Hello,

I'm attempting to compile 4.0.20 and I'm receiving during testing (and sometimes compiling). My config is as follows:

Dual Pentium III 1Ghz
SCSI Raid 10 (~90G)
Red Hat Linux release 7.3 (Valhalla) (not customized)

When I configure with --prefix=/usr/local/mysql4 and then 'make'. The make appears to completes fine. When I run 'make test', I receive the following error:

[...]
Starting Tests

TEST RESULT
------------------------------------------
[... all passed or skipped up to this point]
count_distinct2 [ pass ]
create [ pass ]
ctype_cp1251 [ pass ]
ERROR: /home/bwright/mysql-4.0.20/mysql-test/var/run/master.pid was not created in 30 seconds; Aborting
make: *** [test] Error 1


I've determined this error has something to do with ctype-latin1_de translation not being compiled in and the test script attempting to test for it anyway. If I attempt to compile in the latin1_de translation as --with-extra-charsets=latin1_de (which I assumed would let the test pass), then it gives me the following error during compile:

make[2]: *** No rule to make target `ctype-latin1_de.lo', needed by `bmove_upp.lo'. Stop

This prevents mysql from compiling. In fact, --with-extra-charsets=complex also gives a similar error above with big5 charset. Am I missing some translation parts of the source?

I need to know how to do one of the following:

1) Have the tests complete without the extra latin1_de character set compiled in (how to get rid of or prevent that test or any other similar character set test failure). I need to see all the tests 'pass' or 'skipped' so I know the compile has successfully worked (and the database is functioning).
2) Alternately, how to compile in the ctype-latin1_de character set (or any other character set that causes a test failure) to allow the test to complete successfully.
3) Any other method to allow the tests to complete using release 4.0.20.


If anyone has any suggestions, I'm open. Can I simply remove the ctype-latin1_de test from the mysql-test/t directory? Are there other tests that will fail as a result of this same problem? Is there something else I'm missing (perhaps a language OS component)?

Please refrain from suggesting the following:

1) Use the RPMs. Why not? Using the RPM(s) prevents me from tweaking the code or applying patches if necessary. Plus, I can't compile the binary with the location where it traditionally lives.
2) Install 4.1 alpha. Why not? It's alpha. When it becomes a general release, I'll consider it. Yes, I know this problem is fixed in the 4.1 alpha release. But, I can't use 4.1 until it has been officially released.


Any help is appreciated.  Thanks.

--
Brian Wright ([EMAIL PROTECTED])

--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:    http://lists.mysql.com/[EMAIL PROTECTED]



Reply via email to