Which version of James?
2.1.3
Yes there was 30 ESTABLISHED connections with 26 in the SMTP serviceHave you run netstat to check the active connections?
Is it possible that your service is simply being hit by lots of connections today from internet worms?
Not really the system works fine today
> in the smtp log hundreds of messages like the following: > 13/04/04 10:58:24 DEBUG smtpserver: Calling reset() default Worker #40 > 13/04/04 10:58:24 DEBUG smtpserver: Calling reset() default Worker #40 > ... > Any idea if this can be the source of the problem Actually, that could be helpful. That message is emitted by the Watchdog for that connection when it is reset. The reset happens on one of two conditions: - a new command is entered - a certain amount (20K default) of data is received during processing of the DATA command
The messages started as:
smtpserver-2004-04-13-10-53.log:13/04/04 10:53:27 DEBUG smtpserver: Watchdog default Worker #40 has time to sleep 360000
smtpserver-2004-04-13-10-53.log:13/04/04 10:53:28 DEBUG smtpserver: Calling reset() default Worker #40
smtpserver-2004-04-13-10-53.log:13/04/04 10:53:29 DEBUG smtpserver: Calling reset() default Worker #40
......... 34576 times the same message
smtpserver-2004-04-13-10-53.log:13/04/04 10:58:59 DEBUG smtpserver: Calling stop() default Worker #40
smtpserver-2004-04-13-10-53.log:13/04/04 10:58:59 DEBUG smtpserver: Calling disposeWatchdog() default Worker #40
smtpserver-2004-04-13-10-53.log:13/04/04 10:58:59 DEBUG smtpserver: Watchdog default Worker #40 is exiting run().
then it resume with normal operation
What I find is that when I start the server the number of connection in the connection log is smaller but closer to the
number os connections with NETSTAT but after 2-3 hours the number of connections in the connection log starts to grow
little by little and passes the number of connections in NETSTAT
Like now I have in the connection log
13/04/04 16:41:39 DEBUG connections: Releasing one connection, leaving a total of 62
13/04/04 16:41:39 DEBUG connections: Returning a org.apache.james.util.connection.ServerConnection$ClientConnectionRunner to the pool
13/04/04 16:41:39 DEBUG connections: Retrieving a org.apache.james.util.connection.ServerConnection$ClientConnectionRunner from the pool
13/04/04 16:41:39 DEBUG connections: Adding one connection for a total of 63
13/04/04 16:41:39 DEBUG connections: Starting connection on 200.229.128.62:25 from 200.77.160.40:3805
13/04/04 16:41:39 DEBUG connections: Ending connection on 200.229.128.62:25 from 200.247.6.3:1409
13/04/04 16:41:39 DEBUG connections: Releasing one connection, leaving a total of 62
13/04/04 16:41:39 DEBUG connections: Returning a org.apache.james.util.connection.ServerConnection$ClientConnectionRunner to the pool
13/04/04 16:41:39 DEBUG connections: Ending connection on 200.229.128.62:25 from 67.64.81.156:10326
13/04/04 16:41:39 DEBUG connections: Releasing one connection, leaving a total of 61
13/04/04 16:41:39 DEBUG connections: Returning a org.apache.james.util.connection.ServerConnection$ClientConnectionRunner to the pool
But NETSTAT says:
[EMAIL PROTECTED] logs]# netstat -n | grep ESTA | wc -l
41
[EMAIL PROTECTED] logs]# date
Tue Apr 13 16:41:50 BRT 2004
So after 2 days my server starts to refuse connections because it thinks that there are more than 200 simultaneous connections what is
not correct.
Any idea on how to debug more information about that?
Thanks
Either someone is entering 100s of lines, they are sending you a huge message in a short period of time, or there is a really weird bug in the code. You showed 7 entries at 10:58:24, so that would imply 140K in one second. Are all of the 100s of entries at the same time? What is the rate at which they are logged? I'm leaning towards someone sending a blank command, which would not be logged but would result in the watchdog being reset. --- Noel --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
