mysql> select VERSION();
+-------------+
| VERSION() |
+-------------+
| 3.23.31-log |
+-------------+
1 row in set (0.00 sec)
I swer I'm running 3.23.31 (with pgcc) but let's forget about "Contratulation
message", I think the error is something related to master.info position.
Thanks for your answers to another questions.
Well, give me some advices to get some stability.
I'll prepare a LVS cluster to avoid problems with unexpected corruption, I gave up to
use mysql with pgcc (I'm afraid about it).
I'm planning use a Bi-Directional replication, 2 servers with --log-slave-update and
other required replication stuff, so I'll put them replicating each other and if some
table get corrupted I'll change the LVS switch and my system will work fine until I
fix the problem.
Next time I see some bug I'll send a report with mysqlbug.
Thanks, Best Regards ... Rodolfo Sikora
=================
>>This table is getting corrupted all the time (twice a week).
>
>>We have only SELECTS and INSERTS with little updates. (we have 140.000
>registers growing faster to 1000000 :))
>
>>CREATE TABLE conta (
> idConta int(8) unsigned NOT NULL auto_increment,
> idDominio int(6) unsigned DEFAULT '0' NOT NULL,
> conta varchar(30) DEFAULT '' NOT NULL,
> senha varchar(32) DEFAULT '' NOT NULL,
> usermount varchar(100) DEFAULT '' NOT NULL,
> userlock int(1) DEFAULT '0' NOT NULL,
> senha2 varchar(255) DEFAULT '' NOT NULL,
> PRIMARY KEY (idConta),
> UNIQUE idDominio (idDominio,conta)
>);
>
>>The worst is tha we have another table with 1.000.000
>inserts/updates/deletes/selects and this table never get corrupted.
>
>>What should I do to prevent corruptions. (every day I run an optimize, check
>table... etc..)
>
>Help us find and fix the bug.... You should work on creating a repeatable
>case of corruption. Start by upgrading to 3.23.33 once it is released.
>3.23.33 will print the trouble query to the error log in the crash handler
>along with the backtrace on x86 Linux, so this should make debugging a lot
>easier.
>
>>I'm running mysql under a NetServer LC2000 (HP) with RAID 5 and 1GB RAM,
>PIII 667/600.
>
>>Shoudl I use -O key_buffer and/or table_cache.
>
>This will not help, although different setting that what you have may
>randomly obscure the bug. You should not rely on this, though.
>
>>What really does FLUSH TABLES? it's ok do it once a day?
>
>It closes all the currently open tables and flushed all the buffers related
>to tables to to the disk. This is needed only when you want to be sure that
>what you have on the disk is current, and is normally not necessary.
>
>>By the way I got this interesting message in my slave log:
>
>>010207 12:47:42 Error reading packet from server: Contratulations! You have
>hit the magic number and can win sweepstakes if you report the bug (0)
>
>If your version is pre-3.23.29, you are hitting the bug that has already been
>fixed. This should not happen in newer versions, if it does, it is a bug,
>just like the message says.
>
>Actually, looking at the message, you **are** running an old version. The
>evidence is the misspeled "Congratulations" in the error message. The fix is
>to upgrade - to 3.23.33 once it gets out, or if you cannot wait, 3.23.31.
>
>--
>MySQL Development Team
> __ ___ ___ ____ __
> / |/ /_ __/ __/ __ \/ / Sasha Pachev <[EMAIL PROTECTED]>
> / /|_/ / // /\ \/ /_/ / /__ MySQL AB, http://www.mysql.com/
>/_/ /_/\_, /___/\___\_\___/ Provo, Utah, USA
> <___/
---------------------------------------------------------------------
Before posting, please check:
http://www.mysql.com/manual.php (the manual)
http://lists.mysql.com/ (the list archive)
To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
