Hi

If you log into MySQL on the command line and enter

mysql>SHOW PROCESSLIST ;

this should show you what is running. You may find a runaway query.

also check the temp disk space while the system is stalling - large joins
can use huge amounts of disk space. If MySql  runs out of disk space it
tries to run the query in the space it has, and queuing queries pile up,
giving the situation you describe.

You can also KILL processes found by SHOW PROCESSLIST without needing to do
a full restart.

The default number of max_Max_used_connections is 100

mysql>SHOW PROCESSLIST ;

will show the maximum opened, if it is 100 then you are getting connections
refused which may give the error messages you are getting.

Also check that there are no problems with file permissions for the
temporary tables

HTH

Peter







-----Original Message-----
From: Sebastian Stadtlich [mailto:[EMAIL PROTECTED]
Sent: 25 June 2003 19:25
To: [EMAIL PROTECTED]
Subject: Help : my Server dies : "Got an error reading communication
packets"


Hello All

I'm desperate. Since two days my Server crashed several times.
This is what happens :
The Webserver ist not responding any more. If i'm lucky i can still log
in.
System load ist 90+(!!!) rapidly increasing.
Top shows  a LOT of mysql processes all eating CPU like hell. There are
also
a lot of httpd processes, but that's like it should be. If i stop
apache,
mysql-process minimise too. When i start Apache back up,  Mysql starts
eating
all my CPU again. When i restart mysql, problem goes away and everything
runs smooth for
a while.
All i can find in error files is that there are a lot of errors in the
form :
030625 19:53:57  Aborted connection 23394 to db: 'abc' user: 'xyz' host:
`localhost' (Got timeout reading communication packets)
030625 19:53:57  Aborted connection 23457 to db: 'abc' user: 'xyz' host:
`localhost' (Got timeout reading communication packets)
030625 19:53:57  Aborted connection 23458 to db: 'abc' user: 'xyz' host:
`localhost' (Got timeout reading communication packets)
at least Ten per Second.

All logs are not showing anything unusual.

To me the situation locks like the Socket is blocked and only a restart
can free the mysql Connection.
Can anyone advice? What can i Do? I am absolutely clueless.....
If i can provide anything else to solve the issue? just mail me.


Please Help  :-(
Sebastian

Here are some Facts about the System

The System runs with stupid SUSE Linux 7.2 and is a dual 1200 P3 with
1GB of Memory
and plenty of Disc Space. Since Suse uses a different Version System
than everyone else
in the world i can hardly tell which version runs.

xyz:/etc # mysql -V
mysql  Ver 11.15 Distrib 3.23.37, for suse-linux (i686)

# rpm -q mysql
mysql-3.23.37-58
Apache apache_1.3.27
PHP 4.3.2

my.cnf
---------------------8<--------------------------
port            = 3306
socket          = /var/lib/mysql/mysql.sock


[mysqld]
port            = 3306
socket          = /var/lib/mysql/mysql.sock
skip-locking
set-variable    = key_buffer=96M
set-variable    = max_allowed_packet=8M
set-variable    = table_cache=512
set-variable    = sort_buffer=8192K
set-variable    = net_buffer_length=64K
set-variable    = myisam_sort_buffer_size=64M
set-variable    = max_connections=200

log-bin
server-id       = 1

[mysqldump]
quick
set-variable    = max_allowed_packet=16M

[mysql]
no-auto-rehash
--------------------->8--------------------------












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