a tablespace right after dropping all relations in it. -- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Heikki Linnakangas wrote: Tom Lane wrote: I tend to agree that truncating the file, and extending the fsync request mechanism to actually delete it after the next checkpoint, is the most reasonable route to a fix. Ok, I'll write a patch to do that. What is the argument against making relfilenodes globally unique by adding the xid and epoch of the creating transaction to the filename? Those 64 bits could be stuffed into 13 bytes by base-36 encoding (A-Z,0-9). The maximum length of a relfilenode would then be 10 + 1 + 13 = 24, which any reasonable filesystem should support IMHO. regards, Florian Pflug ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Heikki Linnakangas wrote: Tom Lane wrote: I tend to agree that truncating the file, and extending the fsync request mechanism to actually delete it after the next checkpoint, is the most reasonable route to a fix. Ok, I'll write a patch to do that. What is the argument against making relfilenodes globally unique by adding the xid and epoch of the creating transaction to the filename? Those 64 bits could be stuffed into 13 bytes by base-36 encoding (A-Z,0-9). The maximum length of a relfilenode would then be 10 + 1 + 13 = 24, which any reasonable filesystem should support IMHO. regards, Florian Pflug PS: Sorry if this arrives twice - I'm having a few troubles with my mail setup. ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] dblink un-named connection doesn't get re-used
Sorry for the self-reply... On Oct 18, 2007, at 9:09 AM, Decibel! wrote: Is it intentional that dblink's unnamed connections don't get re-used? From the dblink docs (both 8.1 and HEAD): if only one argument is given, the connection is unnamed; only one unnamed connection can exist at a time So this sounds to me like a bug. -- Decibel!, aka Jim C. Nasby, Database Architect [EMAIL PROTECTED] Give your computer some brain candy! www.distributed.net Team #1828 ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Florian G. Pflug wrote: Heikki Linnakangas wrote: Tom Lane wrote: I tend to agree that truncating the file, and extending the fsync request mechanism to actually delete it after the next checkpoint, is the most reasonable route to a fix. Ok, I'll write a patch to do that. What is the argument against making relfilenodes globally unique by adding the xid and epoch of the creating transaction to the filename? Those 64 bits could be stuffed into 13 bytes by base-36 encoding (A-Z,0-9). The maximum length of a relfilenode would then be 10 + 1 + 13 = 24, which any reasonable filesystem should support IMHO. The size of would be xid + epoch + oid = 96 bits, not 64 bits. That would work, but sounds like a much bigger change. sizeof(RelFileNode) would increase from 12 to 20, so any data structure that deals with RelFileNodes would take more memory. Hash function in buffer manager would get more expensive. I remember seeing that showing up in oprofile sometimes, but it'd need to be benchmarked to see if it really matters. -- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Florian G. Pflug [EMAIL PROTECTED] writes: What is the argument against making relfilenodes globally unique by adding the xid and epoch of the creating transaction to the filename? 1. Zero chance of ever backpatching. (I know I said I wasn't excited about that, but it's still a strike against a proposed fix.) 2. Adds new fields to RelFileNode, which will be a major code change, and possibly a noticeable performance hit (bigger hashtable keys). 3. Adds new columns to pg_class, which is a real PITA ... 4. Breaks oid2name and all similar code that knows about relfilenode. regards, tom lane ---(end of broadcast)--- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] max_prepared_transactions default ... why 5?
Magnus Hagander [EMAIL PROTECTED] writes: Bummer. There are lots of ways to break installcheck though - locale being one I get biten by all the time... Hmm, which locale do you use and what breakage do you see? regards, tom lane ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] Strange error dropping foreign key
Magnus Hagander [EMAIL PROTECTED] writes: db=# alter table isi.items_stat drop constraint items_stat_item_id_fkey; ERROR: items_pkey is an index Context please? regards, tom lane ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
[HACKERS] Strange error dropping foreign key
db=# alter table isi.items_stat drop constraint items_stat_item_id_fkey; ERROR: items_pkey is an index The foreign key points to items.Item_id which is what's indexed by items_pkey. But I only wanted to drop that constraint. /Magnus ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
[HACKERS] Can a C function(server program) be a UDP or TCP server?
Hi there, Is it possible to write a dynamic loaded C function as an UDP or TCP server? What we want to do it is: Add a search function which send a UDP package to remote UDP server and then listen to an UDP port, waiting for the result. Ideally, we don't close the UDP server after the search query end. So we can reuse it for next search. Is it possible? Thanks Billow
Re: [HACKERS] Can a C function(server program) be a UDP or TCP server?
On Thu, 18 Oct 2007 10:55:19 -0400 Billow Gao [EMAIL PROTECTED] wrote: Is it possible to write a dynamic loaded C function as an UDP or TCP server? What we want to do it is: Add a search function which send a UDP package to remote UDP server and then listen to an UDP port, waiting for the result. Ideally, we don't close the UDP server after the search query end. So we can reuse it for next search. Is it possible? Short answer: yes. Slightly longer answer: If you need to ask this quetion then you should really talk to someone about network programming but this is the wrong list. If you are asking if PostgreSQL already does UDP then the answer is no. You need to write a server program that talks UDP in one direction and TCP to PostgreSQL in the other direction. Watch out for security issues. -- D'Arcy J.M. Cain [EMAIL PROTECTED] | Democracy is three wolves http://www.druid.net/darcy/| and a sheep voting on +1 416 425 1212 (DoD#0082)(eNTP) | what's for dinner. ---(end of broadcast)--- TIP 4: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] Can a C function(server program) be a UDP or TCP server?
On Thu, 18 Oct 2007 10:55:19 -0400 Billow Gao [EMAIL PROTECTED] wrote: Is it possible to write a dynamic loaded C function as an UDP or TCP server? What we want to do it is: Add a search function which send a UDP package to remote UDP server and then listen to an UDP port, waiting for the result. Ideally, we don't close the UDP server after the search query end. So we can reuse it for next search. Is it possible? Short answer: yes. Slightly longer answer: If you need to ask this quetion then you should really talk to someone about network programming but this is the wrong list. Thanks for your reply. I can write the network program. But I am not 100% sure whether I can add the c-language function ( http://www.postgresql.org/docs/8.2/interactive/xfunc-c.html) to PostgreSQL. The function will be dynamic loaded by PostgreSQL. I want to know whether there are any limitation on the function I wrote. for example: If I want to write a function: PG_FUNCTION_INFO_V1(c_talktoremoteudp); And use it in PostgreSQL like: = SELECT name, c_talktoremoteudp (emp, 1500) AS overpaid FROM emp WHERE name = 'Bill' OR name = 'Sam'; = The function c_talktoremoteudp will: 1. send udp data to remote udp server 2. monitor an udp port and wait for the reply 3. return the data to the select query. Anyway, I guess that it can be done. I will write a simple function to test it. Ying If you are asking if PostgreSQL already does UDP then the answer is no. You need to write a server program that talks UDP in one direction and TCP to PostgreSQL in the other direction. Watch out for security issues.
Re: [HACKERS] Can a C function(server program) be a UDP or TCP server?
On Thu, 18 Oct 2007 11:24:24 -0400 Billow Gao [EMAIL PROTECTED] wrote: I can write the network program. But I am not 100% sure whether I can add the c-language function ( http://www.postgresql.org/docs/8.2/interactive/xfunc-c.html) to PostgreSQL. The function will be dynamic loaded by PostgreSQL. I want to know whether there are any limitation on the function I wrote. for example: If I want to write a function: PG_FUNCTION_INFO_V1(c_talktoremoteudp); And use it in PostgreSQL like: = SELECT name, c_talktoremoteudp (emp, 1500) AS overpaid FROM emp WHERE name = 'Bill' OR name = 'Sam'; = The function c_talktoremoteudp will: 1. send udp data to remote udp server 2. monitor an udp port and wait for the reply 3. return the data to the select query. I am confused. The dynamic function resides in the server. The query runs in the server. Where is the remoteness in any of this? Are you saying that there is a second server that is not PostgreSQL that uses UDP that you want to communicate with and merge info into the PostgreSQL server from? -- D'Arcy J.M. Cain [EMAIL PROTECTED] | Democracy is three wolves http://www.druid.net/darcy/| and a sheep voting on +1 416 425 1212 (DoD#0082)(eNTP) | what's for dinner. ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] max_prepared_transactions default ... why 5?
On Oct 18, 2007, at 12:07 AM, Bruce Momjian wrote: Josh Berkus wrote: On Wednesday 17 October 2007 21:35, Tom Lane wrote: Josh Berkus [EMAIL PROTECTED] writes: I'm writing up the new GUCs, and noticed that max_prepared_transactions defaults to 5. This is too many for most applications (which don't use them at all) and far too few for applications which use them regularly. I think the intention was to have enough so you could test 'em (in particular, run the regression tests) without eating resources for the majority of installations that aren't using them. Certainly an installation that *is* using 'em would want a higher setting. Yeah, given the amount of memory per xact, I guess we can't actually set the default higher. I just hate to see a setting that is liable to bite someone on the tuchas so easily. They will see the failure at 5 faster and adjust it accordingly. If it was higher they might hit the limit only under heavy load and it would surprise them. Actually, the amount of memory is a reason to default to 0, or change the name, or put a big comment in the config, because I very often saw databases where people had set this to a very high value under the impression that it impacted prepared statements. -- Decibel!, aka Jim C. Nasby, Database Architect [EMAIL PROTECTED] Give your computer some brain candy! www.distributed.net Team #1828 ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] Why copy_relation_data only use wal whenWALarchivingis enabled
Tom Lane wrote: Florian G. Pflug [EMAIL PROTECTED] writes: What is the argument against making relfilenodes globally unique by adding the xid and epoch of the creating transaction to the filename? 1. Zero chance of ever backpatching. (I know I said I wasn't excited about that, but it's still a strike against a proposed fix.) 2. Adds new fields to RelFileNode, which will be a major code change, and possibly a noticeable performance hit (bigger hashtable keys). 3. Adds new columns to pg_class, which is a real PITA ... 4. Breaks oid2name and all similar code that knows about relfilenode. Ah, Ok. I was under the impression that relfilenode in pg_class is a string of some kind. In that case only GetNewRelFileNode would have needed patching... But that is obviously not the case, as I realized now :-( Thanks for setting me straight ;-) regards, Florian Pflug ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] ts_rewrite aggregate API seems mighty ugly
Tom Lane [EMAIL PROTECTED] writes: Gregory Stark [EMAIL PROTECTED] writes: The two-argument form may not be actively broken but it sounds not very integrated. Passing a string which is then planned as an SQL query is not very SQL-ish. True. I'll bet you don't like ts_stat() either. It seems the right way interface here wouldn't be too different from what's there. All we need is a SRF which takes a single tsvector and returns the set of words from it. Then you could do the aggregates yourself in SQL: SELECT count(distinct apodid) as ndoc, count(*) as nentry, element FROM ( SELECT apodid, ts_elements(vector) AS element FROM apod ) GROUP BY element -- Gregory Stark EnterpriseDB http://www.enterprisedb.com ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] Can a C function(server program) be a UDP or TCP server?
D'Arcy J.M. Cain [EMAIL PROTECTED] writes: On Thu, 18 Oct 2007 11:24:24 -0400 And use it in PostgreSQL like: = SELECT name, c_talktoremoteudp(emp, 1500) AS overpaid FROM emp WHERE name = 'Bill' OR name = 'Sam'; = The function c_talktoremoteudp will: 1. send udp data to remote udp server 2. monitor an udp port and wait for the reply 3. return the data to the select query. I am confused. The dynamic function resides in the server. The query runs in the server. Where is the remoteness in any of this? Are you saying that there is a second server that is not PostgreSQL that uses UDP that you want to communicate with and merge info into the PostgreSQL server from? Yeah, what he wants is to implement a function in Postgres which does something like an LDAP or DNS lookup or something like that. Sure you can do this. The only tricky bit is the thing you mentioned about reusing the connection. You could always leave the connection in a safe state and don't need to worry about cleaning it up then you could just store it in a static variable which would be the simplest option. If you want to use Postgres's facilities for allocating memory and cleaning it up when no longer in use you can use some of the Postgres internal API for memory contexts and resource owners. But I don't see any particular reason you should need to worry about this stuff for something simple you're implementing yourself. -- Gregory Stark EnterpriseDB http://www.enterprisedb.com ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] max_prepared_transactions default ... why 5?
Decibel! [EMAIL PROTECTED] writes: Actually, the amount of memory is a reason to default to 0, or change the name, or put a big comment in the config, because I very often saw databases where people had set this to a very high value under the impression that it impacted prepared statements. There's another cost associated with prepared transactions. If it's set to 0 then there's no real reason we need to wal log lock operations. -- Gregory Stark EnterpriseDB http://www.enterprisedb.com ---(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 to the mailing list cleanly
Re: [HACKERS] Can a C function(server program) be a UDP or TCP server?
Thanks. This is what I want to know :-) Regards, Billow Yeah, what he wants is to implement a function in Postgres which does something like an LDAP or DNS lookup or something like that. Sure you can do this. The only tricky bit is the thing you mentioned about reusing the connection. You could always leave the connection in a safe state and don't need to worry about cleaning it up then you could just store it in a static variable which would be the simplest option. If you want to use Postgres's facilities for allocating memory and cleaning it up when no longer in use you can use some of the Postgres internal API for memory contexts and resource owners. But I don't see any particular reason you should need to worry about this stuff for something simple you're implementing yourself. On 10/18/07, D'Arcy J.M. Cain [EMAIL PROTECTED] wrote: On Thu, 18 Oct 2007 11:24:24 -0400 Billow Gao [EMAIL PROTECTED] wrote: I can write the network program. But I am not 100% sure whether I can add the c-language function ( http://www.postgresql.org/docs/8.2/interactive/xfunc-c.html) to PostgreSQL. The function will be dynamic loaded by PostgreSQL. I want to know whether there are any limitation on the function I wrote. for example: If I want to write a function: PG_FUNCTION_INFO_V1(c_talktoremoteudp); And use it in PostgreSQL like: = SELECT name, c_talktoremoteudp (emp, 1500) AS overpaid FROM emp WHERE name = 'Bill' OR name = 'Sam'; = The function c_talktoremoteudp will: 1. send udp data to remote udp server 2. monitor an udp port and wait for the reply 3. return the data to the select query. I am confused. The dynamic function resides in the server. The query runs in the server. Where is the remoteness in any of this? Are you saying that there is a second server that is not PostgreSQL that uses UDP that you want to communicate with and merge info into the PostgreSQL server from? -- D'Arcy J.M. Cain [EMAIL PROTECTED] | Democracy is three wolves http://www.druid.net/darcy/| and a sheep voting on +1 416 425 1212 (DoD#0082)(eNTP) | what's for dinner. ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] max_prepared_transactions default ... why 5?
On Thu, Oct 18, 2007 at 11:23 AM, in message [EMAIL PROTECTED], Gregory Stark [EMAIL PROTECTED] wrote: If it's set to 0 then there's no real reason we need to wal log lock operations. Do we currently take advantage of that fact, or log them anyway? -Kevin ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] Can a C function(server program) be a UDP or TCP server?
On Thursday 18 October 2007 12:27:59 Billow Gao wrote: Thanks. This is what I want to know :-) Regards, Billow Yeah, what he wants is to implement a function in Postgres which does something like an LDAP or DNS lookup or something like that. Sure you can do this. The only tricky bit is the thing you mentioned about reusing the connection. You could always leave the connection in a safe state and don't need to worry about cleaning it up then you could just store it in a static variable which would be the simplest option. If you want to use Postgres's facilities for allocating memory and cleaning it up when no longer in use you can use some of the Postgres internal API for memory contexts and resource owners. But I don't see any particular reason you should need to worry about this stuff for something simple you're implementing yourself. On 10/18/07, D'Arcy J.M. Cain [EMAIL PROTECTED] wrote: On Thu, 18 Oct 2007 11:24:24 -0400 Billow Gao [EMAIL PROTECTED] wrote: I can write the network program. ... Oh my. The worst kind of top-poster: the kind that copies *your* reply from the bottom to the top and top-posts above that. Shudder... :) jan -- -- Jan de Visser [EMAIL PROTECTED] Baruk Khazad! Khazad ai-menu! -- ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] max_prepared_transactions default ... why 5?
Kevin Grittner wrote: On Thu, Oct 18, 2007 at 11:23 AM, in message [EMAIL PROTECTED], Gregory Stark [EMAIL PROTECTED] wrote: If it's set to 0 then there's no real reason we need to wal log lock operations. Do we currently take advantage of that fact, or log them anyway? No, we log them anyway. There's a few other reasons to WAL log lock operations, see comments in heap_lock_tuple: /* * XLOG stuff.You might think that we don't need an XLOG record because * there is no state change worth restoring after a crash.You would be * wrong however: we have just written either a TransactionId or a * MultiXactId that may never have been seen on disk before, and we need * to make sure that there are XLOG entries covering those ID numbers. * Else the same IDs might be re-used after a crash, which would be * disastrous if this page made it to disk before the crash. Essentially * we have to enforce the WAL log-before-data rule even in this case. * (Also, in a PITR log-shipping or 2PC environment, we have to have XLOG * entries for everything anyway.) */ There's also the risk of torn pages. We set the xmax and the XMAX_*_LOCK flag in heap_lock_tuple, and it's possible that only one of those changes is written to disk before crash. -- Heikki Linnakangas EnterpriseDB http://www.enterprisedb.com ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] Proposal: generate_iterator functions
On 10/18/07, Tom Lane [EMAIL PROTECTED] wrote: Merlin Moncure [EMAIL PROTECTED] writes: There was a very similar proposal a little while back (google: array_to_set). I think I like those names better since you are returning a set, not an iterator :-). I agree, this is a very poor choice of name. There should be some reference to arrays in it, for one thing. generate_array_subscripts() maybe? array_to_set or array_expand seem a little better imo (shorter, and symmetry with array_accum()), unless you want to differentiate between internal funcs (array_cat and the like) vs. user funcs. I would prefer a proper C implementation to a solution based around generate_series(). I'm doing a lot of C funcs lately and would be happy taking a stab at this... merlin ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] Proposal: generate_iterator functions
Merlin Moncure [EMAIL PROTECTED] writes: On 10/18/07, Tom Lane [EMAIL PROTECTED] wrote: generate_array_subscripts() maybe? array_to_set or array_expand seem a little better imo (shorter, and symmetry with array_accum()), unless you want to differentiate between internal funcs (array_cat and the like) vs. user funcs. I don't much like either of those, because they seem misleading: what I'd expect from a function named that way is that it returns the *elements* of the array, not their subscripts. Come to think of it, do we have a way of doing that directly? If you only care about accessing the array elements, it seems like dealing in the subscripts is just notational tedium. Perhaps there should be array_expand(anyarray) returns setof anyelement, in addition to the subscript generation function. On the question of being too long, I could live with generate_subscripts(). regards, tom lane ---(end of broadcast)--- TIP 6: explain analyze is your friend
Re: [HACKERS] upgrade from 8.0.3 to 8.1.10 crash
I found when I removed all comments, it worked fine. Any ideas?
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Boergesson,
Cheryl
Sent: Thursday, October 18, 2007 9:24 AM
To: pgsql-hackers@postgresql.org
Subject: [HACKERS] upgrade from 8.0.3 to 8.1.10 crash
Hello. I am trying to upgrade from PostgreSQL 8.0.3 to PostgreSQL
8.1.10. I'm on WindowsXP and I'm compiling with Visual C++ 6.0.
I have a very simple routine that works fine with the 8.0.3 version:
int easy_connect()
{
exec sql connect to my_db as my_cnxtn;
printf (connection results:\n);
printf (code: %d\n, sqlca.sqlcode);
printf (state: %s\n, sqlca.sqlstate);
printf (warn: %s\n, sqlca.sqlwarn);
exec sql begin work;
exec sql commit work;
return 0;
}
When I use the 8.1.10 version, the connection works ( sqlca.sqlcode is
0, sqlca.sqlstate is and sqlca.sqlwarn is blank). But it then
crashes on the exec sql begin work line.
I get the following error in a window:
Runtime Error!
Program: C:\testSQL.exe
This application has requested the Runtime to terminate it in an unusual
way.
Please contact the application's support team for more information.
I removed the 8.0.3 version using remove program from the control
panel. I then went and removed my C:\Program Files\PostgreSQL\8.0
folder, just to make sure I was not including the wrong header files,
linking to the wrong libraries, or calling the wrong version of
ecpg.exe.
Thanks for any help.
-Cheryl
Re: [HACKERS] Proposal: generate_iterator functions
On 10/18/07, Tom Lane [EMAIL PROTECTED] wrote: I don't much like either of those, because they seem misleading: what I'd expect from a function named that way is that it returns the *elements* of the array, not their subscripts. Come to think of it, do we have a way of doing that directly? If you only care about accessing the array elements, it seems like dealing in the subscripts is just notational tedium. Perhaps there should be array_expand(anyarray) returns setof anyelement, in addition to the subscript generation function. On the question of being too long, I could live with generate_subscripts(). how about array_iota? merlin ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] Proposal: generate_iterator functions
Merlin Moncure [EMAIL PROTECTED] writes: On 10/18/07, Tom Lane [EMAIL PROTECTED] wrote: On the question of being too long, I could live with generate_subscripts(). how about array_iota? I think a lot of people wouldn't get the reference. How about array_subscripts()? regards, tom lane ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
Re: [HACKERS] Proposal: generate_iterator functions
Tom Lane wrote: Merlin Moncure [EMAIL PROTECTED] writes: On 10/18/07, Tom Lane [EMAIL PROTECTED] wrote: generate_array_subscripts() maybe? array_to_set or array_expand seem a little better imo (shorter, and symmetry with array_accum()), unless you want to differentiate between internal funcs (array_cat and the like) vs. user funcs. I don't much like either of those, because they seem misleading: what I'd expect from a function named that way is that it returns the *elements* of the array, not their subscripts. Come to think of it, do we have a way of doing that directly? If you only care about accessing the array elements, it seems like dealing in the subscripts is just notational tedium. Perhaps there should be array_expand(anyarray) returns setof anyelement, in addition to the subscript generation function. I think what you're describing is the SQL2003 UNNEST feature. Joe ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] upgrade from 8.0.3 to 8.1.10 crash
Boergesson, Cheryl wrote: I found when I removed all comments, it worked fine. Any ideas? I suggest you add ECPG to the subject line so that the relevant developers notice your problem. -- Alvaro Herrera http://www.amazon.com/gp/registry/CTMLCN8V17R4 The Postgresql hackers have what I call a NASA space shot mentality. Quite refreshing in a world of weekend drag racer developers. (Scott Marlowe) ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
[HACKERS] ECPG crash - upgrade from 8.0.3 to 8.1.10
Hello. I am trying to upgrade from PostgreSQL 8.0.3 to PostgreSQL
8.1.10. I'm on WindowsXP and I'm compiling with Visual C++ 6.0.
I have a very simple routine that works fine with the 8.0.3 version:
int easy_connect()
{
exec sql connect to my_db as my_cnxtn;
printf (connection results:\n);
printf (code: %d\n, sqlca.sqlcode);
printf (state: %s\n, sqlca.sqlstate);
printf (warn: %s\n, sqlca.sqlwarn);
exec sql begin work;
exec sql commit work;
return 0;
}
When I use the 8.1.10 version, the connection works ( sqlca.sqlcode is
0, sqlca.sqlstate is and sqlca.sqlwarn is blank). But it then
crashes on the exec sql begin work line.
I get the following error in a window:
Runtime Error!
Program: C:\testSQL.exe
This application has requested the Runtime to terminate it in an unusual
way.
Please contact the application's support team for more information.
I removed the 8.0.3 version using remove program from the control
panel. I then went and removed my C:\Program Files\PostgreSQL\8.0
folder, just to make sure I was not including the wrong header files,
linking to the wrong libraries, or calling the wrong version of
ecpg.exe.
Oddly, when I removed all comments, it worked fine.
Thanks for any help.
-Cheryl
Re: [HACKERS] Release notes introductory text
On Thu, Oct 18, 2007 at 12:34 AM, in message [EMAIL PROTECTED], Bruce Momjian [EMAIL PROTECTED] wrote: This release represents a major leap forward for PostgreSQL by adding significant new functionality and performance enhancements. This was made possible by a growing community that has dramatically accelerated the pace of development. This release adds the follow major capabilities: If we want to have a short intro to promote the new release, this sounds good to me. -Kevin ---(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 to the mailing list cleanly
Re: [HACKERS] Proposal: generate_iterator functions
On 10/18/07, Tom Lane [EMAIL PROTECTED] wrote: Merlin Moncure [EMAIL PROTECTED] writes: On 10/18/07, Tom Lane [EMAIL PROTECTED] wrote: On the question of being too long, I could live with generate_subscripts(). how about array_iota? I think a lot of people wouldn't get the reference. How about array_subscripts()? works for me.. merlin ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
[HACKERS] I've discovered an error with the tcl pgmail function
Further to the thread Re: Send email from PostgreSQL, may I ? * From: Devrim GUNDUZ devrim ( at ) commandprompt ( dot ) com * To: Gerson Machado gersonamach ( at ) yahoo ( dot ) com ( dot ) br * Subject: Re: Send email from PostgreSQL, may I ? * Date: Fri, 27 Oct 2006 16:07:00 +0300 There's a suggested function in pltclu to send emails from the postgre database. I've worked with this and discovered that there's an error which causes the email to be sent with a blank subject and the subject to be included as part of the body. To overcome this, the following should be done: comment out or remove the two lines:- puts $ServerSocket To: $mailto;# To: puts $ServerSocket From: $mailfrom;# From: Regards, Quentin --- This mail sent through http://webmail.zoom.co.uk ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] I've discovered an error with the tcl pgmail function
[EMAIL PROTECTED] writes: There's a suggested function in pltclu to send emails from the postgre database. There's no such function anywhere in the core Postgres distribution, so I think you've reported this to the wrong place. regards, tom lane ---(end of broadcast)--- TIP 2: Don't 'kill -9' the postmaster
Re: [HACKERS] Proposal: generate_iterator functions
done http://www.pgsql.cz/index.php/Iter%C3%A1tor_pole I'll send patch later Pavel 2007/10/18, Merlin Moncure [EMAIL PROTECTED]: On 10/18/07, Tom Lane [EMAIL PROTECTED] wrote: Merlin Moncure [EMAIL PROTECTED] writes: On 10/18/07, Tom Lane [EMAIL PROTECTED] wrote: On the question of being too long, I could live with generate_subscripts(). how about array_iota? I think a lot of people wouldn't get the reference. How about array_subscripts()? works for me.. merlin ---(end of broadcast)--- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq
Re: [HACKERS] dblink un-named connection doesn't get re-used
Decibel! wrote:
Is it intentional that dblink's unnamed connections don't get re-used?
yes
stats=# select dblink_connect('dbname=stats');
dblink_connect
OK
(1 row)
stats=# select dblink_connect('dbname=postgres');
dblink_connect
OK
(1 row)
AFAIK there's no way I could possibly use or refer to the connection to
stats at this point; so why doesn't dblink close it when I issue the
second connect?
Why doesn't C free allocated memory automatically if you reassign a pointer?
No one has ever complained before, so I can't imagine that the resource
leak is much of an issue in real world cases. But if you don't like the
behavior, patches are gratefully accepted ;-).
Seriously though, I can change it for 8.3, but is it really worth
back-patching?
Joe
---(end of broadcast)---
TIP 6: explain analyze is your friend
Re: [HACKERS] [COMMITTERS] pgsql: Consistently indent release notes for prior releases.
[EMAIL PROTECTED] (Bruce Momjian) writes: Consistently indent release notes for prior releases. Bruce, if you don't revert that patch I will do it for you. Random changes to the release.sgml sections for old releases are an utter nightmare when it comes time to produce back-branch updates. regards, tom lane ---(end of broadcast)--- TIP 5: don't forget to increase your free space map settings
Re: [HACKERS] upgrade from 8.0.3 to 8.1.10 crash
Boergesson, Cheryl [EMAIL PROTECTED] writes: I found when I removed all comments, it worked fine. Any ideas? Oh? Considering that the fragments you've shown us have never had one single comment, that means that no one could possibly have offered you any useful advice. Please, if you would like help with this sort of problem, show a complete self-contained test case. Leaving stuff out is counterproductive, way too often. regards, tom lane ---(end of broadcast)--- TIP 7: You can help support the PostgreSQL project by donating at http://www.postgresql.org/about/donate
[HACKERS] Re: [COMMITTERS] pgsql: Consistently indent release notes for prior releases.
Tom Lane wrote: [EMAIL PROTECTED] (Bruce Momjian) writes: Consistently indent release notes for prior releases. Bruce, if you don't revert that patch I will do it for you. Random changes to the release.sgml sections for old releases are an utter nightmare when it comes time to produce back-branch updates. The problem is that we use those back branches as templates for new minor releases. I could have just done 8.2.X but I figured I should be consistent. I am not sure why it affects back branch updates. The way I create back branches is to modify release.sgml in HEAD and then copy that file to the back branches, and cutt off the top part of release.sgml for the appropriate branch. Also I also don't understand why you are threatening to revert it before even complaining about it to me. -- Bruce Momjian [EMAIL PROTECTED]http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. + ---(end of broadcast)--- TIP 6: explain analyze is your friend
