Re: [BUGS] PostgreSQL client has problems when libbind is installed
Was this problem fixed? Can I request the problem report @ FreeBSD to be closed? Thanks, Palle --On torsdag, juni 12, 2003 18.51.18 -0400 Yves R. Crevecoeur [EMAIL PROTECTED] wrote: Don't break BeOS support. A new version of BeOS will be released very soon. http://www.yellowtab.com http://www.yellowtab.com/board/ http://www.yellowtab.com/support/ Regards, Yves - Original Message - From: Tom Lane [EMAIL PROTECTED] To: Palle Girgensohn [EMAIL PROTECTED] Cc: [EMAIL PROTECTED]; Tom Alsberg [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Thursday, June 12, 2003 2:00 PM Subject: Re: [BUGS] PostgreSQL client has problems when libbind is installed Palle Girgensohn [EMAIL PROTECTED] writes: [ linking libbind causes some obscure problems ] http://www.freebsd.org/cgi/query-pr.cgi?pr=ports/47218 Given that we're not supporting BeOS at the moment anyway, I wonder whether we need libbind on any platform. I know linking it causes some minor problems for me on HPUX (I have to take it out if I want to build with profiling...) Comments anyone? regards, tom lane ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED]) ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faqs/FAQ.html ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED])
Re: [HACKERS] [BUGS] PostgreSQL client has problems when libbind is
Tom sent a mail indicating it was fixed. Thanks, Palle --On fredag, november 07, 2003 09.32.54 -0500 Bruce Momjian [EMAIL PROTECTED] wrote: I don't remember what we did with -lbind, but if someone shows up with BeOS, we will get it working. - -- Palle Girgensohn wrote: Was this problem fixed? Can I request the problem report @ FreeBSD to be closed? Thanks, Palle --On torsdag, juni 12, 2003 18.51.18 -0400 Yves R. Crevecoeur [EMAIL PROTECTED] wrote: Don't break BeOS support. A new version of BeOS will be released very soon. http://www.yellowtab.com http://www.yellowtab.com/board/ http://www.yellowtab.com/support/ Regards, Yves - Original Message - From: Tom Lane [EMAIL PROTECTED] To: Palle Girgensohn [EMAIL PROTECTED] Cc: [EMAIL PROTECTED]; Tom Alsberg [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Thursday, June 12, 2003 2:00 PM Subject: Re: [BUGS] PostgreSQL client has problems when libbind is installed Palle Girgensohn [EMAIL PROTECTED] writes: [ linking libbind causes some obscure problems ] http://www.freebsd.org/cgi/query-pr.cgi?pr=ports/47218 Given that we're not supporting BeOS at the moment anyway, I wonder whether we need libbind on any platform. I know linking it causes some minor problems for me on HPUX (I have to take it out if I want to build with profiling...) Comments anyone? regards, tom lane ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED]) ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faqs/FAQ.html ---(end of broadcast)--- TIP 3: 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 -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end of broadcast)--- TIP 7: don't forget to increase your free space map settings
Re: [BUGS] Bug #926: if old postgresql.jar in CLASSPATH, ant fails
It's been in use for the freebsd port for some months now, so I say yes. ;-) /Palle lördagen den 16 augusti 2003 kl 22.54 skrev Bruce Momjian: Is this patch valid for inclusion in jdbc? --- [EMAIL PROTECTED] wrote: Palle Girgensohn ([EMAIL PROTECTED]) reports a bug with a severity of 3 The lower the number the more severe it is. Short Description if old postgresql.jar in CLASSPATH, ant fails Long Description See http://www.freebsd.org/cgi/query-pr.cgi?pr=48878 If there is an older postgresql.jar file in the ant classpath when building a new postgresql.jar, it will fail. Sample Code Adding includeAntRuntime=no to the compile target in build.xml, as suggested by Tetsurou Okazaki [EMAIL PROTECTED], fixes the problem --- src/interfaces/jdbc/build.xml~ Sun Oct 20 02:10:55 2002 +++ src/interfaces/jdbc/build.xml Mon Mar 3 12:10:37 2003 @@ -101,7 +101,7 @@ !-- This is the core of the driver. It is common for all three versions. -- target name=compile depends=prepare,check_versions,driver -javac srcdir=${srcdir} destdir=${builddir} debug=${debug} +javac includeAntRuntime=no srcdir=${srcdir} destdir=${builddir} debug=${debug} include name=${package}/** / exclude name=${package}/jdbc1/** unless=jdbc1/ No file was uploaded with this report ---(end of broadcast)--- TIP 3: 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 -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
[BUGS] PostgreSQL client has problems when libbind is installed
Hi! This was reported to the FreeBSD bug report system. Under special cirumstances, building postgresql fails. Please see http://www.freebsd.org/cgi/query-pr.cgi?pr=ports/47218 This sounds like a bug in the build process to me. I cannot confirm though. Please check with the original author. /Palle ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
[BUGS] select a.name ... union select a.name ... order by a.name fails in 7.1
Hi! I'm not certain this is correct SQL, and I know the rewrite is easy, but here is a difference I found between 7.0.3 and 7.1: select a.name from users a union select a.name from oldusers a order by a.userid; ERROR: Relation 'a' does not exist This works fine in postgres 7.0.x the simple rewrite is of course 'order by userid', but it is tedious to find all places in our system where this happens. It seems some programmers have used the above syntax a lot :( Is it a bug, or was it a feature in 7.0? /Palle -- Partitur Informationsteknik AB Wenner-Gren Center +46 8 566 280 02 113 46 Stockholm +46 70 785 86 02 Sweden [EMAIL PROTECTED] ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [BUGS] select a.name ... union select a.name ... order by a.name fails in 7.1
Tom, thanks for your input. Now I have something to bash my programmers over the head with ;-) No, just kidding... But as you say, it is plain bad SQL, I realize this fully now. Oddly enough, I have a SQL book that has one or two examples with order by a.foobar. Oh well... Thanks for the prompt answer! /Palle Tom Lane wrote: [EMAIL PROTECTED] writes: I'm not certain this is correct SQL, and I know the rewrite is easy, but here is a difference I found between 7.0.3 and 7.1: select a.name from users a union select a.name from oldusers a order by a.userid; ERROR: Relation 'a' does not exist It's not correct SQL, and Postgres doesn't support it. This works fine in postgres 7.0.x No it doesn't. Consider the following example (done with 7.0.*): play= create table foo (f1 int, f2 int); CREATE play= insert into foo values (1,10); INSERT 1021258 1 play= insert into foo values (2,9); INSERT 1021259 1 play= insert into foo values (3,8); INSERT 1021260 1 play= select a.f1 from foo a union select a.f1 from foo a; f1 1 2 3 (3 rows) -- so far so good, but: play= select a.f1 from foo a union select a.f1 from foo a play- order by a.f2; f1 3 2 1 2 3 (5 rows) -- wow, ORDER BY produces a different resultset! While this is obviously broken, the more fundamental point is that the only sensible ordering of a UNION result is on one of the result columns. Otherwise the ordering isn't well-defined: if the UNION merges equal values of name from the two SELECTs, which SELECT's userid will get used for the sort? So the SQL standard mandates ORDER BY only on output column names or numbers, and we enforce that in 7.1. Prior versions failed to notice that there was a problem, and would do something fairly random instead :-( the simple rewrite is of course 'order by userid', but it is tedious to find all places in our system where this happens. It seems some programmers have used the above syntax a lot :( They haven't thought about whether the query is reasonable. I'd also say that they never looked closely to see if the output they were getting was reasonable ... regards, tom lane ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [BUGS] The Data Base System is in recovery mode
{ RESDOM :resno 3 :restype 25 :restypmod -1 :resname fnamn :reskey 0 :reskeyop 0 :ressortgroupref 0 :resjunk false } :expr { VAR :varno 3 :\ varattno 3 :vartype 25 :vartypmod -1 :varlevelsup 0 :varnoold 3 :varoattno 3}} { TARGETENTRY :resdom { RESDOM :resno 4 :restype 25 :restypm\ od -1 :resname enamn :reskey 0 :reskeyop 0 :ressortgroupref 0 :resjunk false } :expr { VAR :varno 3 :varattno 4 :vartype 25 :vartypmod -1 :\ varlevelsup 0 :varnoold 3 :varoattno 4}} { TARGETENTRY :resdom { RESDOM :resno 5 :restype 25 :restypmod -1 :resname titel :reskey 0 :reskeyo\ p 0 :ressortgroupref 0 :resjunk false } :expr { VAR :varno 3 :varattno 5 :vartype 25 :vartypmod -1 :varlevelsup 0 :varnoold 3 :varoattno 5}\ } { TARGETENTRY :resdom { RESDOM :resno 6 :restype 25 :restypmod -1 :resname telefon :reskey 0 :reskeyop 0 :ressortgroupref 0 :resjunk false\ } :expr { VAR :varno 3 :varattno 10 :vartype 25 :vartypmod -1 :varlevelsup 0 :varnoold 3 :varoattno 10}} { TARGETENTRY :resdom { RESDOM :r\ esno 7 :restype 25 :restypmod -1 :resname mobil :reskey 0 :reskeyop 0 :ressortgroupref 0 :resjunk false } :expr { VAR :varno 3 :varattno 12 \ :vartype 25 :vartypmod -1 :varlevelsup 0 :varnoold 3 :varoattno 12}} { TARGETENTRY ::resdom { RESDOM :resno 8 :restype 25 :restypmod -1 :res\ name email :reskey 0 :reskeyop 0 :ressortgroupref 0 :resjunk false } :expr { VAR :varno 3 :varattno 13 :vartype 25 :vartypmod -1 :varlevels\ up 0 :varnoold 3 :varoattno 13}} { TARGETENTRY :resdom { RESDOM :resno 9 :restype 1184 :restypmod -1 :resname tidpunkt :reskey 0 :reskeyop 0\ :ressortgroupref 0 :resjunk false } :expr { VAR :varno 3 :varattno 16 :vartype 1184 ::vartypmod -1 :varlevelsup 0 :varnoold 3 :varoattno 16\ }}) :qual { EXPR :typeOid 16 :opType and :oper :args ({ CONST :consttype 16 :constlen 1 :constisnull false :constvalue 1 [ 1 0 0 0 ] :\ constbyval true } { EXPR :typeOid 16 :opType op :oper { OPER :opno 1209 :opid 0 :opresulttype 16 } :args ({ EXPR :typeOid 25 :opType func \ :oper { FUNC :funcid 870 :functype 25 :funcisindex false :funcsize 0 :func_fcache @ :0x0 :func_tlist ({ TARGETENTRY :resdom { RESDOM :resno \ 1 :restype 25 :restypmod -1 :resname \\noname :reskey 0 :reskeyop 0 :ressortgroupref 0 :resjunk false } :expr { VAR :varno -1 :varattno 1 \ :vartype 25 :vartypmod -1 :varlevelsup 0 :varnoold -1 :varoattno 1}}) :func_planlist :} :args ({ VAR :varno 3 :varattno 2 :vartype 25 :var\ typmod -1 :varlevelsup 0 :varnoold 3 :varoattno 2})} { EXPR :typeOid 25 :opType func :oper { FUNC :funcid 870 :functype 25 :funcisindex fa\ lse :funcsize 0 :func_fcache @ 0x0 :func_tlist ({ TARGETENTRY :resdom { RESDOM :resno 1 :restype 25 :restypmod -1 :resname \\noname :resk\ ey 0 :reskeyop 0 :ressortgroupref 0 :resjunk false } :expr { VAR :varno -1 :varattno 1 :vartype 25 :vartypmod -1 :varlevelsup 0 :varnoold -\ 1 :varoattno 1}}) :func_planlist ProcessQuery CommitTransactionCommand StartTransactionCommand query: begin transaction ProcessUtility: begin transaction CommitTransactionCommand StartTransactionCommand query: declare curse cursor for select personid,foretag,fnamn,enamn,titel,telefon,mobil,email,tidpunkt, lower(foretag) from wtabniclas order\ by lower(foretag) ProcessQuery CommitTransactionCommand StartTransactionCommand query: fetch forward 20 from curse ProcessUtility: fetch forward 20 from curse CommitTransactionCommand StartTransactionCommand query: abort transaction ProcessUtility: abort transaction CommitTransactionCommand StartTransactionCommand query: select distinct personid,foretag, lower(foretag) from wtabniclas order by lower(foretag) ProcessQuery Server process (pid 6757) exited with status 139 at Wed Oct 18 14:51:53 2000 Terminating any active server processes... Server processes were terminated at Wed Oct 18 14:51:53 2000 Reinitializing shared memory and semaphores DEBUG: Data Base System is starting up at Wed Oct 18 14:51:53 2000 DEBUG: Data Base System was interrupted being in production at Wed Oct 18 14:51:21 2000 DEBUG: Data Base System is in production state at Wed Oct 18 14:51:53 2000 Hope this helps... /Palle Tom Lane [EMAIL PROTECTED] writes: Palle Girgensohn [EMAIL PROTECTED] writes: I have a problem with postgresl-7.0.2 on Freebsd 4-stable. query: declare curse cursor for select personid,foretag,fnamn,enamn,titel,telefon,mobil,email,tidpunkt, lower(foretag) from wtabmaria order by lower(foretag) query: fetch forward 20 from curse Server process (pid 15342) exited with status 139 at Tue Oct 17 14:37:20 2000 Hm. I couldn't duplicate this crash using 7.0.2-plus-some-7.0.3-patches. However I don't recall any bug fixes for cursors in the 7.0.* branch. Could you provide a more complete bug report, like the complete schema for the table? Also, can you provide a gdb traceback from the corefile that the crashing backend hopefully left behind in the database subdirectory ($PGDATA/data/base/yourdb)? regards, tom lane
Re: [BUGS] The Data Base System is in recovery mode
Tom Lane [EMAIL PROTECTED] writes: and GetTemplateEncoding is just about as SEGV-proof as any routine I've ever seen: static inttemplateEncoding; ... int GetTemplateEncoding() { return (templateEncoding); } Heh I speculate that you gave gdb the wrong executable file to compare to the core file ... I too had a hard time believeing it. I used xemacs (M-x gdb-with-core) and used the core dump and /usr/local/pgsql/bin/postgres. I was surprised that there was funcion names, since there shouldn't be any (no debugging symbols). Hence I think too, this backtrace is rubbish. :( I'll see if I can get one tomorrow, but I will have to rebuild postgres with -g and recreate the error. Problem is, I couldn't do it on my machine. Odd... I'll get back to you when I have more info. Cheers, Palle
Re: [BUGS] The Data Base System is in recovery mode
for select personid,foretag,fnamn,enamn,titel,telefon,mobil,email,tidpunkt, lower(foretag) from wtabmaria order by lower(foretag ) ProcessQuery CommitTransactionCommand StartTransactionCommand query: fetch forward 20 from curse ProcessUtility: fetch forward 20 from curse CommitTransactionCommand StartTransactionCommand query: select distinct personid,foretag, lower(foretag) from wtabmaria order by lower(foretag) ProcessQuery Server process (pid 3096) exited with status 139 at Wed Oct 18 09:19:00 2000 Terminating any active server processes... NOTICE: Message from PostgreSQL backend: The Postmaster has informed me that some other backend died abnormally and possibly corrupted shared memory. I have rolled back the current transaction and am going to terminate your database system connection and exit. Please reconnect to the database system and repeat your query. NOTICE: Message from PostgreSQL backend: The Postmaster has informed me that some other backend died abnormally and possibly corrupted shared memory. I have rolled back the current transaction and am going to terminate your database system connection and exit. Please reconnect to the database system and repeat your query. NOTICE: Message from PostgreSQL backend: The Postmaster has informed me that some other backend died abnormally and possibly corrupted shared memory. I have rolled back the current transaction and am going to terminate your database system connection and exit. Please reconnect to the database system and repeat your query. Server processes were terminated at Wed Oct 18 09:19:01 2000 Reinitializing shared memory and semaphores DEBUG: Data Base System is starting up at Wed Oct 18 09:19:01 2000 DEBUG: Data Base System was interrupted being in production at Wed Oct 18 01:12:29 2000 DEBUG: Data Base System is in production state at Wed Oct 18 09:19:01 2000 It looks very much like an out-of-memory error, since now, the fetch is ok, but it fails on the next query, which is a memory hog (~5500 records). I don't know why the progammer initially put the second select inside the transaction, and I have no put an "abort" after the "fetch", to see if I might get away with it now. It seems to work (using -B 2000 too)... The higher I set '-B', the harder it gets to reproduce the error. I *think* I might be able to find a value for '-B' where I can just lower it by one and get it to fail, (given no other users in the database system). Schema: CREATE SEQUENCE "pidseq" start 9700 increment 1 maxvalue 2147483647 minvalue 1 cache 1 ; CREATE TABLE "personer" ( "personid" int4 DEFAULT nextval('pidseq'::text) NOT NULL, "foretag" text, "fnamn" text, "enamn" text, "titel" text, "adress1" text, "adress2" text, "postadress" text, "land" character(2) DEFAULT 'SE' NOT NULL, "telefon" text, "telefax" text, "mobil" text, "email" text, "privemail" text, "url" text, "tidpunkt" timestamp DEFAULT now() NOT NULL, "avdelning" text ); There are about 5500 records in this table. /Palle Tom Lane [EMAIL PROTECTED] writes: Palle Girgensohn [EMAIL PROTECTED] writes: I have a problem with postgresl-7.0.2 on Freebsd 4-stable. query: declare curse cursor for select personid,foretag,fnamn,enamn,titel,telefon,mobil,email,tidpunkt, lower(foretag) from wtabmaria order by lower(foretag) query: fetch forward 20 from curse Server process (pid 15342) exited with status 139 at Tue Oct 17 14:37:20 2000 Hm. I couldn't duplicate this crash using 7.0.2-plus-some-7.0.3-patches. However I don't recall any bug fixes for cursors in the 7.0.* branch. Could you provide a more complete bug report, like the complete schema for the table? Also, can you provide a gdb traceback from the corefile that the crashing backend hopefully left behind in the database subdirectory ($PGDATA/data/base/yourdb)? regards, tom lane -- Partitur Informationsteknik AB Wenner-Gren Center +46 8 566 280 02 113 46 Stockholm +46 70 785 86 02 Sweden [EMAIL PROTECTED]
[BUGS] reproducable command sequence to get mdopen: Couldn't open...
Hi! I have seen this problem on several occasions, using 7.02: # vacuum; NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: RelationIdBuildRelation: smgropen(at_provider_user_info): No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory ERROR: cannot open relation at_provider_user_info To get it back to functional, I have to touch the file data/base/dbname/at_provider_user_info, drop the table and restore it from a backup (if it wasn't empty, which was luckily mostly the case). This works, but there's a bug letting this happen... Here's how I get it happening: Script started on Sun Oct 15 18:06:12 2000 $ psql -e src/core/db/elever-to-person.sql You are now connected as new user nobody. BEGIN; BEGIN CREATE TABLE person (userid text, giver text, first_name text, last_name text, email text, default_language text DEFAULT 'sv'); CREATE CREATE UNIQUE INDEX person_index on person using btree (userid text_ops); CREATE CREATE INDEX person_giver_idx on person (giver); CREATE INSERT INTO person (userid,giver,first_name,last_name,email,default_language) SELECT userid,giver,fnamn,enamn,datorpost,defaultlanguage FROM elever; INSERT 1189717 1 COMMIT; COMMIT BEGIN; BEGIN ALTER TABLE at_provider_user_info RENAME TO tmp; NOTICE: Caution: RENAME TABLE cannot be rolled back, so don't abort now ALTER CREATE TABLE at_provider_user_info (id int4, giver text, name text, order_nr int4, type text, az_only bool DEFAULT false, hideable bool DEFAULT false, is_unique bool DEFAULT false ); CREATE INSERT INTO at_provider_user_info (id,giver,name,order_nr,type,az_only,hideable,is_unique) SELECT *,false,false,false FROM tmp; ERROR: INSERT has more expressions than target columns DROP TABLE tmp; NOTICE: current transaction is aborted, queries ignored until end of transaction block *ABORT STATE* ALTER TABLE at_user_info RENAME TO tmp; NOTICE: current transaction is aborted, queries ignored until end of transaction block *ABORT STATE* CREATE TABLE at_user_info (userid text, info_id int4, content text, visible bool DEFAULT true ); NOTICE: current transaction is aborted, queries ignored until end of transaction block *ABORT STATE* INSERT INTO at_user_info (userid,info_id,content,visible) SELECT *,true FROM tmp; ERROR: Attribute 'visible' is of type 'bool' but expression is of type 'int4' You will need to rewrite or cast the expression DROP TABLE tmp; NOTICE: current transaction is aborted, queries ignored until end of transaction block *ABORT STATE* INSERT INTO at_provider_user_info (id,giver,name,order_nr,type) SELECT nextval('at_provider_user_info_seq'),giver,'Gatuadress',10,'text' FROM providers; NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: RelationIdBuildRelation: smgropen(at_provider_user_info): No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: current transaction is aborted, queries ignored until end of transaction block *ABORT STATE* INSERT INTO at_user_info (userid,info_id,content) SELECT userid,currval('at_provider_user_info_seq'),gatuadress FROM elever; NOTICE: current transaction is aborted, queries ignored until end of transaction block *ABORT STATE* INSERT INTO at_provider_user_info (id,giver,name,order_nr,type) SELECT nextval('at_provider_user_info_seq'),giver,'Postnummer',11,'text' FROM providers; NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: current transaction is aborted, queries ignored until end of transaction block *ABORT STATE* INSERT INTO at_user_info (userid,info_id,content) SELECT userid,currval('at_provider_user_info_seq'),postnummer FROM elever; NOTICE: current transaction is aborted, queries ignored until end of transaction block *ABORT STATE* INSERT INTO at_provider_user_info (id,giver,name,order_nr,type) SELECT nextval('at_provider_user_info_seq'),giver,'Postadress',12,'text' FROM providers; NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: mdopen: couldn't open at_provider_user_info: No such file or directory NOTICE: current transaction is aborted, queries ignored until end of transaction block *ABORT STATE* INSERT INTO at_user_info (userid,info_id,content) SELECT