1 <!DOCTYPE html PUBLIC "-//W3C//DTD html 4.01 transitional//EN">
4 <!-- DOCTYPE html PUBLIC "-//W3C//DTD HTML 3.2//EN" -->
8 <META name="generator" content="HTML Tidy, see www.w3.org">
10 <TITLE>PostgreSQL FAQ</TITLE>
13 <BODY bgcolor="#ffffff" text="#000000" link="#ff0000" vlink="#a00000"
15 <H1>Frequently Asked Questions (FAQ) for PostgreSQL</H1>
17 <P>Last updated: Sat Oct 19 22:58:02 EDT 2002</P>
19 <P>Current maintainer: Bruce Momjian (<A href=
20 "mailto:pgman@candle.pha.pa.us">pgman@candle.pha.pa.us</A>)<BR>
23 <P>The most recent version of this document can be viewed at <A
25 "http://www.PostgreSQL.org/docs/faq-english.html">http://www.PostgreSQL.org/docs/faq-english.html</A>.</P>
27 <P>Platform-specific questions are answered at <A href=
28 "http://www.PostgreSQL.org/users-lounge/docs/faq.html">http://www.PostgreSQL.org/users-lounge/docs/faq.html</A>.</P>
31 <H2 align="center">General Questions</H2>
32 <A href="#1.1">1.1</A>) What is PostgreSQL? How is it pronounced?<BR>
33 <A href="#1.2">1.2</A>) What is the copyright on PostgreSQL?<BR>
34 <A href="#1.3">1.3</A>) What Unix platforms does PostgreSQL run
36 <A href="#1.4">1.4</A>) What non-Unix ports are available?<BR>
37 <A href="#1.5">1.5</A>) Where can I get PostgreSQL?<BR>
38 <A href="#1.6">1.6</A>) Where can I get support?<BR>
39 <A href="#1.7">1.7</A>) What is the latest release?<BR>
40 <A href="#1.8">1.8</A>) What documentation is available?<BR>
41 <A href="#1.9">1.9</A>) How do I find out about known bugs or
43 <A href="#1.10">1.10</A>) How can I learn <SMALL>SQL</SMALL>?<BR>
44 <A href="#1.11">1.11</A>) Is PostgreSQL Y2K compliant?<BR>
45 <A href="#1.12">1.12</A>) How do I join the development team?<BR>
46 <A href="#1.13">1.13</A>) How do I submit a bug report?<BR>
47 <A href="#1.14">1.14</A>) How does PostgreSQL compare to other
48 <SMALL>DBMS</SMALL>s?<BR>
49 <A href="#1.15">1.15</A>) How can I financially assist
53 <H2 align="center">User Client Questions</H2>
54 <A href="#2.1">2.1</A>) Are there <SMALL>ODBC</SMALL> drivers for
56 <A href="#2.2">2.2</A>) What tools are available for using
57 PostgreSQL with Web pages?<BR>
58 <A href="#2.3">2.3</A>) Does PostgreSQL have a graphical user
59 interface? A report generator? An embedded query language
61 <A href="#2.4">2.4</A>) What languages are available to
62 communicate with PostgreSQL?<BR>
65 <H2 align="center">Administrative Questions</H2>
66 <A href="#3.1">3.1</A>) How do I install PostgreSQL somewhere other
67 than <I>/usr/local/pgsql</I>?<BR>
68 <A href="#3.2">3.2</A>) When I start <I>postmaster</I>, I get a
69 <I>Bad System Call</I> or core dumped message. Why?<BR>
70 <A href="#3.3">3.3</A>) When I try to start <I>postmaster</I>, I
71 get <I>IpcMemoryCreate</I> errors. Why?<BR>
72 <A href="#3.4">3.4</A>) When I try to start <I>postmaster</I>, I
73 get <I>IpcSemaphoreCreate</I> errors. Why?<BR>
74 <A href="#3.5">3.5</A>) How do I control connections from other
76 <A href="#3.6">3.6</A>) How do I tune the database engine for
77 better performance?<BR>
78 <A href="#3.7">3.7</A>) What debugging features are available?<BR>
79 <A href="#3.8">3.8</A>) Why do I get <I>"Sorry, too many
80 clients"</I> when trying to connect?<BR>
81 <A href="#3.9">3.9</A>) What is in the <I>pgsql_tmp</I>
83 <A href="#3.10">3.10</A>) Why do I need to do a dump and restore
84 to upgrade PostgreSQL releases?<BR>
87 <H2 align="center">Operational Questions</H2>
88 <A href="#4.1">4.1</A>) What is the difference between binary
89 cursors and normal cursors?<BR>
90 <A href="#4.2">4.2</A>) How do I <SMALL>SELECT</SMALL> only the
91 first few rows of a query?<BR>
92 <A href="#4.3">4.3</A>) How do I get a list of tables or other
93 things I can see in <I>psql</I>?<BR>
94 <A href="#4.4">4.4</A>) How do you remove a column from a
96 <A href="#4.5">4.5</A>) What is the maximum size for a row, a
97 table, and a database?<BR>
98 <A href="#4.6">4.6</A>) How much database disk space is required
99 to store data from a typical text file?<BR>
100 <A href="#4.7">4.7</A>) How do I find out what tables, indexes,
101 databases, and users are defined?<BR>
102 <A href="#4.8">4.8</A>) My queries are slow or don't make use of
103 the indexes. Why?<BR>
104 <A href="#4.9">4.9</A>) How do I see how the query optimizer is
105 evaluating my query?<BR>
106 <A href="#4.10">4.10</A>) What is an R-tree index?<BR>
107 <A href="#4.11">4.11</A>) What is the Genetic Query Optimizer?<BR>
108 <A href="#4.12">4.12</A>) How do I perform regular expression
109 searches and case-insensitive regular expression searches? How do I
110 use an index for case-insensitive searches?<BR>
111 <A href="#4.13">4.13</A>) In a query, how do I detect if a field
112 is <SMALL>NULL</SMALL>?<BR>
113 <A href="#4.14">4.14</A>) What is the difference between the
114 various character types?<BR>
115 <A href="#4.15.1">4.15.1</A>) How do I create a
116 serial/auto-incrementing field?<BR>
117 <A href="#4.15.2">4.15.2</A>) How do I get the value of a
118 <SMALL>SERIAL</SMALL> insert?<BR>
119 <A href="#4.15.3">4.15.3</A>) Don't <I>currval()</I> and
120 <I>nextval()</I> lead to a race condition with other users?<BR>
121 <A href="#4.15.4">4.15.4</A>) Why aren't my sequence numbers
122 reused on transaction abort? Why are there gaps in the numbering of
123 my sequence/SERIAL column?<BR>
124 <A href="#4.16">4.16</A>) What is an <SMALL>OID</SMALL>? What is a
125 <SMALL>TID</SMALL>?<BR>
126 <A href="#4.17">4.17</A>) What is the meaning of some of the terms
127 used in PostgreSQL?<BR>
128 <A href="#4.18">4.18</A>) Why do I get the error <I>"ERROR: Memory
129 exhausted in AllocSetAlloc()"</I>?<BR>
130 <A href="#4.19">4.19</A>) How do I tell what PostgreSQL version I
132 <A href="#4.20">4.20</A>) Why does my large-object operations get
133 <I>"invalid large obj descriptor"</I>?<BR>
134 <A href="#4.21">4.21</A>) How do I create a column that will
135 default to the current time?<BR>
136 <A href="#4.22">4.22</A>) Why are my subqueries using
137 <CODE><SMALL>IN</SMALL></CODE> so slow?<BR>
138 <A href="#4.23">4.23</A>) How do I perform an outer join?<BR>
139 <A href="#4.24">4.24</A>) How do I perform queries using multiple
141 <A href="#4.25">4.25</A>) How do I return multiple rows or columns
143 <A href="#4.26">4.26</A>) Why can't I reliably create/drop
144 temporary tables in PL/PgSQL functions?<BR>
145 <A href="#4.27">4.27</A>) What replication options are available?<BR>
146 <A href="#4.28">4.28</A>) What encryption options are available?<BR>
149 <H2 align="center">Extending PostgreSQL</H2>
150 <A href="#5.1">5.1</A>) I wrote a user-defined function. When I run
151 it in <I>psql</I>, why does it dump core?<BR>
152 <A href="#5.2">5.2</A>) How can I contribute some nifty new types
153 and functions to PostgreSQL?<BR>
154 <A href="#5.3">5.3</A>) How do I write a C function to return a
156 <A href="#5.4">5.4</A>) I have changed a source file. Why does the
157 recompile not see the change?<BR>
161 <H2 align="center">General Questions</H2>
163 <H4><A name="1.1">1.1</A>) What is PostgreSQL? How is it pronounced?</H4>
165 <P>PostgreSQL is pronounced <I>Post-Gres-Q-L</I>.</P>
167 <P>PostgreSQL is an enhancement of the POSTGRES database management
168 system, a next-generation <SMALL>DBMS</SMALL> research prototype.
169 While PostgreSQL retains the powerful data model and rich data
170 types of POSTGRES, it replaces the PostQuel query language with an
171 extended subset of <SMALL>SQL</SMALL>. PostgreSQL is free and the
172 complete source is available.</P>
174 <P>PostgreSQL development is performed by a team of
175 developers who all subscribe to the PostgreSQL development mailing
176 list. The current coordinator is Marc G. Fournier (<A href=
177 "mailto:scrappy@PostgreSQL.org">scrappy@PostgreSQL.org</A>). (See
178 section <a href="#1.6">1.6</a> on how to join). This team is now
179 responsible for all development of PostgreSQL.</P>
181 <P>The authors of PostgreSQL 1.01 were Andrew Yu and Jolly Chen.
182 Many others have contributed to the porting, testing, debugging,
183 and enhancement of the code. The original Postgres code, from which
184 PostgreSQL is derived, was the effort of many graduate students,
185 undergraduate students, and staff programmers working under the
186 direction of Professor Michael Stonebraker at the University of
187 California, Berkeley.</P>
189 <P>The original name of the software at Berkeley was Postgres. When
190 <SMALL>SQL</SMALL> functionality was added in 1995, its name was
191 changed to Postgres95. The name was changed at the end of 1996 to
194 <H4><A name="1.2">1.2</A>) What is the copyright on
197 <P>PostgreSQL is subject to the following COPYRIGHT:</P>
199 <P>PostgreSQL Data Base Management System</P>
201 <P>Portions copyright (c) 1996-2002, PostgreSQL Global Development
202 Group Portions Copyright (c) 1994-6 Regents of the University of
205 <P>Permission to use, copy, modify, and distribute this software
206 and its documentation for any purpose, without fee, and without a
207 written agreement is hereby granted, provided that the above
208 copyright notice and this paragraph and the following two
209 paragraphs appear in all copies.</P>
211 <P>IN NO EVENT SHALL THE UNIVERSITY OF CALIFORNIA BE LIABLE TO ANY
212 PARTY FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL
213 DAMAGES, INCLUDING LOST PROFITS, ARISING OUT OF THE USE OF THIS
214 SOFTWARE AND ITS DOCUMENTATION, EVEN IF THE UNIVERSITY OF
215 CALIFORNIA HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.</P>
217 <P>THE UNIVERSITY OF CALIFORNIA SPECIFICALLY DISCLAIMS ANY
218 WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
219 OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE
220 SOFTWARE PROVIDED HEREUNDER IS ON AN "AS IS" BASIS, AND THE
221 UNIVERSITY OF CALIFORNIA HAS NO OBLIGATIONS TO PROVIDE MAINTENANCE,
222 SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.</P>
224 <P>The above is the BSD license, the classic open-source license.
225 It has no restrictions on how the source code may be used. We like
226 it and have no intention of changing it.</P>
228 <H4><A name="1.3">1.3</A>) What Unix platforms does PostgreSQL run
231 <P>In general, a modern Unix-compatible platform should be able to
232 run PostgreSQL. The platforms that had received explicit testing at
233 the time of release are listed in the installation
236 <H4><A name="1.4">1.4</A>) What non-Unix ports are available?</H4>
238 <P><STRONG>Client</STRONG></P>
240 <P>It is possible to compile the <I>libpq</I> C library, psql, and
241 other interfaces and client applications to run on MS Windows platforms.
242 In this case, the client is running on MS Windows, and communicates
243 via TCP/IP to a server running on one of our supported Unix
244 platforms. A file <I>win32.mak</I> is included in the distribution
245 for making a Win32 <I>libpq</I> library and <I>psql</I>. PostgreSQL
246 also communicates with <SMALL>ODBC</SMALL> clients.</P>
248 <P><STRONG>Server</STRONG></P>
250 <P>The database server can run on Windows NT and Win2k using
251 Cygwin, the Cygnus Unix/NT porting library. See
252 <I>pgsql/doc/FAQ_MSWIN</I> in the distribution or the MS Windows FAQ
253 at <A href="http://www.PostgreSQL.org/docs/faq-mswin.html">
254 http://www.PostgreSQL.org/docs/faq-mswin.html</A>.</P>
256 <p>A native port to MS Win NT/2000/XP is currently being worked
259 <H4><A name="1.5">1.5</A>) Where can I get PostgreSQL?</H4>
261 <P>The primary anonymous ftp site for PostgreSQL is <A href=
262 "ftp://ftp.PostgreSQL.org/pub">ftp://ftp.PostgreSQL.org/pub</A>.
263 For mirror sites, see our main web site.</P>
265 <H4><A name="1.6">1.6</A>) Where can I get support?</H4>
267 <P>The main mailing list is: <A href=
268 "mailto:pgsql-general@PostgreSQL.org">pgsql-general@PostgreSQL.org</A>.
269 It is available for discussion of matters pertaining to PostgreSQL.
270 To subscribe, send mail with the following lines in the body (not
271 the subject line):</P>
278 "mailto:pgsql-general-request@PostgreSQL.org">pgsql-general-request@PostgreSQL.org</A>.</P>
280 <P>There is also a digest list available. To subscribe to this
281 list, send email to: <A href=
282 "mailto:pgsql-general-digest-request@PostgreSQL.org">pgsql-general-digest-request@PostgreSQL.org</A>
289 Digests are sent out to members of this list whenever the main list
290 has received around 30k of messages.
292 <P>The bugs mailing list is available. To subscribe to this list,
293 send email to <A href=
294 "mailto:pgsql-bugs-request@PostgreSQL.org">pgsql-bugs-request@PostgreSQL.org</A>
301 There is also a developers discussion mailing list available. To
302 subscribe to this list, send email to <A href=
303 "mailto:pgsql-hackers-request@PostgreSQL.org">pgsql-hackers-request@PostgreSQL.org</A>
310 <P>Additional mailing lists and information about PostgreSQL can be
311 found via the PostgreSQL WWW home page at:</P>
314 <A href="http://www.PostgreSQL.org">http://www.PostgreSQL.org</A>
317 <P>There is also an IRC channel on EFNet, channel
318 <I>#PostgreSQL</I>. I use the Unix command <CODE>irc -c
319 '#PostgreSQL' "$USER" irc.phoenix.net.</CODE></P>
321 <P>A list of commercial support companies is available at <A href=
322 "http://www.PostgreSQL.org/users-lounge/commercial-support.html">http://www.PostgreSQL.org/users-lounge/commercial-support.html</A>.</P>
324 <H4><A name="1.7">1.7</A>) What is the latest release?</H4>
326 <P>The latest release of PostgreSQL is version 7.2.3.</P>
328 <P>We plan to have major releases every four months.</P>
330 <H4><A name="1.8">1.8</A>) What documentation is available?</H4>
332 <P>Several manuals, manual pages, and some small test examples are
333 included in the distribution. See the <I>/doc</I> directory. You
334 can also browse the manual online at <A href=
335 "http://www.PostgreSQL.org/users-lounge/docs/">http://www.PostgreSQL.org/users-lounge/docs/</A>.</P>
337 <P>There are two PostgreSQL books available online at <A href=
338 "http://www.PostgreSQL.org/docs/awbook.html">http://www.PostgreSQL.org/docs/awbook.html</A>
340 "http://www.commandprompt.com/ppbook/">http://www.commandprompt.com/ppbook/</A>.
341 There is a list of PostgreSQL books available for purchase at <A
343 "http://www.ca.PostgreSQL.org/books/">http://www.ca.PostgreSQL.org/books/</A>.
344 There is also a collection of PostgreSQL technical articles at <A
346 "http://techdocs.PostgreSQL.org/">http://techdocs.PostgreSQL.org/</A>.</P>
348 <P><I>psql</I> has some nice \d commands to show information about
349 types, operators, functions, aggregates, etc.</P>
351 <P>Our web site contains even more documentation.</P>
353 <H4><A name="1.9">1.9</A>) How do I find out about known bugs or
354 missing features?</H4>
356 <P>PostgreSQL supports an extended subset of <SMALL>SQL</SMALL>-92.
357 See our <A href="http://developer.PostgreSQL.org/todo.php">TODO</A>
358 list for known bugs, missing features, and future plans.</P>
360 <H4><A name="1.10">1.10</A>) How can I learn
361 <SMALL>SQL</SMALL>?</H4>
363 <P>The PostgreSQL book at <A href=
364 "http://www.PostgreSQL.org/docs/awbook.html">http://www.PostgreSQL.org/docs/awbook.html</A>
365 teaches <SMALL>SQL</SMALL>. There is another PostgreSQL book at <A
367 "http://www.commandprompt.com/ppbook/">http://www.commandprompt.com/ppbook.</A>
368 There is a nice tutorial at <A href=
369 "http://www.intermedia.net/support/sql/sqltut.shtm">http://www.intermedia.net/support/sql/sqltut.shtm,</A>
371 "http://ourworld.compuserve.com/homepages/graeme_birchall/HTM_COOK.HTM">
372 http://ourworld.compuserve.com/homepages/graeme_birchall/HTM_COOK.HTM,</A>
374 "http://sqlcourse.com/">http://sqlcourse.com.</A></P>
376 <P>Another one is "Teach Yourself SQL in 21 Days, Second Edition"
378 "http://members.tripod.com/er4ebus/sql/index.htm">http://members.tripod.com/er4ebus/sql/index.htm</A></P>
380 <P>Many of our users like <I>The Practical SQL Handbook</I>,
381 Bowman, Judith S., et al., Addison-Wesley. Others like <I>The
382 Complete Reference SQL</I>, Groff et al., McGraw-Hill.</P>
384 <H4><A name="1.11">1.11</A>) Is PostgreSQL Y2K compliant?</H4>
386 <P>Yes, we easily handle dates past the year 2000 AD, and before
389 <H4><A name="1.12">1.12</A>) How do I join the development
392 <P>First, download the latest source and read the PostgreSQL
393 Developers documentation on our web site, or in the distribution.
394 Second, subscribe to the <I>pgsql-hackers</I> and
395 <I>pgsql-patches</I> mailing lists. Third, submit high quality
396 patches to pgsql-patches.</P>
398 <P>There are about a dozen people who have commit privileges to the
399 PostgreSQL <SMALL>CVS</SMALL> archive. They each have submitted so
400 many high-quality patches that it was impossible for the existing
401 committers to keep up, and we had confidence that patches they
402 committed were of high quality.</P>
404 <H4><A name="1.13">1.13</A>) How do I submit a bug report?</H4>
406 <P>Please visit the PostgreSQL BugTool page at <A href=
407 "http://www.PostgreSQL.org/bugs/bugs.php">http://www.PostgreSQL.org/bugs/bugs.php</A>,
408 which gives guidelines and directions on how to submit a
411 <P>Also check out our ftp site <A href=
412 "ftp://ftp.PostgreSQL.org/pub">ftp://ftp.PostgreSQL.org/pub</A> to
413 see if there is a more recent PostgreSQL version or patches.</P>
415 <H4><A name="1.14">1.14</A>) How does PostgreSQL compare to other
416 <SMALL>DBMS</SMALL>s?</H4>
418 <P>There are several ways of measuring software: features,
419 performance, reliability, support, and price.</P>
422 <DT><B>Features</B></DT>
424 <DD>PostgreSQL has most features present in large commercial
425 <SMALL>DBMS</SMALL>s, like transactions, subselects, triggers,
426 views, foreign key referential integrity, and sophisticated
427 locking. We have some features they do not have, like
428 user-defined types, inheritance, rules, and multi-version
429 concurrency control to reduce lock contention.<BR>
433 <DT><B>Performance</B></DT>
435 <DD>PostgreSQL has performance similar to other commercial and
436 open source databases. it is faster for some things, slower for
437 others. In comparison to MySQL or leaner database systems, we are
438 slower on inserts/updates because of transaction overhead. Of
439 course, MySQL does not have any of the features mentioned in the
440 <I>Features</I> section above. We are built for reliability and
441 features, though we continue to improve performance in every
442 release. There is an interesting Web page comparing PostgreSQL to
443 MySQL at <A href= "http://openacs.org/why-not-mysql.html">
445 http://openacs.org/why-not-mysql.html</A><BR>
450 <DT><B>Reliability</B></DT>
452 <DD>We realize that a <SMALL>DBMS</SMALL> must be reliable, or it
453 is worthless. We strive to release well-tested, stable code that
454 has a minimum of bugs. Each release has at least one month of
455 beta testing, and our release history shows that we can provide
456 stable, solid releases that are ready for production use. We
457 believe we compare favorably to other database software in this
462 <DT><B>Support</B></DT>
464 <DD>Our mailing lists provide contact with a large group of developers
465 and users to help resolve any problems encountered. While we cannot
466 guarantee a fix, commercial <SMALL>DBMS</SMALL>s do not always
467 supply a fix either. Direct access to developers, the user
468 community, manuals, and the source code often make PostgreSQL
469 support superior to other <SMALL>DBMS</SMALL>s. There is
470 commercial per-incident support available for those who need it.
471 (See <A href="#1.6">FAQ section 1.6</A>.)<BR>
475 <DT><B>Price</B></DT>
477 <DD>We are free for all use, both commercial and non-commercial.
478 You can add our code to your product with no limitations, except
479 those outlined in our BSD-style license stated above.<BR>
484 <H4><A name="1.15">1.15</A>) How can I financially assist
487 <P>PostgreSQL has had a first-class infrastructure since we started
488 in 1996. This is all thanks to Marc Fournier, who has created
489 and managed this infrastructure over the years.</P>
491 <P>Quality infrastructure is very important to an open-source
492 project. It prevents disruptions that can greatly delay forward
493 movement of the project.</P>
495 <P>Of course, this infrastructure is not cheap. There are a variety
496 of monthly and one-time expenses that are required to keep it
497 going. If you or your company has money it can donate to help fund
498 this effort, please go to <A href="http://store.pgsql.com/shopping/">http://store.pgsql.com/shopping/</A>
499 and make a donation.</P>
501 <P>Although the web page mentions PostgreSQL, Inc, the
502 "contributions" item is solely to support the PostgreSQL project
503 and does not fund any specific company. If you prefer, you can also
504 send a check to the contact address.</P>
507 <H2 align="center">User Client Questions</H2>
509 <H4><A name="2.1">2.1</A>) Are there <SMALL>ODBC</SMALL> drivers
512 <P>There are two <SMALL>ODBC</SMALL> drivers available, PsqlODBC
513 and OpenLink <SMALL>ODBC</SMALL>.</P>
515 <P>PsqlODBC is included in the distribution. More information about
516 it can be gotten from <A href=
517 "ftp://ftp.PostgreSQL.org/pub/odbc/">ftp://ftp.PostgreSQL.org/pub/odbc/</A>.</P>
519 <P>OpenLink <SMALL>ODBC</SMALL> can be gotten from <A href=
520 "http://www.openlinksw.com/">http://www.openlinksw.com</A>. It
521 works with their standard <SMALL>ODBC</SMALL> client software so
522 you'll have PostgreSQL <SMALL>ODBC</SMALL> available on every
523 client platform they support (Win, Mac, Unix, VMS).</P>
525 <P>They will probably be selling this product to people who need
526 commercial-quality support, but a freeware version will always be
527 available. Please send questions to <A href=
528 "mailto:postgres95@openlink.co.uk">postgres95@openlink.co.uk</A>.</P>
530 <P>See also the <A href=
531 "http://www.postgresql.org/devel-corner/docs/programmer/odbc.html">ODBC
532 chapter of the Programmer's Guide</A>.</P>
534 <H4><A name="2.2">2.2</A>) What tools are available for using
535 PostgreSQL with Web pages?</H4>
537 <P>A nice introduction to Database-backed Web pages can be seen at:
538 <A href="http://www.webreview.com">http://www.webreview.com</A></P>
540 <P>For Web integration, PHP is an excellent interface. It is at <A
541 href="http://www.php.net">http://www.php.net</A>.</P>
543 <P>For complex cases, many use the Perl interface and CGI.pm or mod_perl.</P>
545 <H4><A name="2.3">2.3</A>) Does PostgreSQL have a graphical user
546 interface? A report generator? An embedded query language
549 <P>We have a nice graphical user interface called PgAccess which can
550 also be used as a report generator. The Web page is
551 <A href="http://www.pgaccess.org/">http://www.pgaccess.org/</A>.</P>
553 <P>We also include <I>ecpg</I>, which is an embedded SQL query
554 language interface for C.</P>
556 <H4><A name="2.4">2.4</A>) What languages are able to communicate with PostgreSQL?</H4>
563 <LI>C++ (libpq++)</LI>
565 <LI>Embedded C (ecpg)</LI>
569 <LI>Perl (DBD::Pg and perl5)</LI>
573 <LI>Python (PyGreSQL)</LI>
575 <LI>TCL (libpgtcl)</LI>
577 <LI>C Easy API (libpgeasy)</LI>
579 <LI>PHP ('pg_' functions, Pear::DB)</LI>
581 <P>Additional interfaces are available at
582 <a href="http://www.PostgreSQL.org/interfaces.html">http://www.PostgreSQL.org/interfaces.html</A>
584 <a href="http://gborg.PostgreSQL.org">http://gborg.PostgreSQL.org</A>.
588 <H2 align="center">Administrative Questions</H2>
590 <H4><A name="3.1">3.1</A>) How do I install PostgreSQL somewhere
591 other than <I>/usr/local/pgsql</I>?</H4>
593 <P>Specify the <I>--prefix</I> option when running
594 <I>configure</I>.</P>
596 <H4><A name="3.2">3.2</A>) When I start <I>postmaster</I>, I get a
597 <I>Bad System Call</I> or core dumped message. Why?</H4>
599 <P>It could be a variety of problems, but first check to see that
600 you have System V extensions installed in your kernel. PostgreSQL
601 requires kernel support for shared memory and semaphores.</P>
603 <H4><A name="3.3">3.3</A>) When I try to start <I>postmaster</I>, I
604 get <I>IpcMemoryCreate</I> errors. Why?</H4>
606 <P>You either do not have shared memory configured properly in your
607 kernel or you need to enlarge the shared memory available in the
608 kernel. The exact amount you need depends on your architecture and
609 how many buffers and backend processes you configure for
610 <I>postmaster</I>. For most systems, with default numbers of
611 buffers and processes, you need a minimum of ~1 MB. See the <A
613 "http://www.postgresql.org/idocs/index.php?kernel-resources.html">PostgreSQL
614 Administrator's Guide</A> for more detailed information about
615 shared memory and semaphores.</P>
617 <H4><A name="3.4">3.4</A>) When I try to start <I>postmaster</I>, I
618 get <I>IpcSemaphoreCreate</I> errors. Why?</H4>
620 <P>If the error message is <I>IpcSemaphoreCreate: semget failed (No
621 space left on device)</I> then your kernel is not configured with
622 enough semaphores. Postgres needs one semaphore per potential
623 backend process. A temporary solution is to start <I>postmaster</I>
624 with a smaller limit on the number of backend processes. Use
625 <I>-N</I> with a parameter less than the default of 32. A more
626 permanent solution is to increase your kernel's
627 <SMALL>SEMMNS</SMALL> and <SMALL>SEMMNI</SMALL> parameters.</P>
629 <P>Inoperative semaphores can also cause crashes during heavy
632 <P>If the error message is something else, you might not have
633 semaphore support configured in your kernel at all. See the
634 PostgreSQL Administrator's Guide for more detailed information
635 about shared memory and semaphores.</P>
637 <H4><A name="3.5">3.5</A>) How do I control connections from other
640 <P>By default, PostgreSQL only allows connections from the local
641 machine using Unix domain sockets. Other machines will not be able
642 to connect unless you add the <I>-i</I> flag to <I>postmaster</I>,
643 <B>and</B> enable host-based authentication by modifying the file
644 <I>$PGDATA/pg_hba.conf</I> accordingly. This will allow TCP/IP
647 <H4><A name="3.6">3.6</A>) How do I tune the database engine for
648 better performance?</H4>
650 <P>Certainly, indexes can speed up queries. The
651 <SMALL>EXPLAIN</SMALL> command allows you to see how PostgreSQL is
652 interpreting your query, and which indexes are being used.</P>
654 <P>If you are doing many <SMALL>INSERTs</SMALL>, consider doing
655 them in a large batch using the <SMALL>COPY</SMALL> command. This
656 is much faster than individual <SMALL>INSERTS.</SMALL> Second,
657 statements not in a <SMALL>BEGIN WORK/COMMIT</SMALL> transaction
658 block are considered to be in their own transaction. Consider
659 performing several statements in a single transaction block. This
660 reduces the transaction overhead. Also, consider dropping and
661 recreating indexes when making large data changes.</P>
663 <P>There are several tuning options. You can disable <I>fsync()</I>
664 by starting <I>postmaster</I> with a <I>-o -F</I> option. This will
665 prevent <I>fsync()</I>s from flushing to disk after every
668 <P>You can also use the <I>postmaster</I> <I>-B</I> option to
669 increase the number of shared memory buffers used by the backend
670 processes. If you make this parameter too high, the
671 <I>postmaster</I> may not start because you have exceeded your
672 kernel's limit on shared memory space. Each buffer is 8K and the
673 default is 64 buffers.</P>
675 <P>You can also use the backend <I>-S</I> option to increase the
676 maximum amount of memory used by the backend process for temporary
677 sorts. The <I>-S</I> value is measured in kilobytes, and the
678 default is 512 (i.e. 512K).</P>
680 <P>You can also use the <SMALL>CLUSTER</SMALL> command to group
681 data in tables to match an index. See the <SMALL>CLUSTER</SMALL>
682 manual page for more details.</P>
684 <H4><A name="3.7">3.7</A>) What debugging features are
687 <P>PostgreSQL has several features that report status information
688 that can be valuable for debugging purposes.</P>
690 <P>First, by running <I>configure</I> with the --enable-cassert
691 option, many <I>assert()</I>s monitor the progress of the backend
692 and halt the program when something unexpected occurs.</P>
694 <P>Both <I>postmaster</I> and <I>postgres</I> have several debug
695 options available. First, whenever you start <I>postmaster</I>,
696 make sure you send the standard output and error to a log file,
700 ./bin/postmaster >server.log 2>&1 &
703 <P>This will put a server.log file in the top-level PostgreSQL
704 directory. This file contains useful information about problems or
705 errors encountered by the server. <I>Postmaster</I> has a <I>-d</I>
706 option that allows even more detailed information to be reported.
707 The <I>-d</I> option takes a number that specifies the debug level.
708 Be warned that high debug level values generate large log
711 <P>If <I>postmaster</I> is not running, you can actually run the
712 <I>postgres</I> backend from the command line, and type your
713 <SMALL>SQL</SMALL> statement directly. This is recommended
714 <B>only</B> for debugging purposes. Note that a newline terminates
715 the query, not a semicolon. If you have compiled with debugging
716 symbols, you can use a debugger to see what is happening. Because
717 the backend was not started from <I>postmaster</I>, it is not
718 running in an identical environment and locking/backend interaction
719 problems may not be duplicated.</P>
721 <P>If <I>postmaster</I> is running, start <I>psql</I> in one
722 window, then find the <SMALL>PID</SMALL> of the <I>postgres</I>
723 process used by <I>psql</I>. Use a debugger to attach to the
724 <I>postgres</I> <SMALL>PID.</SMALL> You can set breakpoints in the
725 debugger and issue queries from <I>psql</I>. If you are debugging
726 <I>postgres</I> startup, you can set PGOPTIONS="-W n", then start
727 <I>psql</I>. This will cause startup to delay for <I>n</I> seconds
728 so you can attach to the process with the debugger, set any
729 breakpoints, and continue through the startup sequence.</P>
731 <P>The <I>postgres</I> program has <I>-s, -A</I>, and <I>-t</I>
732 options that can be very useful for debugging and performance
735 <P>You can also compile with profiling to see what functions are
736 taking execution time. The backend profile files will be deposited
737 in the <I>pgsql/data/base/dbname</I> directory. The client profile
738 file will be put in the client's current directory. Linux requires
739 a compile with <I>-DLINUX_PROFILE</I> for proper profiling.</P>
741 <H4><A name="3.8">3.8</A>) Why do I get <I>"Sorry, too many
742 clients"</I> when trying to connect?</H4>
744 <P>You need to increase <I>postmaster</I>'s limit on how many
745 concurrent backend processes it can start.</P>
747 <P>The default limit is 32 processes. You can increase it by
748 restarting <I>postmaster</I> with a suitable <I>-N</I> value or
749 modifying <I>postgresql.conf</I>.</P>
751 <P>Note that if you make <I>-N</I> larger than 32, you must also
752 increase <I>-B</I> beyond its default of 64; <I>-B</I> must be at
753 least twice <I>-N</I>, and probably should be more than that for
754 best performance. For large numbers of backend processes, you are
755 also likely to find that you need to increase various Unix kernel
756 configuration parameters. Things to check include the maximum size
757 of shared memory blocks, <SMALL>SHMMAX;</SMALL> the maximum number
758 of semaphores, <SMALL>SEMMNS</SMALL> and <SMALL>SEMMNI;</SMALL> the
759 maximum number of processes, <SMALL>NPROC;</SMALL> the maximum
760 number of processes per user, <SMALL>MAXUPRC;</SMALL> and the
761 maximum number of open files, <SMALL>NFILE</SMALL> and
762 <SMALL>NINODE.</SMALL> The reason that PostgreSQL has a limit on
763 the number of allowed backend processes is so your system won't run
764 out of resources.</P>
766 <P>In PostgreSQL versions prior to 6.5, the maximum number of
767 backends was 64, and changing it required a rebuild after altering
768 the MaxBackendId constant in
769 <I>include/storage/sinvaladt.h</I>.</P>
771 <H4><A name="3.9">3.9</A>) What are the <I>pgsql_tmp</I>
774 <P>They are temporary files generated by the query executor. For
775 example, if a sort needs to be done to satisfy an <SMALL>ORDER
776 BY,</SMALL> and the sort requires more space than the backend's
777 <I>-S</I> parameter allows, then temporary files are created to
778 hold the extra data.</P>
780 <P>The temporary files are usually deleted automatically, but might
781 remain if a backend crashes during a sort. A stop and restart of the
782 <I>postmaster</I> will remove files from those directories.</P>
784 <H4><A name="3.10">3.10</A>) Why do I need to do a dump and restore
785 to upgrade between major PostgreSQL releases?</H4>
787 <P>The PostgreSQL team makes only small changes between minor releases,
788 so upgrading from 7.2 to 7.2.1 does not require a dump and restore.
789 However, major releases (e.g. from 7.2 to 7.3) often change the internal
790 format of system tables and data files. These changes are often complex,
791 so we don't maintain backward compatability for data files. A dump outputs
792 data in a generic format that can then be loaded in using the new internal
795 <P>In releases where the on-disk format does not change, the
796 <I>pg_upgrade</I> script can be used to upgrade without a dump/restore.
797 The release notes mention whether <I>pg_upgrade</I> is available for the
802 <H2 align="center">Operational Questions</H2>
804 <H4><A name="4.1">4.1</A>) What is the difference between binary
805 cursors and normal cursors?</H4>
807 <P>See the <SMALL>DECLARE</SMALL> manual page for a
810 <H4><A name="4.2">4.2</A>) How do I <SMALL>SELECT</SMALL> only the
811 first few rows of a query?</H4>
813 <P>See the <SMALL>FETCH</SMALL> manual page, or use
814 <SMALL>SELECT</SMALL> ... <SMALL>LIMIT</SMALL>....</P>
816 <P>The entire query may have to be evaluated, even if you only want
817 the first few rows. Consider using a query that has an <SMALL>ORDER
818 BY.</SMALL> If there is an index that matches the <SMALL>ORDER
819 BY</SMALL>, PostgreSQL may be able to evaluate only the first few
820 records requested, or the entire query may have to be evaluated
821 until the desired rows have been generated.</P>
823 <H4><A name="4.3">4.3</A>) How do I get a list of tables or other
824 things I can see in <I>psql</I>?</H4>
826 <P>You can read the source code for <I>psql</I> in file
827 <I>pgsql/src/bin/psql/describe.c</I>. It contains
828 <SMALL>SQL</SMALL> commands that generate the output for psql's
829 backslash commands. You can also start <I>psql</I> with the
830 <I>-E</I> option so it will print out the queries it uses to
831 execute the commands you give.</P>
833 <H4><A name="4.4">4.4</A>) How do you remove a column from a
836 <P>This functionality was added in release 7.3 with
837 <SMALL>ALTER TABLE DROP COLUMN</SMALL>. In earlier versions,
841 LOCK TABLE old_table;
842 SELECT ... -- select all columns but the one you want to remove
845 DROP TABLE old_table;
846 ALTER TABLE new_table RENAME TO old_table;
850 <H4><A name="4.5">4.5</A>) What is the maximum size for a row, a
851 table, and a database?</H4>
853 <P>These are the limits:</P>
855 Maximum size for a database? unlimited (1 TB databases exist)
856 Maximum size for a table? 16 TB
857 Maximum size for a row? 1.6TB
858 Maximum size for a field? 1 GB
859 Maximum number of rows in a table? unlimited
860 Maximum number of columns in a table? 250-1600 depending on column types
861 Maximum number of indexes on a table? unlimited
864 Of course, these are not actually unlimited, but limited to
865 available disk space and memory/swap space. Performance may suffer
866 when these values get unusually large.
868 <P>The maximum table size of 16 TB does not require large file
869 support from the operating system. Large tables are stored as
870 multiple 1 GB files so file system size limits are not
873 <P>The maximum table size and maximum number of columns can be
874 increased if the default block size is increased to 32k.</P>
876 <H4><A name="4.6">4.6</A>) How much database disk space is required
877 to store data from a typical text file?</H4>
879 <P>A PostgreSQL database may require up to five times the disk
880 space to store data from a text file.</P>
882 <P>As an example, consider a file of 100,000 lines with an integer
883 and text description on each line. Suppose the text string
884 avergages twenty bytes in length. The flat file would be 2.8 MB.
885 The size of the PostgreSQL database file containing this data can
886 be estimated as 6.4 MB:</P>
888 36 bytes: each row header (approximate)
889 24 bytes: one int field and one text field
890 + 4 bytes: pointer on page to tuple
891 ----------------------------------------
894 The data page size in PostgreSQL is 8192 bytes (8 KB), so:
897 ------------------- = 128 rows per database page (rounded down)
901 -------------------- = 782 database pages (rounded up)
904 782 database pages * 8192 bytes per page = 6,406,144 bytes (6.4 MB)
907 <P>Indexes do not require as much overhead, but do contain the data
908 that is being indexed, so they can be large also.</P>
910 <P><SMALL>NULL</SMALL>s are stored in bitmaps, so they
911 use very little space.</P>
913 <H4><A name="4.7">4.7</A>) How do I find out what tables, indexes,
914 databases, and users are defined?</H4>
916 <P><I>psql</I> has a variety of backslash commands to show such
917 information. Use \? to see them. There are also system tables
918 beginning with <I>pg_</I> that describe these too. Also, <I>psql
919 -l</I> will list all databases.</P>
921 <P>Also try the file <I>pgsql/src/tutorial/syscat.source</I>. It
922 illustrates many of the <SMALL>SELECT</SMALL>s needed to get
923 information from the database system tables.</P>
925 <H4><A name="4.8">4.8</A>) My queries are slow or don't make use of
926 the indexes. Why?</H4>
927 Indexes are not automatically used by every query. Indexes are only
928 used if the table is larger than a minimum size, and the query
929 selects only a small percentage of the rows in the table. This is
930 because the random disk access caused by an index scan can be
931 slower than a straight read through the table, or sequential scan.
933 <P>To determine if an index should be used, PostgreSQL must have
934 statistics about the table. These statistics are collected using
935 <SMALL>VACUUM ANALYZE</SMALL>, or simply <SMALL>ANALYZE</SMALL>.
936 Using statistics, the optimizer knows how many rows are in the
937 table, and can better determine if indexes should be used.
938 Statistics are also valuable in determining optimal join order and
939 join methods. Statistics collection should be performed
940 periodically as the contents of the table change.</P>
942 <P>Indexes are normally not used for <SMALL>ORDER BY</SMALL> or to
943 perform joins. A sequential scan followed by an explicit sort is
944 usually faster than an index scan of a large table.</P>
945 However, <SMALL>LIMIT</SMALL> combined with <SMALL>ORDER BY</SMALL>
946 often will use an index because only a small portion of the table
947 is returned. In fact, though MAX() and MIN() don't use indexes,
948 it is possible to retrieve such values using an index with ORDER BY
953 ORDER BY col [ DESC ]
957 <P>When using wild-card operators such as <SMALL>LIKE</SMALL> or
958 <I>~</I>, indexes can only be used in certain circumstances:
960 <LI>The beginning of the search string must be anchored to the start
961 of the string, i.e.:</LI>
963 <LI><SMALL>LIKE</SMALL> patterns must not start with <I>%.</I></LI>
964 <LI><I>~</I> (regular expression) patterns must start with
967 <LI>The search string can not start with a character class,
969 <LI>Case-insensitive searches such as <SMALL>ILIKE</SMALL> and
970 <I>~*</I> do not utilise indexes. Instead, use functional
971 indexes, which are described in section <a href="#4.12">4.12</a>.</LI>
972 <LI>The default <I>C</I> locale must be used during
977 <H4><A name="4.9">4.9</A>) How do I see how the query optimizer is
978 evaluating my query?</H4>
980 <P>See the <SMALL>EXPLAIN</SMALL> manual page.</P>
982 <H4><A name="4.10">4.10</A>) What is an R-tree index?</H4>
984 <P>An R-tree index is used for indexing spatial data. A hash index
985 can't handle range searches. A B-tree index only handles range
986 searches in a single dimension. R-trees can handle
987 multi-dimensional data. For example, if an R-tree index can be
988 built on an attribute of type <I>point</I>, the system can more
989 efficiently answer queries such as "select all points within a
990 bounding rectangle."</P>
992 <P>The canonical paper that describes the original R-tree design
995 <P>Guttman, A. "R-trees: A Dynamic Index Structure for Spatial
996 Searching." Proceedings of the 1984 ACM SIGMOD Int'l Conf on Mgmt
999 <P>You can also find this paper in Stonebraker's "Readings in
1000 Database Systems".</P>
1002 <P>Built-in R-trees can handle polygons and boxes. In theory,
1003 R-trees can be extended to handle higher number of dimensions. In
1004 practice, extending R-trees requires a bit of work and we don't
1005 currently have any documentation on how to do it.</P>
1007 <H4><A name="4.11">4.11</A>) What is the Genetic Query
1010 <P>The <SMALL>GEQO</SMALL> module speeds query optimization when
1011 joining many tables by means of a Genetic Algorithm (GA). It allows
1012 the handling of large join queries through nonexhaustive
1015 <H4><A name="4.12">4.12</A>) How do I perform regular expression
1016 searches and case-insensitive regular expression searches? How do I
1017 use an index for case-insensitive searches?</H4>
1019 <P>The <I>~</I> operator does regular expression matching, and
1020 <I>~*</I> does case-insensitive regular expression matching. The
1021 case-insensitive variant of <SMALL>LIKE</SMALL> is called
1022 <SMALL>ILIKE</SMALL>.</P>
1024 <P>Case-insensitive equality comparisons are normally expressed
1029 WHERE lower(col) = 'abc';
1032 This will not use an standard index. However, if you create a
1033 functional index, it will be used:
1035 CREATE INDEX tabindex ON tab (lower(col));
1038 <H4><A name="4.13">4.13</A>) In a query, how do I detect if a field
1039 is <SMALL>NULL</SMALL>?</H4>
1041 <P>You test the column with <SMALL>IS NULL</SMALL> and <SMALL>IS
1042 NOT NULL</SMALL>.</P>
1044 <H4><A name="4.14">4.14</A>) What is the difference between the
1045 various character types?</H4>
1047 Type Internal Name Notes
1048 --------------------------------------------------
1049 "char" char 1 character
1050 CHAR(n) bpchar blank padded to the specified fixed length
1051 VARCHAR(n) varchar size specifies maximum length, no padding
1052 TEXT text no specific upper limit on length
1053 BYTEA bytea variable-length byte array (null-byte safe)
1056 <P>You will see the internal name when examining system catalogs
1057 and in some error messages.</P>
1059 <P>The last four types above are "varlena" types (i.e., the first
1060 four bytes on disk are the length, followed by the data). Thus the
1061 actual space used is slightly greater than the declared size.
1062 However, these data types are also subject to compression or being
1063 stored out-of-line by <SMALL>TOAST</SMALL>, so the space on disk
1064 might also be less than expected.</P>
1066 <P><SMALL>CHAR(n)</SMALL> is best when storing strings that are
1067 usually the same length. <SMALL>VARCHAR(n)</SMALL> is best when
1068 storing variable-length strings but it limits how long a string can
1069 be. <SMALL>TEXT</SMALL> is for strings of unlimited length, maximum
1070 1 gigabyte. <SMALL>BYTEA</SMALL> is for storing binary data,
1071 particularly values that include <SMALL>NULL</SMALL> bytes. These
1072 types have similar performance characteristics.</P>
1074 <H4><A name="4.15.1">4.15.1</A>) How do I create a
1075 serial/auto-incrementing field?</H4>
1077 <P>PostgreSQL supports a <SMALL>SERIAL</SMALL> data type. It
1078 auto-creates a sequence and index on the column. For example,
1081 CREATE TABLE person (
1087 is automatically translated into this:
1089 CREATE SEQUENCE person_id_seq;
1090 CREATE TABLE person (
1091 id INT4 NOT NULL DEFAULT nextval('person_id_seq'),
1094 CREATE UNIQUE INDEX person_id_key ON person ( id );
1097 See the <I>create_sequence</I> manual page for more information
1098 about sequences. You can also use each row's <I>OID</I> field as a
1099 unique value. However, if you need to dump and reload the database,
1100 you need to use <I>pg_dump</I>'s <I>-o</I> option or <SMALL>COPY
1101 WITH OIDS</SMALL> option to preserve the <SMALL>OID</SMALL>s.
1103 <H4><A name="4.15.2">4.15.2</A>) How do I get the value of a
1104 <SMALL>SERIAL</SMALL> insert?</H4>
1106 <P>One approach is to retrieve the next <SMALL>SERIAL</SMALL> value
1107 from the sequence object with the <I>nextval()</I> function
1108 <I>before</I> inserting and then insert it explicitly. Using the
1109 example table in <A href="#4.15.1">4.15.1</A>, an example in a
1110 pseudo-language would look like this:</P>
1112 new_id = execute("SELECT nextval('person_id_seq')");
1113 execute("INSERT INTO person (id, name) VALUES (new_id, 'Blaise Pascal')");
1116 You would then also have the new value stored in
1117 <CODE>new_id</CODE> for use in other queries (e.g., as a foreign
1118 key to the <CODE>person</CODE> table). Note that the name of the
1119 automatically created <SMALL>SEQUENCE</SMALL> object will be named
1120 <<I>table</I>>_<<I>serialcolumn</I>>_<I>seq</I>, where
1121 <I>table</I> and <I>serialcolumn</I> are the names of your table
1122 and your <SMALL>SERIAL</SMALL> column, respectively.
1124 <P>Alternatively, you could retrieve the assigned
1125 <SMALL>SERIAL</SMALL> value with the <I>currval()</I> function
1126 <I>after</I> it was inserted by default, e.g.,</P>
1128 execute("INSERT INTO person (name) VALUES ('Blaise Pascal')");
1129 new_id = execute("SELECT currval('person_id_seq')");
1132 Finally, you could use the <A href="#4.16"><SMALL>OID</SMALL></A>
1133 returned from the <SMALL>INSERT</SMALL> statement to look up the
1134 default value, though this is probably the least portable approach.
1135 In Perl, using DBI with Edmund Mergl's DBD::Pg module, the oid
1136 value is made available via <I>$sth->{pg_oid_status}</I> after
1137 <I>$sth->execute()</I>.
1139 <H4><A name="4.15.3">4.15.3</A>) Don't <I>currval()</I> and
1140 <I>nextval()</I> lead to a race condition with other users?</H4>
1142 <P>No. <I>currval()</I> returns the current value assigned by your
1143 backend, not by all users.</P>
1145 <H4><A name="4.15.4">4.15.4</A>) Why aren't my sequence numbers
1146 reused on transaction abort? Why are there gaps in the numbering of
1147 my sequence/SERIAL column?</H4>
1149 <P>To improve concurrency, sequence values are given out to running
1150 transactions as needed and are not locked until the transaction
1151 completes. This causes gaps in numbering from aborted
1154 <H4><A name="4.16">4.16</A>) What is an <SMALL>OID</SMALL>? What is
1155 a <SMALL>TID</SMALL>?</H4>
1157 <P><SMALL>OID</SMALL>s are PostgreSQL's answer to unique row ids.
1158 Every row that is created in PostgreSQL gets a unique
1159 <SMALL>OID</SMALL>. All <SMALL>OID</SMALL>s generated during
1160 <I>initdb</I> are less than 16384 (from
1161 <I>include/access/transam.h</I>). All user-created
1162 <SMALL>OID</SMALL>s are equal to or greater than this. By default,
1163 all these <SMALL>OID</SMALL>s are unique not only within a table or
1164 database, but unique within the entire PostgreSQL installation.</P>
1166 <P>PostgreSQL uses <SMALL>OID</SMALL>s in its internal system
1167 tables to link rows between tables. These <SMALL>OID</SMALL>s can
1168 be used to identify specific user rows and used in joins. It is
1169 recommended you use column type <SMALL>OID</SMALL> to store
1170 <SMALL>OID</SMALL> values. You can create an index on the
1171 <SMALL>OID</SMALL> field for faster access.</P>
1173 <P>O<SMALL>ID</SMALL>s are assigned to all new rows from a central
1174 area that is used by all databases. If you want to change the
1175 <SMALL>OID</SMALL> to something else, or if you want to make a copy
1176 of the table, with the original <SMALL>OID</SMALL>s, there is no
1177 reason you can't do it:</P>
1179 CREATE TABLE new_table(old_oid oid, mycol int);
1180 SELECT old_oid, mycol INTO new FROM old;
1181 COPY new TO '/tmp/pgtable';
1183 COPY new WITH OIDS FROM '/tmp/pgtable';
1186 CREATE TABLE new_table (mycol int);
1187 INSERT INTO new_table (oid, mycol) SELECT oid, mycol FROM old_table;
1189 <P>O<SMALL>ID</SMALL>s are stored as 4-byte integers, and will
1190 overflow at 4 billion. No one has reported this ever happening, and
1191 we plan to have the limit removed before anyone does.</P>
1193 <P>T<SMALL>ID</SMALL>s are used to identify specific physical rows
1194 with block and offset values. T<SMALL>ID</SMALL>s change after rows
1195 are modified or reloaded. They are used by index entries to point
1196 to physical rows.</P>
1198 <H4><A name="4.17">4.17</A>) What is the meaning of some of the
1199 terms used in PostgreSQL?</H4>
1201 <P>Some of the source code and older documentation use terms that
1202 have more common usage. Here are some:</P>
1205 <LI>table, relation, class</LI>
1207 <LI>row, record, tuple</LI>
1209 <LI>column, field, attribute</LI>
1211 <LI>retrieve, select</LI>
1213 <LI>replace, update</LI>
1215 <LI>append, insert</LI>
1217 <LI><SMALL>OID</SMALL>, serial value</LI>
1219 <LI>portal, cursor</LI>
1221 <LI>range variable, table name, table alias</LI>
1224 <P>A list of general database terms can be found at: <A href=
1225 "http://hea-www.harvard.edu/MST/simul/software/docs/pkgs/pgsql/glossary/glossary.html">http://hea-www.harvard.edu/MST/simul/software/docs/pkgs/pgsql/glossary/glossary.html</A></P>
1227 <H4><A name="4.18">4.18</A>) Why do I get the error <I>"ERROR:
1228 Memory exhausted in AllocSetAlloc()"</I>?</H4>
1230 <P>You probably have run out of virtual memory on your system,
1231 or your kernel has a low limit for certain resources. Try this
1232 before starting <I>postmaster</I>:</P>
1238 Depending on your shell, only one of these may succeed, but it will
1239 set your process data segment limit much higher and perhaps allow
1240 the query to complete. This command applies to the current process,
1241 and all subprocesses created after the command is run. If you are
1242 having a problem with the <SMALL>SQL</SMALL> client because the
1243 backend is returning too much data, try it before starting the
1246 <H4><A name="4.19">4.19</A>) How do I tell what PostgreSQL version
1249 <P>From <I>psql</I>, type <CODE>SELECT version();</CODE></P>
1251 <H4><A name="4.20">4.20</A>) Why does my large-object operations
1252 get <I>"invalid large obj descriptor"</I>?</H4>
1254 <P>You need to put <CODE>BEGIN WORK</CODE> and <CODE>COMMIT</CODE>
1255 around any use of a large object handle, that is, surrounding
1256 <CODE>lo_open</CODE> ... <CODE>lo_close.</CODE></P>
1258 <P>Currently PostgreSQL enforces the rule by closing large object
1259 handles at transaction commit. So the first attempt to do anything
1260 with the handle will draw <I>invalid large obj descriptor</I>. So
1261 code that used to work (at least most of the time) will now
1262 generate that error message if you fail to use a transaction.</P>
1264 <P>If you are using a client interface like <SMALL>ODBC</SMALL> you
1265 may need to set <CODE>auto-commit off.</CODE></P>
1267 <H4><A name="4.21">4.21</A>) How do I create a column that will
1268 default to the current time?</H4>
1270 <P>Use <I>CURRENT_TIMESTAMP</I>:</P>
1272 <CODE>CREATE TABLE test (x int, modtime timestamp DEFAULT CURRENT_TIMESTAMP );
1276 <H4><A name="4.22">4.22</A>) Why are my subqueries using
1277 <CODE><SMALL>IN</SMALL></CODE> so slow?</H4>
1279 <P>Currently, we join subqueries to outer queries by sequentially
1280 scanning the result of the subquery for each row of the outer
1281 query. If the subquery returns only a few rows and the outer query
1282 returns many rows, <CODE><SMALL>IN</SMALL></CODE> is fastest. To
1283 speed up other queries, replace <CODE>IN</CODE> with
1284 <CODE>EXISTS</CODE>:</P>
1287 WHERE col IN (SELECT subcol FROM subtab);
1292 WHERE EXISTS (SELECT subcol FROM subtab WHERE subcol = col);
1295 For this to be fast, <CODE>subcol</CODE> should be an indexed column.
1296 We hope to fix this limitation in a future release.
1298 <H4><A name="4.23">4.23</A>) How do I perform an outer join?</H4>
1300 <P>PostgreSQL supports outer joins using the SQL standard syntax.
1301 Here are two examples:</P>
1304 FROM t1 LEFT OUTER JOIN t2 ON (t1.col = t2.col);
1309 FROM t1 LEFT OUTER JOIN t2 USING (col);
1312 <P>These identical queries join t1.col to t2.col, and also return
1313 any unjoined rows in t1 (those with no match in t2). A
1314 <SMALL>RIGHT</SMALL> join would add unjoined rows of t2. A
1315 <SMALL>FULL</SMALL> join would return the matched rows plus all
1316 unjoined rows from t1 and t2. The word <SMALL>OUTER</SMALL> is
1317 optional and is assumed in <SMALL>LEFT</SMALL>,
1318 <SMALL>RIGHT</SMALL>, and <SMALL>FULL</SMALL> joins. Ordinary joins
1319 are called <SMALL>INNER</SMALL> joins.</P>
1321 <P>In previous releases, outer joins can be simulated using
1322 <SMALL>UNION</SMALL> and <SMALL>NOT IN</SMALL>. For example, when
1323 joining <I>tab1</I> and <I>tab2</I>, the following query does an
1324 <I>outer</I> join of the two tables:<BR>
1328 SELECT tab1.col1, tab2.col2
1330 WHERE tab1.col1 = tab2.col1
1332 SELECT tab1.col1, NULL
1334 WHERE tab1.col1 NOT IN (SELECT tab2.col1 FROM tab2)
1338 <H4><A name="4.24">4.24</A>) How do I perform queries using
1339 multiple databases?</H4>
1341 <P>There is no way to query a database other than the current one.
1342 Because PostgreSQL loads database-specific system catalogs, it is
1343 uncertain how a cross-database query should even behave.</P>
1345 <P><I>/contrib/dblink</I> allows cross-database queries using
1346 function calls. Of course, a client can make simultaneous
1347 connections to different databases and merge the results on the
1350 <H4><A name="4.25">4.25</A>) How do I return multiple rows or
1351 columns from a function?</H4>
1353 <P>You can return result sets from PL/pgSQL functions using
1354 <I>refcursors</I>. See <A href=
1355 "http://www.PostgreSQL.org/idocs/index.php?plpgsql-cursors.html">
1356 http://www.PostgreSQL.org/idocs/index.php?plpgsql-cursors.html</A>,
1357 section 23.7.3.3.</P>
1359 <H4><A name="4.26">4.26</A>) Why can't I reliably create/drop
1360 temporary tables in PL/PgSQL functions?</H4>
1361 <P>PL/PgSQL caches function contents, and an unfortunate side effect
1362 is that if a PL/PgSQL function accesses a temporary table, and that
1363 table is later dropped and recreated, and the function called
1364 again, the function will fail because the cached function contents
1365 still point to the old temporary table. The solution is to use
1366 <SMALL>EXECUTE</SMALL> for temporary table access in PL/PgSQL. This
1367 will cause the query to be reparsed every time.</P>
1369 <H4><A name="4.27">4.27</A>) What replication options are available?
1371 <P>There are several master/slave replication options available.
1372 These allow only the master to make database changes and the slave
1373 can only do database reads. The bottom of <a
1374 href="http://gborg.PostgreSQL.org/genpage?replication_research">
1375 http://gborg.PostgreSQL.org/genpage?replication_research</a> lists
1376 them. A multi-master replication solution is being worked on at <a
1377 href="http://gborg.PostgreSQL.org/project/pgreplication/projdisplay.php">http://gborg.PostgreSQL.org/project/pgreplication/projdisplay.php</a>.</P>
1379 <H4><A name="4.28">4.28</A>) What encryption options are available?
1382 <LI><I>/contrib/pgcrypto</I> contains many encryption functions for
1383 use in <SMALL>SQL</SMALL> queries.</LI>
1384 <LI>The only way to encrypt transmission from the client to the
1385 server is by using <I>hostssl</I> in <I>pg_hba.conf</I>.</LI>
1386 <LI>Database user passwords are automatically encrypted when stored
1387 in version 7.3. In previous versions, you must enable the option
1388 <i>PASSWORD_ENCRYPTION</i> in <i>postgresql.conf</i>.</LI>
1389 <LI>The server can run using an encrypted file system.</LI>
1394 <H2 align="center">Extending PostgreSQL</H2>
1396 <H4><A name="5.1">5.1</A>) I wrote a user-defined function. When I
1397 run it in <I>psql</I>, why does it dump core?</H4>
1399 <P>The problem could be a number of things. Try testing your
1400 user-defined function in a stand-alone test program first.</P>
1402 <H4><A name="5.2">5.2</A>) How can I contribute some nifty new
1403 types and functions to PostgreSQL?</H4>
1405 <P>Send your extensions to the <I>pgsql-hackers</I> mailing list,
1406 and they will eventually end up in the <I>contrib/</I>
1409 <H4><A name="5.3">5.3</A>) How do I write a C function to return a
1412 <P>In versions of PostgreSQL beginning with 7.3, table-returning
1413 functions are fully supported in C, PL/PgSQL, and SQL. See the
1414 Programmer's Guide for more information. An example of a
1415 table-returning function defined in C can be found in
1416 contrib/tablefunc.</P>
1418 <H4><A name="5.4">5.4</A>) I have changed a source file. Why does
1419 the recompile not see the change?</H4>
1421 <P>The <I>Makefiles</I> do not have the proper dependencies for
1422 include files. You have to do a <I>make clean</I> and then another
1423 <I>make</I>. If you are using <SMALL>GCC</SMALL> you can use the
1424 <I>--enable-depend</I> option of <I>configure</I> to have the
1425 compiler compute the dependencies automatically.</P>