Error Log for Crashing Server

Hello,

I have been trying to troubleshoot my dedicated server and have been looking in var/log/

I was told to specifically look in the messages file to find out what was going on before the server went down.

What happens is I can not view the site on the net nor can I get root access, from there I reboot and everthing is good to go for a few days. Then same deal.

The logs say before the reset....any ideas or direction I should go in?

Mar 19 07:20:59 u1528761 xinetd[2896]: START: smtp pid=3175 from=117.199.178.11
Mar 19 07:21:29 u1528761 xinetd[2896]: EXIT: smtp status=1 pid=3175 duration=30(sec)
Mar 19 07:21:58 u1528761 xinetd[2896]: START: smtp pid=3177 from=121.132.176.661
Mar 19 07:21:59 u1528761 xinetd[2896]: START: smtp pid=3179 from=125.57.102.141
Mar 19 07:22:03 u1528761 xinetd[2896]: EXIT: smtp status=0 pid=3179 duration=4(sec)
Mar 19 07:22:03 u1528761 xinetd[2896]: START: smtp pid=3197 from=88.251.266.54

Thank you,


Erik

Erik, what do you have hosted on your server?

Im using plesk and I host about 10 very basic websites...ive used a few plug in applications (blog, chat) and one site has ssl with its own dedicated IP address and finally 1and1 setup a mail server for me so that I could send outgoing mail...I question if setting up that mail server is the problem as the crashing started around that time...

Ive also read that 1and1 recently did an upgrade and that has been causing a number of servers to go offline.

Here are my last logs from last nights crash

Mar 22 03:59:55 u15287612 xinetd[2896]: START: smtp pid=30867 from=92.80.84.5
Mar 22 03:59:55 u15287612 xinetd[2896]: START: smtp pid=30868 from=190.42.21.125
Mar 22 04:00:00 u15287612 xinetd[2896]: START: smtp pid=30870 from=212.33.234.58
Mar 22 04:00:05 u15287612 xinetd[2896]: EXIT: smtp status=1 pid=30866 duration=31(sec)
Mar 22 04:00:05 u15287612 xinetd[2896]: EXIT: smtp status=1 pid=30868 duration=10(sec)
Mar 22 04:00:07 u15287612 dhclient: DHCPREQUEST on eth0 to 74.208.8.250 port 67

Thank you for any help you can provide

Plz do the following steps.

1. take the server back up
2. Reinstall the server ( a latest version )
3. Restore all database
4. after taking backup verify the backup results.

Regards

Have you fixed your issue?

Mysql error log is overwritten whenever the server crashes.

Not sure if it's normal 5.1 behavior?

How to repeat:
Crash the DB voluntarily. I usually do huge FEDERATED queries for that :)

Example log after a crash:

root@dblog-10:/data/mysql# head dblog-10.err

Number of processes running now: 0
070313 15:37:32 mysqld restarted
070313 15:37:32 [Warning] Although a path was specified for the --log-slow-queries option,
log tables are used. To enable logging to files use the --log-output option.
070313 15:37:33 InnoDB: Database was not shut down normally!
InnoDB: Starting crash recovery.
InnoDB: Reading tablespace information from the .ibd files...
InnoDB: Restoring possible half-written data pages from the doublewrite
InnoDB: buffer...
070313 15:37:39 InnoDB: Starting log scan based on checkpoint.......

What do you mean you don't see an error in the log? A crash always leaves something about a trace.

Try to find a pattern in the crashes:
Is the crash x time since the startup? Is it every x time?
Maybe you can find any commands or logging that is printed just before it crashes?