I ran into a problem where erlang R15 had some segfaults i could not determine. In the #erlang Channel on Freenode i was told to try the newest R16B release of erlang. Instead of building R16B myself with configure and make, i searched for a deb repository and found one at erlang-solutions.com.
Check out this site, there is a howto for debian and rpm based distributions:
Every now and then i get some mails from my hosters monitoring that something at my server is not running. As always i have a look and this time stumbled across this problem.
The VM could not be started, which resulted in this log output.
Starting container ...
vzquota : (warning) Incorrect quota shutdown for id 101, recalculating disk usage
vzquota : (error) quota check : lstat `sess_ehnttr7nr4kp2tcri7982fr0h2': Input/output error
vzquota on failed 
TASK ERROR: command 'vzctl start 101' failed: exit code 60
Searching for ‘exit code 60′ brought me on the right tracks. The solution was to use this commands:
root@server ~ # vzquota off 101
vzquota : (error) Quota is not running for id 101
vzquota : (warning) Repairing quota: it was incorrectly marked as running for id 101
root@server ~ # vzquota on 101
The problem seems to be a “corrupted” quota file, because the VM has not been shut down correctly and the quota file still assumed the VM was running.
I recently had to mount a system with sshfs and ended up with a error.
Aim was to mount the folder ‘backups’ in the home directory of the user ‘bob’.
/usr/bin/sshfs firstname.lastname@example.org:/backups/ /var/my_backups
email@example.com:/backups/: No such file or directory
The error seems confusing at first, but does say exactly what is wrong.
Solution would be using this target path:
/usr/bin/sshfs firstname.lastname@example.org:backups/ /var/my_backups
As you can see, the first slash of ‘backups’ is gone. This is correct, because after the ‘:’, a absolute path can be given to sshfs for mounting. In my case, i wanted to work relative to the user home directory, so i simply hat do ommit the first slash.
Hope this helps someone.
I am using this cronjob for monitoring updates and SMART values.
# Monitoring cronjob
33 3 * * 0 root apt-get update ; apt-get -q --simulate upgrade
34 3 * * 0 root smartctl -a /dev/sda ; smartctl -a /dev/sdb
Dont forget to reload cron!
You will need this package: “smartmontools” for “smartctl”.
The output will be send via mail at root. You can configure that in “/etc/aliases” and then running “newaliases”.
I recently got the error “Argument list too long” when trying to remove many files in a folder:
$ rm *
bash: /bin/rm: Argument list too long
If all the files do not contain whitespace or special chars, this will work:
Otherwise you you need a solution with special handling of filenames.
I use Pidgin for some IRC channels but there is a problem with some channels with special chars in their channel topic text. There seem to be some special chars that are used for colors and similar, but pidgin does not handle them correctly in the xml files like blist.xml (buddylist).
More about that bug in this tickets:
My fix for the bug: Creating a start script for pidgin, that converts the badly escaped stuff in the xml files:
if [ -f ~/.purple/blist.xml ]
mv ~/.purple/blist.xml ~/.purple/blist.xml.tmp
cat ~/.purple/blist.xml.tmp | sed 's/\&#x/\&#x/g' > ~/.purple/blist.xml
If your REALY REALY REALY have to disable SeLinux, check out this file:
If you want to disable your firewall on your centos machine, try this command:
either then manually changing the file
Also check out the other “system-config-*” commands
I recently installed CentOS 6.2 in a VirtualBox and was pretty astonished, because it did not have networking or a eth0 working.
You need to change this file:
More info about that problem:
If more than one person has to manage a server, it might come in handy to have some kind of versioning of the config files.
But also having the possibility to keep track of local configuration changes in your system could be very usefull.
I found 2 tools that can help you:
etckeeper is a collection of tools to let /etc be stored in a git, mercurial, darcs, or bzr repository. It hooks into apt (and other package managers including yum and pacman-g2) to automatically commit changes made to /etc during package upgrades. It tracks file metadata that revison control systems do not normally support, but that is important for /etc, such as the permissions of /etc/shadow. It’s quite modular and configurable, while also being simple to use if you understand the basics of working with revision control.
$ apt-get install etckeeper
Changetrack logs modifications of a set of files, and allows recovery of the tracked files from any stage of development. The changes are presented in a powerful web-based form, a text file, or an email message. A handful of options allow situation-specific configuration, but the code is readily available for more complex modifications.
$ apt-get install changetrack