Server Horror

HowTo: OpenSolaris with Nagios (The contrib pkg repo)

Posted in Uncategorized by serverhorror on 2009-11-26

I wanted to get Nagios running on OpenSolaris, honoring the pkg System I tried:

pkg search nagios

No dice, but I hadn’t added the contrib repo yet and didn’t know about it up until that point :)

Just add

as a pkg source and then install Nagios. Very nice!

HowTo: Create an OpenSolaris Zone

Posted in Uncategorized by serverhorror on 2009-11-24

OK this is my first encounter with OpenSolaris Zones, so take all off the following with a grain of salt.

To make it (hopefully) easy to destroy the zone we create a ZFS Filesystem which will hold all of our zones: (more…)

Tagged with: , , , ,

Easy Installation of VirtualBox on OpenSolaris (and maybe other stuff)

Posted in Uncategorized by serverhorror on 2009-11-24

So I was looking how to install VirtualBox under OpenSolaris. My first thought was a simple

pkg install virtualbox

or similiar would be enough. It isn’t :). (more…)

Tagged with: , , , , ,

OpenSolaris Release Repository

Posted in Uncategorized by serverhorror on 2009-11-23

Should I ever want to switch between release and dev repositories this is just a quick note to the corresponding links:

Tagged with: , , , , ,

Running OpenSolaris now (for better or worse)! – Yay?

Posted in Uncategorized by serverhorror on 2009-11-23

So, I just switched to OpenSolaris on my workstation. In the sense of “Eating one’s own dog food” I may have a better argumentation on wether I should like it or not. So far everything seems to work althoug I don’t quite have a mirrored ZFS root pool in case one disk dies.

I also nearly immediately switched to the “OpenSolaris Development Release Packaging Repository” – actually that’s a lie. I’m switching to it right at this very moment….

$ pfexec pkg set-publisher -O http://pkg.opensolaris.org/dev opensolaris.org
$ pfexec pkg image-update
Creating Plan /

Maybe, just maybe I’ll stick to it….

Tagged with: , , , , , ,

HowTo: Mirror the ZFS rpool

Posted in Uncategorized by serverhorror on 2009-11-15

Let’s use the shiny ZFS to have a mirrored pool which we can use to boot in case one of the hard disks fails. This involves several steps : (more…)

Tagged with: , , , , ,

Where are the BSD Licensed “Enterprise Filesystems”?

Posted in bsd, tools, unix by serverhorror on 2009-10-25

A very simple question.

Where is the “Enterprise Filesystem” licensed in a pure BSD way?

I know ZFS has been implemented by OpenSolaris (well yeah…that’s the origin of ZFS) as well as FreeBSD but the CDDL is not without critique in the BSD community. The OpenBSD people consider the CDDL even more restrivtive than GNU/GPL.

So a question to all the guys and girls who are able to create a filesystem (I can’t my coding skills are nowhere near to be able to create one), where is the BSD licensed filesystem with all the new features that make a hype?

I’m thinking about stuff like this (thou maybe not complete, somebody may have an absolutely usefull new idea that obsoletes all these features

*This may not be a Problem now, but I do get why Sun chose to make ZFS a 128-bit filesystem

RE: Is there a best distro? (LinuxJournal)

Posted in Uncategorized by serverhorror on 2009-10-19

LinuxJournal is asking for the best Linux Distro, first I thought

It’s actually quite easy to answer (especially to someone who will understand the implications):

There is no “best Linux”

Just asko what’s the best commercial Unix out there. One could
say: “They all start at the same point” it’s a kernel and some
toolset you can then use and (re)script to your needs.

The best ist actually the one that meets your requirements best.

“All of the distributions start from the same point”

I think that’s just plain wrong. Yes they do have the same kernel (do they really? – Debian GNU/kFreeBSD). But even switching from Ubuntu to Debian or vice versa has a lot of hurdles to take (“/etc/inittab” on Ubuntu?) - Upstart in Debian/Stable?

I’d rather suggest FreeBSD or OpenBSD (4.6 is just out, I suggest everyone to give it a try) - or is crucial because choosing between the 2 is (IMHO) even wider apart than choosing between RHEL and Debian. Note: I suggest this iff[sic] you want to stay in an enviroment that stays mostly the same thruought all servers and desktops

FreeBSD is FreeBSD is FreeBSD – on every box.

Linux is not “just Linux” from that point there’s just too much diversity among the distros. Even with a single Distro you can have large differences, of course you can take any BSD apart so that it won’t look like the original distribution anymore, but staying within the universe of best practices you won’t have as much diversity between FreeBSD installations as between Linux installations within a single distribution.

Selection of rich applications

The key word here is Selection. Compare RHELs selection to Debians selection. You wouldn’t think that an enterprise grade distro doesn’t even have Nagios in the repositories - aren’t the enterprise people monitoring their servers too?

My choice?

Simple, give me the problem description and I’ll have a look which of the distributions (or even non-Linux – read BSDs, or even non-FOSS) solutions will suit best. It doesn’t help to get RHEL if they don’t support the software you need, but it’s neither helpful to stay with Debian if you need proven Enterprise grade support contrancts with 4 hours of maximum time to react. (Btw: OpenSolaris does have the option to buy support from Sun)

Of course that would be only the start of a discussion with someone who already knows enough about *NIX to discuss on such a Level. As you said dear LinuxJournal:

This was not someone who is unfamiliar with technology, or UNIX for that matter, but someone who is one of us

µ-awk

Posted in tools, unix by serverhorror on 2009-10-05

My personal µ-awk tutorial, just some notes I like not to forget

Print the average of a column:

# grep daemon_with_interesting_column_eight daemon.log |\
  awk '{sum+=$8} END { print "Average = ",sum/NR}'

Print the average of a column fullfilling a constraint:

# awk '/$8>=0.02/  {sum+=$8} END { print "Average = ",sum/NR}'

I need a Wiki (or two, or three) – Quick!

Posted in Uncategorized by serverhorror on 2009-08-17

OK so you need a Wiki in your company? – Not much of a problem just grab a copy of some wiki and there you go. Now that is fine for very small companies. You may need to have a Wiki for each project or departement and have some user rights within those Wikis. What are you gonna do? Create a whole new instance all the way thru everytime? I recommend a Wiki-Farm!

Let’s have some requirements:

  • ACL or any kind of permissions for pages
  • Shared User Base (mentioned below as an exercise to the reader…)
  • Quick Setup of new Wikis (more…)