This will be a little bigger thread (and I hope it will grow with time) about working with hard disk drives. All about filesystem creation, filesystem health checks and other stuff for HDD.
[Read more…]
Remove logs older than X days
The simplest way to remove files older than $NUMBER of days..
Dry run (files will be NOT deleted, you will see a list with files older than 100 days)
1 | find -mtime +100 | less |
Another dry run, see what would be delete:
1 2 3 | find -mtime +600 | xargs echo rm rm /var/log/squid3/access.log.0 rm /var/log/squid3/access.log.2 |
Finally if you decided to remove the files – the final command to delete them:
1 | find -mtime +600 | xargs rm |
Bash history timestamps
Maybe you know this problem, if more than one user has root access to a server you have very little chances to control who was it. The only hook could be date and time. Default if you issue the “history” command you see a command number and the operation.
Put the next line in your .bashrc and you will see date and time in your history:
1 2 | # Bash Timestamps export HISTTIMEFORMAT="%F %T " |
Finely to make it work execute:
1 | source .bashrc |
Your new history format:
1 2 3 4 5 | 536 2011-05-31 22:06:44 less /var/log/syslog 537 2011-05-31 22:06:46 ll 538 2011-05-31 22:06:54 vim rssh.pl 539 2011-05-31 22:07:07 perl rssh.pl 540 2011-05-31 22:07:16 ss -lnp |
SSH aliases
I think you now this problem, you have a lot different hosts with different ssh ports.
It takes to much time to write the full ssh commands and recognize all of your ports:
Easy solution – use aliases:
1 2 3 4 5 6 | vim .ssh/config # Add this Host piper HostName piper User root Port 1022 |
1 | ssh piper |
‘Argument list too long’
root@web2:$ rm pe-warn-*.bz2
-bash: /bin/rm: Argument list too long
This peoblem happens when you are trying to delete too many files in a directory at the same time – it seems rm has special limits …
To solve the problem:
Use:
1 | find . -name 'pe-warn-*.bz2' | xargs rm |
or
1 | find . -name "pe-warn-*.bz2" -delete |