main.file_contents failure due to missing INFO_SRC in source tarball

[David Smid]

Compiling from source on Scientific Linux 6.2 with these settings

tar -zxf Percona-Server-5-1.5.23-rel25.3.tar.gz

rm -rf debug release tmp
mkdir -p rpm/{BUILD,RPMS,SOURCES,SPECS,SRPMS} tmp
# Build debug binaries
mkdir debug
(
  cd debug
  cmake ../Percona-Server-5.5.23-rel25.3 -DBUILD_CONFIG=mysql_release -DCMAKE_BUILD_TYPE=Debug -DCMAKE_INSTALL_PREFIX:PATH=/opt/percona-5.5.23 -DINSTALL_LAYOUT:STRING=STANDALONE -DMYSQL_DATADIR:PATH=/home/mysql/percona-5.5/data -DWITH_READLINE:BOOL=ON -DEXTRA_CHARSETS:STRING=all -DWITH_EMBEDDED_SERVER=OFF

  make VERBOSE=1
  make test
)
(
cp tests-ignored.txt debug/mysql-test/
cp tests-ignored.txt Percona-Server-5.5.23-rel25.3/mysql-test/
cd debug/mysql-test/
./mysql-test-run.pl --skip-test-list=tests-ignored.txt

And the tests fail (and the error code does not tell what file broke the build)

[19:13:12]: [Step 1/3] main.events_scheduling [ pass ] 6179
[19:13:13]: [Step 1/3] main.execution_constants [ pass ] 251
[19:13:13]: [Step 1/3] main.explain [ pass ] 529
[19:13:14]: [Step 1/3] main.file_contents [ fail ]
[19:13:14]: [Step 1/3] Test ended at 2012-05-25 19:13:14
[19:13:14]: [Step 1/3] CURRENT_TEST: main.file_contents
[19:13:14]: [Step 1/3] --- /home/blue-01/buildAgent/work/690e7e30210d5d15/Percona-Server-5.5.23-rel25.3/mysql-test/r/file_contents.result 2012-04-25 00:56:36.000000000 +0300
[19:13:14]: [Step 1/3] +++ /home/blue-01/buildAgent/work/690e7e30210d5d15/Percona-Server-5.5.23-rel25.3/mysql-test/r/file_contents.reject 2012-05-25 20:13:13.922701058 +0300
[19:13:14]: [Step 1/3] @@ -1,6 +1,6 @@
[19:13:14]: [Step 1/3]
[19:13:14]: [Step 1/3] Checking 'INFO_SRC' and 'INFO_BIN'
[19:13:14]: [Step 1/3] -INFO_SRC: Found MySQL version number / Found BZR revision id
[19:13:14]: [Step 1/3] +INFO_SRC: Found MySQL version number / No line 'revision-id: .....'
[19:13:14]: [Step 1/3] INFO_BIN: Found 'Compiler ... used' line / Found 'Feature flags' line
[19:13:14]: [Step 1/3]
[19:13:14]: [Step 1/3] End of tests
[19:13:14]: [Step 1/3] mysqltest: Result length mismatch
[19:13:14]: [Step 1/3] - saving '/home/blue-01/buildAgent/work/690e7e30210d5d15/debug/mysql-test/var/log/main.file_contents/' to '/home/blue-01/buildAgent/work/690e7e30210d5d15/debug/mysql-test/var/log/main.file_contents/'
[19:13:14]: [Step 1/3] Only 1331 of 2384 completed.
[19:13:14]: [Step 1/3] mysql-test-run: *** ERROR: Not all tests completed
[19:13:14]: [Step 1/3] Process exited with code 1
[19:13:14]: [Step 1/3] Step Compile source code (Command Line) failed

[Stewart Smith (stewart)]

On Tue, 29 May 2012 07:27:18 -0000, David Smid <email address hidden> wrote:
> [19:13:14]: [Step 1/3] main.file_contents [
> fail ]

Can you please retry with current trunk? I *think* we had a problem like
this at some point in the past but it is since fixed....

It also may be due to source tree layout, the file_contents test is a
bit strange... does building from the bzr tree work and pass tests?

--
Stewart Smith

[Raghavendra D Prabhu (raghavendra-prabhu)]

Tested with PS 5.5.27-28.1 tar.gz.

./mysql-test-run.pl --vardir=$HOME/mysql --suite=main,binlog,innodb
Logging: /tmp/Percona-Server-5.5.27-rel28.1/mysql-test/mysql-test-run.pl --vardir=/home/raghu/mysql --suite=main,binlog,innodb
120919 6:08:05 [Note] Flashcache bypass: disabled
120919 6:08:05 [Note] Flashcache setup error is : open flash device failed

120919 6:08:05 [Note] Plugin 'FEDERATED' is disabled.
MySQL Version 5.5.27
Checking supported features...
 - skipping ndbcluster
 - SSL connections supported
 - binaries are debug compiled
Using suites: main,binlog,innodb
Collecting tests...
 - adding combinations for binlog
vardir: /home/raghu/mysql
Checking leftover processes...
Removing old var directory...
Creating var directory '/home/raghu/mysql'...
Installing system database...
Using server port 57166

==============================================================================

TEST RESULT TIME (ms) or COMMENT
--------------------------------------------------------------------------

worker[1] Using MTR_BUILD_THREAD 300, with reserved ports 13000..13009
binlog.binlog_multi_engine [ skipped ] No ndbcluster tests(--skip-ndbcluster)
main.alter_table-big [ skipped ] Test needs 'big-test' option
main.count_distinct3 [ skipped ] Test needs 'big-test' option
main.create-big [ skipped ] Test needs 'big-test' option
main.events_stress [ skipped ] Test needs 'big-test' option
main.events_time_zone [ skipped ] Test needs 'big-test' option
main.information_schema-big [ skipped ] Test needs 'big-test' option
main.merge-big [ skipped ] Test needs 'big-test' option
main.mysqlbinlog_row_big [ skipped ] Test needs 'big-test' option
main.ssl-big [ skipped ] Test needs 'big-test' option
main.type_newdecimal-big [ skipped ] Test needs 'big-test' option
main.variables-big [ skipped ] Test needs 'big-test' option
main.archive-big [ disabled ] Bug#11817185 2011-03-10 Anitha Disabled since this leads to timeout on Solaris Sparc
main.log_tables-big [ disabled ] Bug#11756699 2010-11-15 mattiasj report already exists
main.lowercase_table3 [ disabled ] Bug#11762269 2010-06-30 alik main.lowercase_table3 on Mac OSX
main.mysql_embedded [ disabled ] Bug#12561297 2011-05-14 Anitha Dependent on PB2 changes - eventum#41836
main.read_many_rows_innodb [ disabled ] Bug#11748886 2010-11-15 mattiasj report already exists
main.sum_distinct-big [ disabled ] Bug#11764126 2010-11-15 mattiasj was not tested
binlog.binlog_spurious_ddl_errors [ disabled ] BUG#11761680 2010-06-03 alik binlog_spurious_ddl_errors.test fails, thus disabled
binlog.binlog_truncate_innodb [ disabled ] BUG#11764459 2010-10-20 anitha Originally disabled due to BUG#42643. Product bug fixed, but test changes needed
main.percona_log_slow_admin_statements-config [ pass ] ...

[Laurynas Biveinis (laurynas-biveinis)]

Raghu -

Can you add --force --max-test-fail=0 parameters to mysql-test-run invocation and report any other failures to other bugs? Let's keep this one for the file_contents failure.

The main.file_contents failure looks like a packaging issue: INFO_SRC is missing from the Percona Server source tarball.

[David Smid (datasmid)]

Still a problem in 5.5.27-rel29.0

[14:02:17]: main.events_scheduling [ pass ] 6114
[14:02:18]: main.execution_constants [ pass ] 450
[14:02:19]: main.explain [ pass ] 301
[14:02:19]: main.file_contents [ fail ]
[14:02:19]: Test ended at 2012-10-16 14:02:19
[14:02:19]: CURRENT_TEST: main.file_contents
[14:02:19]: --- /home/brown-02_q_nl_eng_tc_blue01/buildAgent/work/f56cead8f56ffa2/Percona-Server-5.5.27-rel29.0/mysql-test/r/file_contents.result 2012-10-11 21:00:57.000000000 +0300
[14:02:19]: +++ /home/brown-02_q_nl_eng_tc_blue01/buildAgent/work/f56cead8f56ffa2/Percona-Server-5.5.27-rel29.0/mysql-test/r/file_contents.reject 2012-10-16 15:02:19.626251348 +0300
[14:02:19]: @@ -1,6 +1,6 @@
[14:02:19]:
[14:02:19]: Checking 'INFO_SRC' and 'INFO_BIN'
[14:02:19]: -INFO_SRC: Found MySQL version number / Found BZR revision id
[14:02:19]: +INFO_SRC: Found MySQL version number / No line 'revision-id: .....'
[14:02:19]: INFO_BIN: Found 'Compiler ... used' line / Found 'Feature flags' line
[14:02:19]:
[14:02:19]: End of tests
[14:02:19]: mysqltest: Result length mismatch
[14:02:19]: - saving '/home/brown-02_q_nl_eng_tc_blue01/buildAgent/work/f56cead8f56ffa2/debug/mysql-test/var/log/main.file_contents/' to '/home/brown-02_q_nl_eng_tc_blue01/buildAgent/work/f56cead8f56ffa2/debug/mysql-test/var/log/main.file_contents/'
[14:02:19]: Only 1350 of 2429 completed.
[14:02:19]: mysql-test-run: *** ERROR: Not all tests completed

[Stewart Smith (stewart)]

This should be fixed in the tree re-org to make Percona Server look more like a branch of MySQL and very much fix how source tarballs are generated.

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PS-1918