Recreation of SSL certificates script for OES1 and OES2

Mike — Fri, 13 Aug 2010 08:16:27 +0000

I’ve posted a couple of time but on Coolsolution from Novell there is a post about a script who can recreate all the SSL certificates on the right place for your OES1 and OES2 server.

As of Augustus 12 there is a new version out by jmeldrum.

Fixes and Enhancements:

The script will now check if your are root
OES2 x86_64 is now supported
A relative path to the .pfx file can now be used.

This script will now do pre and post checks to see if the certificates are good or bad
Color was also added for easier reading

No SSH access for vSphere 4.1

Mike — Mon, 26 Jul 2010 13:44:53 +0000

After a upgrade to VMware vSphere 4.1 I couldn’t SSH to the Service Console. First I thought the password wasn’t correct, but after checking to login on the console directly I noticed that the password is correct. When I enabled rootlogin in the /etc/ssh/sshd_config -> permitrootlogin yes, I could login as root through SSH, so SSHD was working. So I must be a new security feature in vSphere 4.1. Then I changed the vSphere permission in the vSphere client to for my user to Administrator. After this I was able to SSH to the server. I’m haven’t had the time to check witch feature is responsible for this but I’m going to.

PFN_LIST_CORRUPT with VMware Workstation 7

Mike — Thu, 22 Jul 2010 07:59:09 +0000

During a lab of the N309o course (Deploying Novell Open Enterprise Server 2 for Linux) a student experienced a blue screen on this VMware Host workstation. This workstation was running Windows XP 64 bit with VMware Workstation build 203739.

The student was trying to install a package on a SLED 11 virtual machine through Yast. For this package Yast needed the SLED 11 DVD. The virtual CD-ROM player was configured as Auto, witch means that the physical CD-ROM player of the host is connected to the virtual CD-ROM player of the virtual machines.

If you Google on the error PFN_LIST_CORRUPT you will find that this error is mostly RAM corruption. In this case VMware Workstation passes wrong information to the memory therefore the blue screen.

Resolution: Configure the virtual CD-ROM player to use a ISO file.

I didn’t test this with VMware Workstation version 7.1. But this is on my list.

NSS on LVM

Mike — Tue, 20 Jul 2010 13:02:29 +0000

By default is you create a NSS pool or volume, these are management by EVMS. EVMS is a disk manager original developed by IBM.
This is a quote about EVMS from the sourceforge website

The Enterprise Volume Management System (EVMS) Project has the goal of providing unparalleled flexibility and extensibility in managing storage. It represents a new approach to logical volume management for Linux. The architecture introduces a plug-in model that allows for easy expansion and customization of various levels of volume management.

EVMS provides a single, unified system for handling all of your storage management tasks. EVMS recognizes all of the disks on your system and allows for a variety of partitioning schemes. Software-RAID and logical volume groups can be managed in EVMS. Filesystems can be created and checked, and are automatically updated when changes are made to the underlying volumes. With EVMS, there is no longer a need for several individual utilities for performing each of these tasks.

In 2006 IBM discontinued development of the project. There have been no further releases until know. Therefor Novell announced in 2008 that they will be moving from EVMS to LVM for NSS.

From OES2 it’s possible to run NSS on LVM with some with some disadvantages:

no iManager support
no NSS tools who uses EVMS (like software RAID and the creation, modifiying and deletion of patitions)
no NSSMU support (the feature F4 update NDS works!)
no Clustering

Advantages of NSS on LVM are:

Better support of LVM in Linux
Creation of a aligned partition (VMware)

So how can we create a NSS Pool with LVM?

Create a empty (aligned) partition with Yast
Create the NSS Pool
mkfs -t nsspool -n nsspoolname devname
Create a directory for mouting the NSS Pool
mkdir /mnt/pooldir
Mount the pool
mount -t nsspool devname nsspoolmountpoint -o name=nsspoolname
Add the NSS pool to eDirectory
NSSMU | Pools | F4 (NDS Update)
Add the NSS Pool mountpoint to the fstab file
devname nsspoolmountpoint nsspool auto,rw,name=nsspoolname

And know the NSS Volume

Create the NSS volume
mkfs -t nssvol -n nssvolname nsspoolname
Create the mountpoint for the NSS volume
mdkir /media/nss/nssvolname
Mount the NSS volume on the mountpoint
mount -t nssvol nssvolname volmountpoint -o name=nssvolname
Update eDirectory
NSSMU | Volume | F4 (NDS Update)
Add the NSS volume mountpoint to fstab file
nssvolname nssvolumemountpoint nssvol auto,rw,name=nssvolname

Now you have a NSS Pool and Volume with LVM

openSUSE 11.3 is out!

Mike — Mon, 19 Jul 2010 11:20:13 +0000

Just a quick copy/paste action from the opensuse.org site:

openSUSE 11.3 is finally out! The openSUSE Project is pleased to announce the release of openSUSE 11.3. openSUSE 11.3 includes new versions of GNOME, KDE, OpenOffice.org, Firefox, the Linux kernel, and many, many more updates and improvements. In 11.3 you’ll find more than 1,000 open source desktop applications. openSUSE also includes a full suite of server software and a rich selection of open source development tools. Get it today!

Novellcongres.nl presentation: OES2 on vSphere best practice

Mike — Thu, 08 Jul 2010 12:10:43 +0000

In June I gave a presentation OES2 on vSphere best practice. After this session I got a lot of possible response from
the people in the studio.

Although you can download the presentation a long time from the novellcongres.nl site, I thought to post it out here also. One disadvantage, because the congres is called novellconges.nl the presentation is in Dutch. But although Dutch isn’t your native language, I guess you can make some sense reading the presentation and looking at the commands. Especially the commands about aligning NSS on a VMware virtual disk is nice!

I only have 1 condition if you download the presentation. You must give me feedback

Change duplex setting on vSphere Service Console

Mike — Tue, 29 Jun 2010 17:20:06 +0000

During a training a student accidentally changed the duplex setting of the vmnic for the service console to fixed 1000Mbit full duplex. This isn’t so bad, if the vmnic wasn’t on a 100Mbit switch!

So I had to go to the service console to change the duplex setting back to auto.
I expected to use the esxcfg-vswif command. But with this command you cannot change the duplex setting. After some investigation I discovered to use normal Linux command like ethtool.

To see the current settings:

ethtool vmnic1

To change the settings to 1000Mbit full duplex:

ethtool -s vmnic1 autoneg off
ethtool -s vmnic 1 speed 1000 duplex full

To change it backup again to auto:

ethtool -s vmnic1 autoneg on

Twitter

Mike — Fri, 25 Jun 2010 13:45:40 +0000

It took a while but inspirited by a girlfriend Maggie van Kempen I registered for a Twitter account. I plan to use my Twitter not to let you all know if I go to the toilet or something like that, but just to announce new interesting blog articles written by me. So if you like to get notified through Twitter, please click the follow me link!

Configuring a SLES DHCP failover

Mike — Wed, 23 Jun 2010 13:40:27 +0000

In most networks DHCP is one of the single-point-of-failures. If DHCP failes, every devices relying on DHCP cannot function.

In order to provide a high availability to your DHCP server, set up a backup DHCP server running the same Linux distribution and version, as follows:

Stop DHCP on both DHCP servers (rcdhcpd stop).

On your primary DHCP server, add lines to the /etc/dhcpd.conf file, similar to following example:

failover peer "dhcp" {
   primary;
   address 192.168.1.10;          # local host IP address
   port 647;                      # make sure this port is not used by other
                                    programs
   peer address 192.168.1.11;     # backup host IP address
   peer port 647;
   max-response-delay 60;
   max-unacked-updates 10;
   mclt 600;
   split 128;
   load balance max seconds 3;
}
include "/etc/dhcpd.master";

On your backup DHCP server, add lines to the /etc/dhcpd.conf file, similar to following example:

failover peer "dhcp" {
   secondary;
   address 192.168.1.11;         # local host IP address
   port 647;                     # make sure this port is not used by other
                                   programs
   peer address 192.168.1.10;    # primary host IP address
   peer port 647;
   max-response-delay 60;
   max-unacked-updates 10;
}
include "/etc/dhcpd.master";

Add the following lines to the end of /etc/dhcpd.master on both DHCP servers. On SUSE Linux Enterprise Server, copy this file to /var/lib/dhcp/etc.

subnet 192.168.1.0 netmask 255.255.255.0 {
  pool {
    failover peer "dhcp";
    range 192.168.1.100 192.168.1.200;
    deny dynamic bootp clients;
  }
  option routers 192.168.1.1;
}

Start DHCP on both DHCP servers (rcdhcpd start.

After these configurations are complete, the primary DHCP server, which is usually a management server or install server, responds to DHCP client requests. If the primary DHCP server goes offline, the backup DHCP server automatically takes its place and provides DHCP service.

VMware and Novell Expand Strategic Partnership to Deliver and Support SUSE Linux Enterprise Server for VMware vSphere Environments

Mike — Thu, 10 Jun 2010 06:36:51 +0000

Just a quick copy-past from a press release

VMware and Novell today announced an expansion to their strategic partnership with an original equipment manufacturer (OEM) agreement through which VMware will distribute and support the SUSE Linux Enterprise Server operating system. Under the agreement, VMware also intends to standardize its virtual appliance-based product offerings on SUSE Linux Enterprise Server.
Read whole article

Of course I think this is great! For the last 5 years now, my main target is SLES/OES on VMware. So that VMware will be using SLES for the Virtual Appliances.. Great!