Log Problem and bash script

Posted by GvWorker on Server Fault See other posts from Server Fault or by GvWorker
Published on 2010-06-07T08:16:57Z Indexed on 2010/06/07 8:23 UTC
Read the original article Hit count: 200

Filed under:
|
|

Hello Guys,

I have 11 Debian servers running on rackspace cloud hosting. All running VHCS2 for hosting management. 1 server is used for application and 10 are used for only smtp. My question is regarding smtp servers. Each server hosted 1 domain. My problem is when my client use smtp there's a log created in this directory

/var/log/

but within 24 hours drives are full and server refuse all smtp connections. Then I deleted the logs and ran following command to check the disk space.

df -h

but it shows hdd still full and server is still refusing the smtp connections. Then I ran following command to see the truth

du --max-depth=1 -h

It shows the truth. The real disk space used. Then I rebooted the server and now server working fine. But after few hours same situation happened. Then I created the following script.

#!/bin/sh

rm -fr /var/log/*
rm -fr /var/log/apache2/*.log
rm -fr /var/log/apache2/*.log.*
rm -fr /var/log/apache2/users/*
rm -fr /var/log/apache2/backup/*

reboot

It worked for days but after that logs are again filling the hdd.

Now I want the following solutions. If anybody can help me.

  1. When I delete files from server hdd will free up without rebooting
  2. Log should be in specific range. Like a specific size of file where old data overwrite with new data

© Server Fault or respective owner

Related posts about debian

Related posts about logs