Hi,
I can confirm this on Linux i386 with official binaries on MyISAM
tables:
mysql check table blast;
DELETE FROM ++---+--+--+
| Table | Op| Msg_type | Msg_text |
++---+--+--+
| test_mmo.blast | check |
Hi,
actually I know about current problem in 3.23.49a on linux at least -
there's something weird with the cache I think. I've posted some reports
with my ideas to the bugs list already. For you:
I can get rid of such problems whenever they appear by doing :
mysqladmin flush-tables
They
the 'truncate table' trick.
Note the the script 'mysqlhotcopy' only copies the header of the .MYI
file..
Regards,
Monty
BTW: And here's status of my current knowledges ... ;)
http://lists.mysql.com/cgi-ez/ezmlm-cgi?9:mss:11886:200204:jjojehdjpfcddnaofmmb
--
Martin Mokrejs
[EMAIL PROTECTED
: --extended-insert breaks import of some tables through pipe on
similar thread in the bugs list.
Note the last one succeeeds - this is a current workaround for me.
Martin MOKREJS wrote:
On Fri, 15 Mar 2002, Michael Widenius wrote:
Martin And if user sets net_buffer_length=80M, then happens what
chown -R mysql.mysql /var/lib/mysql
That's because there's user=root defined in safe_mysqld.
I just put there mysql and it works fine. I don't know how you, but I
don't like the .pid and .err file mixed with databases in the datadir.
MYSQL_UNIX_PORT=${MYSQL_UNIX_PORT:-/tmp/mysql.sock}
Hi,
use "mysqladmin variables" command to see current settings. I guess
your mysqld does not read the config file you have edited. ;-) It
happened to me me also few days ago. Remember mysqld looks for
/etc/my.cnf and $DATADIR/var/my.cnf if I remember well. Maybe put the
path to config file just