Home Home > Systems-management
Sign up | Login

Deprecation notice: openSUSE Lizards user blog platform is deprecated, and will remain read only for the time being. Learn more...

Archive for the ‘Systems Management’ Category

Problems installing software in openSUSE?, Simple solution!

August 11th, 2010 by

Sometime ago I had a little problem with Zypper at the time of adding some packages. The same problem extended to YaST and well, the end of the world.

Possibly there was a integrity problem between the updater and the database RPM, or whatever, the true thing is that the error goes like:

Tamaño total a descargar: 174,1 MiB. Después de la operación se utilizarán 565,0 KiB adicionales.
¿Desea continuar? [s/n/?] (s): s
Descargando paquete graphviz-2.20.2-45.4.1.i586 (1/153), 868,0 KiB (2,2 MiB desempaquetado)
Descargando delta: ./rpm/i586/graphviz-2.20.2-45.3_45.4.1.i586.delta.rpm, 30,0 KiB
Obteniendo: graphviz-2.20.2-45.3_45.4.1.i586.delta.rpm [hecho (19,6 KiB/s)]
Aplicando delta: ./graphviz-2.20.2-45.3_45.4.1.i586.delta.rpm [hecho]
Instalando graphviz-2.20.2-45.4.1 [error]
La instalación de graphviz-2.20.2-45.4.1 ha fallado:
(con –nodeps –force) Error: Subprocess failed. Error: RPM fallido: error: db4 error(-30987) from dbcursor->c_get: DB_PAGE_NOTFOUND: Requested page not found
error: error(-30987) getting “” records from Requireversion index
error: db4 error(-30987) from dbcursor->c_get: DB_PAGE_NOTFOUND: Requested page not found
error: error(-30987) getting “” records from Requireversion index
error: db4 error(-30987) from dbcursor->c_get: DB_PAGE_NOTFOUND: Requested page not found

If you have some error like this one, the solution its extremely simple, just run in the command line interface:

sudo rpm --rebuilddb && sudo zypper clean -a && sudo zypper ref

This will rebuild the RPM database and then refresh the repos. 😉

Software search trick

August 4th, 2010 by

Do you use software.opensuse.org and get an error message “search limit reached” when searching for a generic term like “perl” or “kde”? Here is the solution:
The software search now also supports matching for exact package names, just put your search string into double quotes! See for example perl or kde4

Road to 11.3 : when pattern are not your friend, pre selection can be a trap

June 10th, 2010 by

So it’s time to take some hours to test our future version.

Today I start a fresh M7/Factory install : booting from pxe. The test case is build quickly a minimal server text mode.

Just uncheck the auto configuration, we are after all linux admin. Choose your partition keyboard, language (en recommanded for server) etc … normal.

Just before starting the install check software :  click on installation resume . You will discover that base-system-pattern would like to install a kernel-desktop, wtf why we want a server !

So there’s a new ticket about that : https://bugzilla.novell.com/show_bug.cgi?id=613216

I express the fact that it would be nice to have a new pattern selected when we choose minimal install server text mode.

And you what about your opinion about pre-selection or having a base-system-server pattern … Please comment, & vote on bugzilla

A pattern guru wanted to build a patch for that.

Your own OEM configuration: YaST Firstboot

May 13th, 2010 by

Have you ever thought how users should configure their systems, deployed by AutoYaST or kiwi? One of possible answers is called YaST Firstboot.

The YaST firstboot utility is a special kind of configuration workflow that can be run after the basic system is installed. It is started on the first boot of the system and guides a user through a series of steps that allow for easier configuration of their desktops. YaST firstboot does not run by default and has to be configured to run by the user or the system administrator. It is useful for image deployments where the system in the image is already configured (read: AutoYaST, SUSE_Studio, KIWI), and end-user should do only the last few steps, like setting the root password.

Enabling Firstboot

To enable running firstboot configuration sequence on the machine, it is necessary to:

  • Have yast2-firstboot package installed. Depending on your installation method, this means adding the package to the list of packages to be installed (e.g. for AutoYaST) or that are part of installation image.
  • Create the empty file /var/lib/YaST2/reconfig_system. If this file is present on system boot, firstboot configuration sequence is started. YaST Firstboot removes the file when the configuration is done.

Customizing YaST Firstboot

There are two files that control the behavior of Firstboot: the firstboot control file (firstboot.xml) and the sysconfig file /etc/sysconfig/firstboot. The control file defines the steps that should be part of your configuration sequence, in the sysconfig file it is possible to define custom messages and paths to various files.

Customized Messages

Most important texts configurable in sysconfig file is the text shown in the License Agreement screens. For license texts, there are sysconfig variables FIRSTBOOT_LICENSE_DIR and FIRSTBOOT_NOVELL_LICENSE_DIR.

The license text is read from the file ‘license.txt’ or from ‘license_<locale>.txt’. The license texts of the Novell base product are by default installed to the directory /etc/YaST2/licenses/base/ — you can set different value to FIRSTBOOT_NOVELL_LICENSE_DIR if they are elsewhere. Use FIRSTBOOT_LICENSE_DIR to indicate a path to directory containing vendor licence texts; it is preferred to put these license texts into another subdirectory of /etc/YaST2/licenses/.

To show two license texts in one dialog (typically the one from vendor and one from Novell), use ‘firstboot_license_novell’ step in your firstboot.xml file. This client will use the license texts specified by both FIRSTBOOT_LICENSE_DIR and FIRSTBOOT_NOVELL_LICENSE_DIR.

Customized Workflow

The default firstboot workflow can be controled using one single file which uses the same syntax as the control.xml file used to control the complete installation. The firstboot control file consists of workflow and proposal configurations and can be used to add or remove configuration screens depending on the end configuration of the system. The file firstboot.xml is installed with the yast2-firstboot package and can be found at the following location: /etc/YaST2/firstboot.xml.

This file can be modified to match the post installation requirements of the product in question. In addition to the default and pre-installed components, custom screens can be added to enable maximal flexiblity during post installation. Look into the example firstboot.xml file coming with your yast2-firstboot package for more available steps.

Custom Scripts

Not everything can be achieved with already prepared steps, and it is usually not necessary to write your own ycp dialogs.

You can add schell scripts to be executed at the end of the firstboot configuration. Scripts should be placed in /usr/share/firstboot/scripts or in a custom location that can be set using the variable SCRIPT_DIR of /etc/sysconfig/firstboot configuration file. The scripts are executed in alphabetical order of their names.

AutoYaST Support

It is possible to configure the firstboot process as a part of autoinstallation, so the system can be installed with most of the default values set via AutoYaST profile, leaving the rest to the end user during the firstboot sequence.

As a part of autoinstallation configuration, you need to provide all the changes mentioned above:

  • Customize /etc/sysconfig/firstboot: it can be done e.g. via Sysconfig Editor in System section of AutoYaST configuration module.
  • Provide customized firstboot.xml file and point to its location in FIRSTBOOT_CONTROL_FILE value of /etc/sysconfig/firstboot.
  • Enable Firstboot: do it via GUI in Misc/Firstboot section of AutoYaST configuration module or manually by adding the section
<firstboot>
   <firstboot_enabled config:type="boolean">true</firstboot_enabled>
</firstboot>

to your AutoYaST profile.

This is the shortened version of the article published at http://en.opensuse.org/YaST_Firstboot

Small script to monitor service/process

April 5th, 2010 by

Small script to monitor a process and restart it if found dead, I am sure there are many more ways to do this, here is one: servicemon

News from the Zypper Revolution

April 5th, 2010 by

Maybe revolution is a bit strong, but at this hour in the night I can probably be excused for using a bit of hyperbole – besides, nothing great in the world has ever been accomplished without passion as Hegel would have it, so I am on solid ground here.

I have been going over customer feedback from Novell’s Brainshare conference for my internal “Systems Management Zeitgeist” report, and there are a couple of points I just had to share with you all as they are plain simply inspirational.

Our update stack is, well, zippy.  Like greased lightning, according to this happy SUSE Linux Enterprise customer:

Zypper updates a Linux system across major versions in 5 minutes, full Oracle server update done in 15 minutes

We of course appreciate speed in of itself, as a technical achievement powered by enhancements like libsat and DeltaRPM, and Community users share this point of view with us.  But Enterprise users have a different and equally valid point of view: administrator time is costly, and while many management consoles exist, industry data shows that tools do help, but not nearly enough: administrators are still involved, personally, in most Systems Management tasks —  I could quote analyst data, but not at this wee hour, so just trust me on this point.

This one medium-sized customer actually took the time to calculate what the time savings meant to his business:

The faster update stack is resulting in 56,000 dollars in [operational budget] savings

It is not everyday a customer gives you a precise dollar number in describing what a technology’s impact is on his expenses — So I just had to share it with you all, it is such a nice commentary on our effort’s tangible impact.

I can hear some of you wonder why I blogged this on my Lizard’s Community account, rather than on Novell’s Corporate site, since I am talking up Enterprise distro data and as the Systems Management guy I really have either option.  Good question!  I could say it is because I was not in the mood to dig up analyst quotes, and this setting allows me to be more cavalier and just waltz over those references, but there is a more important reason, read on.

We in the Systems Management team happen to think sleep is for the weak, and have been cooking up our next scheme for improvement — but we need your help to get there.

As the keenest observers among you have long ago noticed, with the 11.2 release we declared “zypper dup” a supported migration path, and received some accolades for it already.  But we all know that live distro upgrade migration across major version changes is a big endeavor, and we would like to solicit your help in improving it: if you have the time and inclination to test zypper dup and provide a properly filed bug report of any kinks you might discover, we would be delighted to use your feedback to improve the 11.3 implementation of this process.

Just a word of caution: comments to this entry, or bugs filed without sufficient data to be analyzed, are not going to further the result we all seek.  If you report something, make sure enough data to reproduce the issue is included, and that you are able to provide additional data upon request of the developer handling your report: if we cannot reproduce a problem, we cannot fix it.

Thanks in advance to those among you joining us in this effort!

Into the Cloud

March 29th, 2010 by

Setting up your own Cloud infrastructure on SUSE has just become a lot easier. You can now use Kiwi and a mostly pre-configured set-up to build your own Cloud node images. Once these images are built setting up your Cloud can be accomplished in a few minutes.

Checkout the Cloud Recipe in the Kiwi Cookbook.

Happy Hacking

Li-f-e updated

March 24th, 2010 by

openSUSE Education team is happy to announce the availability of the updated openSUSE Education Li-f-e DVD iso. The Linux for Education (Li-f-e) contains the wide selection of education, development, office, as well as multimedia packs to meet all possible computing needs of students, teachers and parents.

Some of the highlights of this update:

Desktop Environments:

Additions:

Updates:

  • All official updates to openSUSE 11.2 since its release
  • LTSP 0.5.1.99, includes fat-client support
  • Banshee 1.6 RC1
  • Code::Blocks SVN 6182
  • and of course most of the education packages like gcompris and tux4kids suite got updated.

Download:

Direct Download | metalink | torrent | md5sum

More mirrors on sourceforge

More information here: http://en.opensuse.org/Education/Live

Have a lot of fun

Your openSUSE Education team

Cheat ocfs2-tools

February 9th, 2010 by

When running ocfs2 on cLVM, openais+pacemaker setup, you may run into something like described below.

The symptoms:

umount /data1
mspx1d0:~ # tunefs.ocfs2 –fs-features=sparse,unwritten /dev/system/data1
tunefs.ocfs2: Unable to access cluster service while opening device “/dev/system/data1”
mspx1d0:~ # mount /dev/system/data1 /data1
mspx1d0:~ # tunefs.ocfs2 –fs-features=sparse,unwritten /dev/system/data1
tunefs.ocfs2: Configuration error discovered while opening device “/dev/system/data1”

Here is how to get past cluster check when running ocfs2-tools.

From hb_gui or crm stop mounting of the file system and ocf::ocfs2:o2cb resources.

Install ocfs2-tools-o2cb package

Do normal stack configuration using /etc/init.d/o2cb configure, but do not auto start on boot. Read the fine ocfs2 manual.

Bring ocfs2 online: /etc/init.d/o2cb online ocfs2

Start ocfs2 service: rcocfs2 start

Update cluster stack on your partition with ocfs2 filesystem: tunefs.ocfs2 –update-cluster-stack /dev/system/data1

Do your thing with ocfs2-tools, such as: tunefs.ocfs2 -v –fs-features=sparse,unwritten /dev/system/data1

Stop ocfs2 service and unload o2cb: rcocfs2 stop && /etc/init.d/o2cb offline ocfs2 && /etc/init.d/o2cb unload

Re-enable ocf::ocfs2:o2cb resources and run tunefs.ocfs2 –update-cluster-stack /dev/system/data1 again. Re-enable mounting from hb_gui.  Hope there is a simpler way somewhere that I am not aware of.

WebYaST Status Module

December 15th, 2009 by

After releasing WebYaST 1.0 last week we have had a look to the status module again and have come to the conclusion that we can do it better. 🙂

The current version looks more like a medical monitoring system that no one understand:

current status page

Technical background

The system data will be collected by a service called collectd (www.collectd.org). This data will be stored in a rrd-database which is located in

/var/lib/collectd/<system-name>/[cpu-0|cpu-1|df|interface|load|memory|….]

An own file/database will for each value will be generated.

The upper screen shot shows that an own graphic will be shown for each value. This is not really user friendly. We would like to show only values which provides “really” interesting information for the sysadmin or user. These values should also be grouped in order to reduce the amount of graphs.

Another lack of information is the labeling of the graphs. There is no description what the values mean and in which period they have been evaluated.

The aim

would be to give the user a nice overview about the system status. Additional, there should be an alarm system available if system values are exceeded. E.g. there should be sent an email to the system administrator if no disk space is available anymore.
The Administrator should also be able to configure “his” needed system values in a nice way.

Suggestion for configuration

configure

There are three parts:

  • General
    General setting like period time of shown data, email address in the case of occurred errors and the color of displayed limits.
  • Graphics
    Defining graphs with

    • labels
    • scales
    • flag for cumulate values
      This means that the values in the graph are added. E.g. the component df of collectd returns two values of a partition (free, used). These values will be cumulated in the graph like this:
      collecdtd disk status
  • Enable Values/Limits
    • Defining which data of collectd will be shown in the graphics described above. This enable us to group data in one graph, defining color and label of each collectd value:
      collecdtd cpu2
    • Defining limits and send an email if the value is overstepped longer than the defined period (Alarm after (min.))

The Result

should be an amount of view graphs which include interesting data only. These values are checked periodically and an email will be sent if they exceed a defined limit.

I am asking for

your opinion about:

  • Do you know a much more fancy way for displaying system status data ?
  • Which kind of Data would be interesting for you (CPU, memory, disk-space,….)?
  • When and for which kind of data would you like to get an email if limits have been exceeded?

So, before changing this module we would like to ask you what you would like to see. 🙂

Thanks in advance.