[[Mathias Chauvin]] - IT stuff

To enable SSH on ALOM, the ALOM variable “if_connection” has to be set to ssh, followed by ALOM reset:

sc> setsc if_connection ssh
sc> resetsc -y

In order to improve the process of deploying RedHat servers in a development lab, a new script has been written.
This script is named:

kick-setup.sh

The aim of this script is to set the kickstart environment for a new RedHat release on a Kickstart server running Solaris 10 or a flavor of Linux OS in a fully automated manner.
Two steps only are now required to prepare the Kickstart environment for a new RedHat release (can simply be a new RedHat update as well):

1. First, connect to the RedHat network https://rhn.redhat.com to download the RedHat DVD iso you need and that is not already setup in the kickstart server.
   Place the downloaded iso file in your iso repository. Note that the path to the isos must look like:

     /path/to/isos/linux/RHEL?/

   where ? is the number of the RedHat release.
2. Connect to your Kickstart server and run the script with the appropriate options.

Here below the help for the script, as well as an example:

root@ksserver:/export/kickstart# /opt/mc-thias/bin/kick-setup.sh -h  

This script intends to setup the whole Kickstart server 
for a specific release and update of RedHat Enterprise Linux.
This setup will be done according to a few required parameters, 
and thanks to the RHEL DVD iso.

kick-setup.sh is to be used with four mandatory parameters:
  - The release number of the RedHat Enterprise Linux to be deployed. 
    This number is an integer, for instance 5 for RedHat 5.
    This option is:
      -r num
  - The update number of the RedHat Enterprise Linux release to be deployed. 
    This number is an integer, as 4 for RedHat 5 Update 4.
    This option is:
      -u num
  - The arch of the RedHat Enterprise Linux to be deployed. 
    This is either 32b or 64b for 32 or 64 bits x86. 
    This option is:
      -m 32b|64b
  - The iso file of the RedHat EL that is to be setup on the Kickstart server.
    This iso should have first been downloaded from rhn.redhat.com
    This option is:
      -i iso_name

Examples:
To deploy RHEL6 U2 for x86 32bits, you would run the following:
  ./kick-setup.sh -r 6 -u 2 -m 32b -i rhel-server-6.2-i386-dvd.iso

Hit the Read More link to get the script and learn more on how to use it

Read more »

1/ Presentation
For some reasons, mainly reporting/quality ones , one might have to export a list of the clients currently backed up by a Networker server.

When it comes to access the Networker configuration from the command lines a few tools appear to be very useful:

• mminfo
• nsradmin

among others. Therefore you are invited to have a look at the man pages of these tools.

2/ The request one can face
In order to comply with some reporting/quality policies, one might want to get the list of Networker clients currently in use, with their Description, Groups, Save set, Browse Policy and Retention Policy.
The nsradmin tool will then be used.
Let’s see how the nsradmin script looks like:

[root@nwserver ~]# cat /tmp/clientlist.cmd
. type: nsr client;scheduled backup: Enabled
show name;group;comment;"save set";"browse policy";"retention policy"
print
root@nwserver:~#

Running this script leads to the result one might expect, displaying every enabled clients, with their name, group, comment/description, save set and browse and retention policies. But the format of the output is not quite easy to work with when building reports (integration in an Excel spreadsheet for instance).
Let’s see an example:

[root@nwserver ~]# nsradmin -i /tmp/clientlist.cmd
[...]
                       name: gw.domain.com;
                     comment: Gateway main;
               browse policy: Year;
            retention policy: 10 ans;
                       group: FULL 1st quarter;
                    save set: All;

                        name: logs.domain.com;
                     comment: Logs server;
               browse policy: Year;
            retention policy: 3 ans;
                       group: Full Q2 Q3 Q4;
                    save set: All;
[...]

3/ Building a CSV format report from these data
In order to make these data easily workable, a Perl script is being used, translating this raw output from nsradmin to a CSV format.

You can fin the script here

Have a look at the content of this script as you might want to modify some parameters:

$nsradmin_cmd = "/nsr/sbin/nsradmin";
$tmp_cmd_file = "/tmp/clientlist.cmd";
$tmp_csv_file = "/tmp/clientlist.csv";

where tmp_csv_file is the CSV output file.

The script is quite simple. It is simply using a hash of hashes to parse the output from the nsradmin script:

%result = ( "_name" => {
                comment => "Description",
                groups => {
                        group   => "Backup Group",
                        saveset => "Save set",
                        brpol   => "Browse Policy",
                        retpol  => "Retention Policy"
                        },
                },
);

This part of the script initializes the hash of hashes with entries that stand for the header of the resulting table.
Then the hash of hashes is built using a for loop:

$svrname="";
$savegrp="";
$grpidx=0;
for $rawlist_el(@rawlist) {
        $field = substr($rawlist_el,0,28);
        $field =~ s/^\s+|\s+$//g;
        chomp $field;
        $data = substr($rawlist_el,30);
        chomp $data;
        chop $data;
        switch ($field) {
                case "name"     { $svrname = $data; $grpidx++; }
                case "comment"  { $result{"$svrname"}{"comment"} = "$data"; }
                case "group"    { 
                       $result{"$svrname"}{"groups$grpidx"}{"group"} = "$data "; }
                case "save set" {
                       $result{"$svrname"}{"groups$grpidx"}{"saveset"} .= "$data "; }
                case "browse policy"    { 
                       $result{"$svrname"}{"groups$grpidx"}{"brpol"} .= "$data "; }
                case "retention policy" { 
                       $result{"$svrname"}{"groups$grpidx"}{"retpol"} .= "$data "; }
                case "" { 
                       $result{"$svrname"}{"groups$grpidx"}{"saveset"} .= "$data "; }
        }
}

An index (grpidx) is used as a client can belong to multiple groups.
The values 28 and 30 might be adjusted but are a match on the server that I’m working with.
They are corresponding to the number of char used for the first field in the nsradmin output.

4/ Use of the output
Once the CSV file has been generated, it can be imported in Excel or in any Spreadsheet tool, and can be easily manipulated to generate reports.

Enjoy…

I’ve been using an USB disk attached to an Airport Extreme lately, and for some any reasons I go, after being able to use it as a Time Machine backup disk, I received the message

Time Machine could not complete the backup. 
The backup disk image"/Volume/.....sparsebundle " is already in use.

Searching for a solution on the web, most of the time this issue is said to be fixed by deleting the sparsebundle file and reinitiating the backup. You’re probably thinking that this is not exactly what can be “acceptable“.

Hopefully, and thanks to a cnet review, I figure out that the following steps can fix the issue, and keep your previous backups safe:

• Access the Airport Extreme using the AirPort Utility
• Go to Manual Setup

• Go to Disks

  
• Click on Disconnect All Users…
• At this stage, unplug or turn the external HD off, the plug it back or turn it on

Everything should be back to normal

The following is a documentation that can be usefull for those who want to access their Unix data from a Windows workstation (Why don’t you use Linux/Mac OS workstations?)

Let start with this:

• The ADS domain name is the same as the DNS name: nulldom.com
• The ADS Kerberos realm is the same as the DNS domain name (written in capitals): NULLDOM.COM

In the following step-by-step the server that will join the domain is named sol10svr.nulldom.com, and is installed with Solaris 10u8 (running on x86).

Don’t loose a sec, click here to

Read more »