Re: [BUGS] ECPG CONNECT TO DEFAULT segfault
On Tue, Dec 28, 2004 at 11:15:52PM -0700, Michael Fuhr wrote: > PostgreSQL 8.0.0rc2, 7.4.6 > Solaris 9, FreeBSD 4.11-PRERELEASE > > ECPG's CONNECT TO DEFAULT causes a segmentation fault. This > originally came up in the "Connection without database name" thread > in pgsql-hackers: > This is exactly what I did fix with the work mentioned in the hackers thread. It should work nicely with CVS HEAD. Could you please try this? BTW Tom, I just found that my problem with PGDATABASE was a local one. It surely works, so the oracle style syntax will remain. That is you can also write EXEC SQL CONNECT to connect to the default database using the given username. Michael -- Michael Meskes Email: Michael at Fam-Meskes dot De ICQ: 179140304, AIM/Yahoo: michaelmeskes, Jabber: [EMAIL PROTECTED] Go SF 49ers! Go Rhein Fire! Use Debian GNU/Linux! Use PostgreSQL! ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
[BUGS] BUG #1365: Loose connection in idle mode
The following bug has been logged online: Bug reference: 1365 Logged by: Jean Claude Email address: [EMAIL PROTECTED] PostgreSQL version: 8.0 Operating system: W2K Description:Loose connection in idle mode Details: I didn't found any related topic about timeout connection with a non activity of clients. In fact the server disconnect any client after ~10 minutes of inactivity. I try it with my program that i am developping, psql and pgadmin III. All do the same after arround 10 minutes with inactivity or idle, the server diconnect by logging a message "unexpected EOF on client connection". Hoping to find out the reason or maybe a bug but all my project is in standby now to fix this problem. Thanks in advance for all your help JClaude ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
[BUGS] unpredictable cost in explain
Hello !
Please help me, I tried to make request run faster
1st request (original):
SELECT dy.*, sat.dton as satdton
FROM dyBilling dy, saType sat
WHERE dy.dt>='20-12-2004' AND dy.dt<='20-12-2004' AND
dy.servid=sat.servid AND dy.dt BETWEEN sat.dton AND sat.dtoff
it have execution time 4174,514 ms and
following query plan:
Nested Loop (cost=0.00..11.57 rows=1 width=41)
-> Index Scan using dybilling_dt_servid_sid on dybilling dy
(cost=0.00..5.69 rows=1 width=37)
Index Cond: ((dt >= '20-12-2004'::date) AND (dt <=
'20-12-2004'::date))
-> Index Scan using satype_dton_dtoff_servid on satype sat
(cost=0.00..5.86 rows=1 width=12)
Index Cond: (("outer".dt >= sat.dton) AND ("outer".dt <=
sat.dtoff) AND ("outer".servid = sat.servid))
2nd request:
SELECT dy.*, sat.dton as satdton
FROM dyBilling dy, saType sat
WHERE dy.dt='20-12-2004' AND -- <- diff only here
dy.servid=sat.servid AND dy.dt BETWEEN sat.dton AND sat.dtoff
it have execution time 106,340 ms and
following query plan:
Merge Join (cost=3973.95..4037.21 rows=195 width=41)
Merge Cond: ("outer".servid = "inner".servid)
Join Filter: (("outer".dt >= "inner".dton) AND ("outer".dt <=
"inner".dtoff))
-> Sort (cost=3451.52..3453.99 rows=989 width=37)
Sort Key: dy.servid
-> Index Scan using dybilling_dt_servid_sid on dybilling dy
(cost=0.00..3402.32 rows=989 width=37)
Index Cond: (dt = '20-12-2004'::date)
-> Sort (cost=522.43..538.58 rows=6460 width=12)
Sort Key: sat.servid
-> Seq Scan on satype sat (cost=0.00..113.60 rows=6460 width=12)
So,
Estimated Cost Real Execution Time, ms
1st request0.00..11.57 4174,514
2ns request3973.95..4037.21106,340
Why? Is this EXPLAIN's behaviour correct ?
Results reproducable any number of times, no any caching effects etc.
Some maybe useful info:
isbs=# select count(*) from dyBilling;
-
1144078
isbs=# select count(*) from saType;
---
6472
isbs=# select version();
-
PostgreSQL 7.4.1 on i386-unknown-freebsd4.7, compiled by GCC 2.95.4
Thanks.
--
best regards,
Ruslan A Dautkhanov [EMAIL PROTECTED]
smime.p7s
Description: S/MIME Cryptographic Signature
Re: [BUGS] RC2 on OS X can not make check
I'll try again this morning without the diable-shared. However, I tried this because of a recomendation for using libpq. When I tried to compile my app (using xcode on OS X) it complained about the dylib I will try again in a little while and let you know. Ted --- Tom Lane <[EMAIL PROTECTED]> wrote: > Theodore Petrosky <[EMAIL PROTECTED]> writes: > > I just found the time to try 8.0RC2 on OSX and I > have > > a problem with 'make check'. > > > Here is my configure.. > > > ./configure --disable-shared ... > > --disable-shared breaks plpgsql, and probably a few > other things. I'm > not exactly sure why we still advertise it as a > supported option. > > regards, tom lane > __ Do you Yahoo!? Jazz up your holiday email with celebrity designs. Learn more. http://celebrity.mail.yahoo.com ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [BUGS] ECPG CONNECT TO DEFAULT segfault
On Wed, Dec 29, 2004 at 10:32:21AM +0100, Michael Meskes wrote: > On Tue, Dec 28, 2004 at 11:15:52PM -0700, Michael Fuhr wrote: > > > > ECPG's CONNECT TO DEFAULT causes a segmentation fault. This > > originally came up in the "Connection without database name" thread > > in pgsql-hackers: > > This is exactly what I did fix with the work mentioned in the hackers > thread. It should work nicely with CVS HEAD. Could you please try this? Try what? I don't see a patch, link, or suggested fix in the thread -- just your original message and the two followups by Tom Lane and me. Or was the fix discussed in a different thread? -- Michael Fuhr http://www.fuhr.org/~mfuhr/ ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [BUGS] BUG #1365: Loose connection in idle mode
"Jean Claude" <[EMAIL PROTECTED]> writes: > I didn't found any related topic about timeout connection with a non > activity of clients. In fact the server disconnect any client after ~10 > minutes of inactivity. There is no disconnect timeout in Postgres. This sort of thing has been reported before, and it's always been due to overaggressive dropping of idle connections by a firewall in between the clients and server. NAT routers have to keep per-connection state about the address mapping, and they will drop that state after some period of no traffic. 10 minutes seems unreasonably short though. I hope it's configurable on your box. If not, you can either buy a more reasonably designed router, or tweak your clients so they do something at least once every few minutes... regards, tom lane ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
[BUGS] plperl and pgport
Hi, Compiling 8.0.0RC2 on AIX 4.3 on an RS600 with gcc, in src/pl/plperl, got the following error: gcc -O2 -Wall -Wmissing-prototypes -Wpointer-arith -fno-strict-aliasing -Wl,-bnoentry -Wl,-H512 -Wl,-bM:SRE -o libplperl.so libplperl.a -L../../../src/port -L/usr/local/lib -L/da23/d56/croquet/readline/lib /usr/local/lib/perl5/5.6.0/aix/auto/DynaLoader/DynaLoader.a -L/usr/local/lib/perl5/5.6.0/aix/CORE -lperl -lnsl -ldbm -ldl -lld -lm -lC -lc -lbsd -lPW -liconv -Wl,-bI:../../../src/backend/postgres.imp -Wl,-bE:libplperl.exp ld: 0711-317 ERROR: Undefined symbol: .pg_strcasecmp ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more information. collect2: ld returned 8 exit status Adding -lpgport to this command line allowed the build to continue. Is this the proper place to send build errors like this? What's a useful subject line? Nick Addington ---(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
[BUGS] compiling on AIX
Hi, Compiling 8.0.0RC2 on AIX 4.3 on an RS600 with gcc, in src/backend, got the following error: gcc -O2 -Wall -Wmissing-prototypes -Wpointer-arith -fno-strict-aliasing -L../../src/port -L/usr/local/lib -L/da23/d56/croquet/readline/lib access/SUBSYS.o bootstrap/SUBSYS.o catalog/SUBSYS.o parser/SUBSYS.o commands/SUBSYS.o executor/SUBSYS.o lib/SUBSYS.o libpq/SUBSYS.o main/SUBSYS.o nodes/SUBSYS.o optimizer/SUBSYS.o port/SUBSYS.o postmaster/SUBSYS.o regex/SUBSYS.o rewrite/SUBSYS.o storage/SUBSYS.o tcop/SUBSYS.o utils/SUBSYS.o ../../src/timezone/SUBSYS.o -Wl,-bE:../../src/backend/postgres.imp -lpgport_srv -lz -lreadline -ltermcap -lPW -lld -lnsl -ldl -lm -o postgres ld: 0711-781 ERROR: TOC overflow. TOC size: 72616 Maximum size: 65536 collect2: ld returned 12 exit status make: *** [postgres] Error 1 Adding "-Xlinker -bbigtoc" to this command line (enabling the "bigtoc" linker option) allowed the build to continue, but this is the ld man page's comment: bigtoc Generates extra code if the size of the table of contents (TOC) is greater than 64KB. Extra code is needed for every reference to a TOC symbol that cannot be addressed with a 16-bit offset. Because a program containing generated code may have poor performance, you should reduce the number of TOC entries needed by the program before using this option. The default is the nobigtoc option. Please advise. Nick Addington ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
[BUGS] BUG #1360: Use PostgreSQL OLE DB Provider in VS.NET 2003
The following bug has been logged online: Bug reference: 1360 Logged by: Tran Duc Khoa Email address: [EMAIL PROTECTED] PostgreSQL version: 8.0 Operating system: Windows Advance Server 2000 Description:Use PostgreSQL OLE DB Provider in VS.NET 2003 Details: Hi all In VS.NET 2003, I use PostgreSQL OLE DB Provider (pgoledb.dll in folder ..\postgreSQL\8.0.0-beta4\PgOleDb) to connect to PostgreSQL 8.0 beta5 database. When I create a Data connection, an error happend "Not enough storage is available to complete this operation" ---(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
[BUGS] postgresql-8.0.0-rc2.msi failed to install on Dynamic NTFS Partition
POSTGRESQL BUG REPORT TEMPLATE Your name : Egon Phillips Your email address : [EMAIL PROTECTED] System Configuration - Architecture (example: Intel Pentium) : AMD 2500XP Operating System (example: Linux 2.4.18) : Win2000 5.00.2195, sp4 PostgreSQL version (example: PostgreSQL-8.0): postgresql-8.0.0-rc2.msi Compiler used (example: gcc 2.95.2) : none Please enter a FULL description of your problem: postgresql-8.0.0-rc2.msi will not install on drive e:. The following directory trees are created: "E:\Program Files\PostgreSQL\8.0.0-rc2\data" "E:\Program Files\PostgreSQL\8.0.0-rc2\tmp" I've attempted implementing the work-around as described in the following: "Running & Installing PostgreSQL On Native Windows FAQ, 3.5) I'm getting permissions errors when installing/running initdb. But, the error persists as described in 3.5. I solve the problem by installing the postgresql-8.0.0-rc2.msi in: "C:\Program Files\PostgreSQL\8.0.0-rc2" I believe the correct result should be to install postgresql-8.0.0-rc2.msi in: "E:\Program Files\PostgreSQL\8.0.0-rc2" Please describe a way to repeat the problem. Please try to provide a concise reproducible example, if at all possible: Create a "Program Files" subdirectory tree on a drive other than "C:\", I would suggest trying "E:\". Apply the workaround described in "Running & Installing PostgreSQL On Native Windows FAQ, 3.5). Attempt to install postgresql-8.0.0-rc2.msi in "E:\Program Files". Drive "E:\" is an NTFS Dynamic Partition. -- Reproduction from commands in description. If you know how this problem might be fixed, list the solution below: - The Brockton Initiative (http://brockton.dyndns.org) Egon Phillips, Chair ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
[BUGS] BUG #1361: errors in install
The following bug has been logged online: Bug reference: 1361 Logged by: Email address: [EMAIL PROTECTED] PostgreSQL version: 7.4.6 Operating system: Linux Description:errors in install Details: Hi, I have two errors : * When i want to create a DB, he replies : "relocation error: createdb: undefined symbol: get_progname" * I can't start or stop the database. It replies that i have an old version of PG, but i don't know how to delete them end reinstall it. Can anybody help me? Thanks. M. DANG ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faqs/FAQ.html
[BUGS] German Setup Win32 (V8 RC2)
Dear Win32-Setup team, I have a few comments concerning the german setup. There are some small typos in the wizard Step "Dienste-Konfiguration". 1. "Bitte hiere einen..." -> "Bitte hier einen..." 2. "abbrechen" -> "Abbrechen" 3. "zurück" -> "Zurück" 4. "nächster" -> "Weiter" 5. The Labels (static text) are to small Please see attached Bitmap. Regards, Pit ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
[BUGS] BUG #1357: Installation via Terminal Service
The following bug has been logged online: Bug reference: 1357 Logged by: Winfried Maus Email address: [EMAIL PROTECTED] PostgreSQL version: 8.0 Operating system: Windows Server 2003 Description:Installation via Terminal Service Details: When trying to install PostgreSQL 8.0.0 RC2 within a CONSOLE Terminal Session, the installer prompts a dialog box, that PostgreSQL cannot be installed using a terminal session but must be run from the console. Since RDP Client 5.2 supports the /console switch to force a terminal session into console mode which works fine with every other software I tried to install, I consider this behaviour a bug in the PostgreSQL installer. Merry x-mas and greetings from Germany. :-) ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
[BUGS] PostgreSQL 8.0.0 Release Candidate 2 announcement
pls, link to the most important info - the release notes detailing what's changed since a) RC1 abd b) 7.4. ---(end of broadcast)--- TIP 7: don't forget to increase your free space map settings
[BUGS] BUG #1359: there is a problem in "SELECT ....GROUP BY"
The following bug has been logged online: Bug reference: 1359 Logged by: jingold chen Email address: [EMAIL PROTECTED] PostgreSQL version: 8.0 Operating system: window xp home edition Description:there is a problem in "SELECT GROUP BY" Details: I have install a postgresql 8.0 RC2 for window. And Create a database with "Unicode" But I send a query at "pgAdminIII 1.2": "select recie_man,count(id) from public.case_table group by 1 order by 2" It return a recordset with duplicate record. --- the table(case_table) definition: CREATE TABLE public.case_table ( id int4 NOT NULL, caseclass char(1), linkman char(24), emailaddr char(32), address text, phone char(16), recie_man char(64), ) WITH OIDS; I don't know what worry in it. Is it the postgresql 8.0 bug? I have do the same test in postgresql 7.4 for linux and gain the right result (It return a recordset without duplicate) ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faqs/FAQ.html
[BUGS] German Win32 Setup (V8 RC2)
The password hint is not written in correct german : Better: Das angegebene Kennwort ist nicht sicher genug. Soll das Installationsprogramm es durch ein sichereres, zufälliges ersetzen? Regards, Pit PS: The setup seems to have problems displaying german umlauts. Please see attached screenshot <> ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
[BUGS] BUG #1363: Loose connection in idle mode
The following bug has been logged online: Bug reference: 1363 Logged by: Jean Claude Email address: [EMAIL PROTECTED] PostgreSQL version: 8.0 Operating system: W2K Description:Loose connection in idle mode Details: I didn't found any related topic about timeout connection with a non activity of clients. In fact the server disconnect any client after ~10 minutes of inactivity. I try it with my program that i am developping, psql and pgadmin III. All do the same after arround 10 minutes with inactivity or idle, the server diconnect by logging a message "unexpected EOF on client connection". Hoping to find out the reason or maybe a bug but all my project is in standby now to fix this problem. Thanks in advance for all your help JClaude ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send "unregister YourEmailAddressHere" to [EMAIL PROTECTED])
[BUGS] function return type for triggers in postgresql 8.0 rc2
I am porting an existing database to postgresql and am translating some triggers. I have the following objects: A table named 'currentstatus' a function which I orignally defined as: CREATE OR REPLACE FUNCTION "CurrentStatus_DTrigx"() RETURNS opaque AS $BODY$ Begin if (select COUNT(*) FROM deleted, Stations WHERE deleted.CurrentStatusCode = Stations.CurrentStatusCode) > 0 then RAISE EXCEPTION 'RAISERROR(778501, 16, 1)'; ROLLBACK TRANSACTION; end if; END $BODY$ LANGUAGE 'plpgsql' VOLATILE; a trigger I defined as: create trigger csd before delete on currentstatus for each row execute procedure public.CurrentStatus_DTrig() after struggling with 'function not defined errors', and unable to see what was wrong I redefined the function as: CREATE OR REPLACE FUNCTION currentstatus_dtrig() RETURNS opaque AS $BODY$ Begin if (select COUNT(*) FROM deleted, Stations WHERE deleted.CurrentStatusCode = Stations.CurrentStatusCode) > 0 then RAISE EXCEPTION 'RAISERROR(778501, 16, 1)'; ROLLBACK TRANSACTION; end if; END $BODY$ LANGUAGE 'plpgsql' VOLATILE; and got the message to the effect of 'redefining return type as trigger', and the following function was created: -- Function: currentstatus_dtrig() -- DROP FUNCTION currentstatus_dtrig(); CREATE OR REPLACE FUNCTION currentstatus_dtrig() RETURNS "trigger" AS $BODY$ Begin if (select COUNT(*) FROM deleted, Stations WHERE deleted.CurrentStatusCode = Stations.CurrentStatusCode) > 0 then RAISE EXCEPTION 'RAISERROR(778501, 16, 1)'; ROLLBACK TRANSACTION; end if; END $BODY$ LANGUAGE 'plpgsql' VOLATILE; ALTER FUNCTION currentstatus_dtrig() OWNER TO postgres; GRANT EXECUTE ON FUNCTION currentstatus_dtrig() TO public; GRANT EXECUTE ON FUNCTION currentstatus_dtrig() TO postgres; This is strange, AFAICT I followed the documentation setting up the trigger and functions (correct parameter signature and return type). Did I do something wrong or is this a RC2 issue? Thanks, Pedro Lopez ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faqs/FAQ.html
[BUGS] BUG #1364: using field names or table names with blanks
The following bug has been logged online: Bug reference: 1364 Logged by: Pit Mueller Email address: [EMAIL PROTECTED] PostgreSQL version: 8.0 Operating system: WINXP Description:using field names or table names with blanks Details: MS SQL Server uses a quite elegant method for using field names and table names: select [field] from [table] f.ex. select [my field] from [the big table] PostgreSQL 8.0 should support this notation too. Brackets are not supported at the moment -> the SQL parser notifies an syntax error ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
[BUGS] BUG #1362: Storing large object like 35mb files.
The following bug has been logged online: Bug reference: 1362 Logged by: Email address: [EMAIL PROTECTED] PostgreSQL version: 7.4.6 Operating system: linux Description:Storing large object like 35mb files. Details: There is problem how to store large files in postgres.Which datatypes available for storing large file like 35MB. ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
[BUGS] connection error
Hello I am Manish.Well i configured Freeradius with Postgresql on Redhat Fedora.well when i start the freeradius i got this error as below: Couldn't connect socket to PostgreSQL server [EMAIL PROTECTED]:radius Error: rlm_sql_postgresql: Postgresql error 'could not connect to server: Connection refused ?Is the server running on host localhost and accepting ?TCP/IP connections on port 5432? ' Anticipatinf a favourable response from you. Thanking you Manish
Re: [BUGS] plperl and pgport
Nicolas Addington <[EMAIL PROTECTED]> writes: > Compiling 8.0.0RC2 on AIX 4.3 on an RS600 with gcc, in src/pl/plperl, got > the following error: > gcc -O2 -Wall -Wmissing-prototypes -Wpointer-arith -fno-strict-aliasing > -Wl,-bnoentry -Wl,-H512 -Wl,-bM:SRE -o libplperl.so libplperl.a > -L../../../src/port -L/usr/local/lib -L/da23/d56/croquet/readline/lib > /usr/local/lib/perl5/5.6.0/aix/auto/DynaLoader/DynaLoader.a > -L/usr/local/lib/perl5/5.6.0/aix/CORE -lperl -lnsl -ldbm -ldl -lld -lm -lC > -lc -lbsd -lPW -liconv -Wl,-bI:../../../src/backend/postgres.imp > -Wl,-bE:libplperl.exp > ld: 0711-317 ERROR: Undefined symbol: .pg_strcasecmp > ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more > information. > collect2: ld returned 8 exit status > Adding -lpgport to this command line allowed the build to continue. That seems like the wrong fix. Why isn't the linker finding pg_strcasecmp in the main backend? regards, tom lane ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [BUGS] function return type for triggers in postgresql 8.0 rc2
"PEDRO LOPEZ" <[EMAIL PROTECTED]> writes: > after struggling with 'function not defined errors', and unable to see what > was wrong I redefined the function as: > CREATE OR REPLACE FUNCTION currentstatus_dtrig() RETURNS opaque AS > ... > and got the message to the effect of 'redefining return type as trigger', Yup. > This is strange, AFAICT I followed the documentation setting up the trigger > and functions (correct parameter signature and return type). It's been a long time (at least since 7.3) since the documentation recommended using OPAQUE for anything. regards, tom lane ---(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
Re: [BUGS] BUG #1364: using field names or table names with blanks
"Pit Mueller" <[EMAIL PROTECTED]> writes: > MS SQL Server uses a quite elegant method for using field names and table > names: > select [field] from [table] > f.ex. select [my field] from [the big table] One man's "elegant" is another's "ugly and nonstandard". > PostgreSQL 8.0 should support this notation too. Not likely. Use double quotes like the SQL standard tells you to: select "my field" from "the big table" One would hope that SQL Server takes this too, since they claim some amount of spec compliance. regards, tom lane ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
[BUGS] BUG #1366: spam from yourselves
The following bug has been logged online: Bug reference: 1366 Logged by: keith berrington Email address: [EMAIL PROTECTED] PostgreSQL version: 8.0 Operating system: not applicable Description:spam from yourselves Details: I have been trying unsuccessfully for over 2 months to remove myself from the mailing list. Despite attempting to unsubscibe, sending spam reports to spamcop and notifying the authorities you take no notice of me. What do i need to do to unsubscibe. Please do not tell me to email "unregister YourEmailAddressHere" to [EMAIL PROTECTED])as it does not work. ---(end of broadcast)--- TIP 8: explain analyze is your friend
