Regression test instructions and analysis. The PostgreSQL regression tests are a comprehensive set of tests for the SQL implementation embedded in PostgreSQL. They test standard SQL operations as well as the extended capabilities of PostgreSQL. There are two different ways in which the regression tests can be run: the "sequential" method and the "parallel" method. The sequential method runs each test script in turn, whereas the parallel method starts up multiple server processes to run groups of tests in parallel. Parallel testing gives confidence that interprocess communication and locking are working correctly. Another key difference is that the sequential test procedure uses an already-installed postmaster, whereas the parallel test procedure tests a system that has been built but not yet installed. (The parallel test script actually does an installation into a temporary directory and fires up a private postmaster therein.) Some properly installed and fully functional PostgreSQL installations can "fail" some of these regression tests due to artifacts of floating point representation and time zone support. The tests are currently evaluated using a simple diff comparison against the outputs generated on a reference system, so the results are sensitive to small system differences. When a test is reported as "failed", always examine the differences between expected and actual results; you may well find that the differences are not significant. The regression tests were originally developed by Jolly Chen and Andrew Yu, and were extensively revised/repackaged by Marc Fournier and Thomas Lockhart. From PostgreSQL v6.1 onward the regression tests are current for every official release. The regression testing notes below assume the following (except where noted): Commands are Unix-compatible. See note below. Defaults are used except where noted. User postgres is the Postgres superuser. The source path is /usr/src/pgsql (other paths are possible). The runtime path is /usr/local/pgsql (other paths are possible). Normally, the regression tests should be run as the postgres user since the 'src/test/regress' directory and sub-directories are owned by the postgres user. If you run the regression test as another user the 'src/test/regress' directory tree must be writeable by that user. It was formerly necessary to run the postmaster with system time zone set to PST, but this is no longer required. You can run the regression tests under your normal postmaster configuration. The test script will set the PGTZ environment variable to ensure that timezone-dependent tests produce the expected results. However, your system must provide library support for the PST8PDT time zone, or the timezone-dependent tests will fail. To verify that your machine does have this support, type the following: setenv TZ PST8PDT date The "date" command above should have returned the current system time in the PST8PDT time zone. If the PST8PDT database is not available, then your system may have returned the time in GMT. If the PST8PDT time zone is not available, you can set the time zone rules explicitly: setenv PGTZ PST8PDT7,M04.01.0,M10.05.03 The directory layout for the regression test area is: Kerberos Directory Description Directory Description input Source files that are converted using make all into some of the .sql files in the sql subdirectory. output Source files that are converted using make all into .out files in the expected subdirectory. sql .sql files used to perform the regression tests. expected .out files that represent what we expect the results to look like. results .out files that contain what the results actually look like. Also used as temporary storage for table copy testing. tmp_check Temporary installation created by parallel testing script. Commands were tested on RedHat Linux version 4.2 using the bash shell. Except where noted, they will probably work on most systems. Commands like ps and tar vary wildly on what options you should use on each platform. Use common sense before typing in these commands. Prepare the files needed for the regression test with: cd /usr/src/pgsql/src/test/regress gmake clean gmake all You can skip "gmake clean" if this is the first time you are running the tests. This step compiles a C program with PostgreSQL extension functions into a shared library. Localized SQL scripts and output-comparison files are also created for the tests that need them. The localization replaces macros in the source files with absolute pathnames and user names. If you intend to use the "sequential" test procedure, which tests an already-installed postmaster, be sure that the postmaster is running. If it isn't already running, start the postmaster in an available window by typing postmaster or start the postmaster daemon running in the background by typing cd nohup postmaster > regress.log 2>&1 & The latter is probably preferable, since the regression test log will be quite lengthy (60K or so, in Postgres 7.0) and you might want to review it for clues if things go wrong. Do not run postmaster from the root account. Run the regression tests. For a sequential test, type cd /usr/src/pgsql/src/test/regress gmake runtest For a parallel test, type cd /usr/src/pgsql/src/test/regress gmake runcheck The sequential test just runs the test scripts using your already-running postmaster. The parallel test will perform a complete installation of Postgres into a temporary directory, start a private postmaster therein, and then run the test scripts. Finally it will kill the private postmaster (but the temporary directory isn't removed automatically). You should get on the screen (and also written to file ./regress.out) a series of statements stating which tests passed and which tests failed. Please note that it can be normal for some of the tests to "fail" due to platform-specific variations. See the next section for details on determining whether a "failure" is significant. Some of the tests, notably "numeric", can take a while, especially on slower platforms. Have patience. After running the tests and examining the results, type cd /usr/src/pgsql/src/test/regress gmake clean to recover the temporary disk space used by the tests. If you ran a sequential test, also type dropdb regression The actual outputs of the regression tests are in files in the ./results directory. The test script uses diff to compare each output file against the reference outputs stored in the ./expected directory. Any differences are saved for your inspection in ./regression.diffs. (Or you can run diff yourself, if you prefer.) The files might not compare exactly. The test script will report any difference as a "failure", but the difference might be due to small cross-system differences in error message wording, math library behavior, etc. "Failures" of this type do not indicate a problem with Postgres. Thus, it is necessary to examine the actual differences for each "failed" test to determine whether there is really a problem. The following paragraphs attempt to provide some guidance in determining whether a difference is significant or not. Some of the regression tests involve intentional invalid input values. Error messages can come from either the Postgres code or from the host platform system routines. In the latter case, the messages may vary between platforms, but should reflect similar information. These differences in messages will result in a "failed" regression test which can be validated by inspection. Most of the date and time results are dependent on timezone environment. The reference files are generated for timezone PST8PDT (Berkeley, California) and there will be apparent failures if the tests are not run with that timezone setting. The regression test driver sets environment variable PGTZ to PST8PDT to ensure proper results. Some of the queries in the "timestamp" test will fail if you run the test on the day of a daylight-savings time changeover, or the day before or after one. These queries assume that the intervals between midnight yesterday, midnight today and midnight tomorrow are exactly twenty-four hours ... which is wrong if daylight-savings time went into or out of effect meanwhile. There appear to be some systems which do not accept the recommended syntax for explicitly setting the local time zone rules; you may need to use a different PGTZ setting on such machines. Some systems using older timezone libraries fail to apply daylight-savings corrections to pre-1970 dates, causing pre-1970 PDT times to be displayed in PST instead. This will result in localized differences in the test results. Some of the tests involve computing 64-bit (float8) numbers from table columns. Differences in results involving mathematical functions of float8 columns have been observed. The float8 and geometry tests are particularly prone to small differences across platforms. Human eyeball comparison is needed to determine the real significance of these differences which are usually 10 places to the right of the decimal point. Some systems signal errors from pow() and exp() differently from the mechanism expected by the current Postgres code. Several of the tests involve operations on geographic date about the Oakland/Berkley CA street map. The map data is expressed as polygons whose vertices are represented as pairs of float8 numbers (decimal latitude and longitude). Initially, some tables are created and loaded with geographic data, then some views are created which join two tables using the polygon intersection operator (##), then a select is done on the view. When comparing the results from different platforms, differences occur in the 2nd or 3rd place to the right of the decimal point. The SQL statements where these problems occur are the following: QUERY: SELECT * from street; QUERY: SELECT * from iexit; There is at least one case in the "random" test script that is intended to produce random results. This causes random to fail the regression test once in a while (perhaps once in every five to ten trials). Typing diff results/random.out expected/random.out should produce only one or a few lines of differences. You need not worry unless the random test always fails in repeated attempts. (On the other hand, if the random test is never reported to fail even in many trials of the regress tests, you probably should worry.) The ./expected/*.out files were adapted from the original monolithic expected.input file provided by Jolly Chen et al. Newer versions of these files generated on various development machines have been substituted after careful (?) inspection. Many of the development machines are running a Unix OS variant (FreeBSD, Linux, etc) on Ix86 hardware. The original expected.input file was created on a SPARC Solaris 2.4 system using the postgres5-1.02a5.tar.gz source tree. It was compared with a file created on an I386 Solaris 2.4 system and the differences were only in the floating point polygons in the 3rd digit to the right of the decimal point. The original sample.regress.out file was from the postgres-1.01 release constructed by Jolly Chen. It may have been created on a DEC ALPHA machine as the Makefile.global in the postgres-1.01 release has PORTNAME=alpha. Since some of the tests inherently produce platform-specific results, we have provided a way to supply platform-specific result comparison files. Frequently, the same variation applies to multiple platforms; rather than supplying a separate comparison file for every platform, there is a mapping file that defines which comparison file to use. So, to eliminate bogus test "failures" for a particular platform, you must choose or make a variant result file, and then add a line to the mapping file, which is "resultmap". Each line in the mapping file is of the form testname/platformnamepattern=comparisonfilename The test name is just the name of the particular regression test module. The platform name pattern is a pattern in the style of expr(1) (that is, a regular expression with an implicit ^ anchor at the start). It is matched against the platform name as printed by config.guess. The comparison file name is the name of the substitute result comparison file. For example: the int2 regress test includes a deliberate entry of a value that is too large to fit in int2. The specific error message that is produced is platform-dependent; our reference platform emits ERROR: pg_atoi: error reading "100000": Numerical result out of range but a fair number of other Unix platforms emit ERROR: pg_atoi: error reading "100000": Result too large Therefore, we provide a variant comparison file, int2-too-large.out, that includes this spelling of the error message. To silence the bogus "failure" message on HPPA platforms, resultmap includes int2/hppa=int2-too-large which will trigger on any machine for which config.guess's output begins with 'hppa'. Other lines in resultmap select the variant comparison file for other platforms where it's appropriate.