Setting Up a SPARCstation
The “images” directory on the CD-ROM and the ftp server contain both the floppy files (boot-v0.img and boot-v2.img) and a separate tftpboot.img file. As with the floppy images, the latest tftboot.img file can downloaded from one of the Red Hat mirror sites.
It is this latter file that is required for a network boot. The method used to configure the boot server is almost identical for Linux, SunOS and Solaris systems. We'll use an example configuration, where our boot server is a machine called tyne.gaijin.co.jp and the SPARCstation client is coquet.gaijin.co.jp. The server IP address is 172.17.172.50.
Assign an IP address for your new machine on the same sub-net as the boot server. In our example, the next available IP address for the client “coquet” is 172.17.172.52. It's hardware Ethernet address (MAC) is 8:0:20:3:9:96.
Create a directory for the tftp boot files on the server. SunOS and Solaris machines use /tftpboot by default, and we can use this directory on a Linux system, too.
Copy the tftpboot.img file to the newly created directory.
Create a symbolic link from the tftpboot.img file to a unique file name which the SPARCstation boot PROM requests across the net. The format of the symbolic link is <CLIENT_HEX_IP_ADDRESS>.<ARCHITECTURE>. Take the IP address which you assigned in the first step and convert it, section by section, to hex and then add the architecture of your system. In our example, we need to convert 172.17.172.52 into hex words and, since we aren't exactly sure what architecture our new system is, we'll create links for both 4c and 4m machines.
172 = AC 17 = 11 172 = AC 52 = 34 ln -s ./tftpboot.img AC11AC34.SUN4C ln -s ./tftpboot.img AC11AC34.SUN4M
Note that unlike SunOS and Solaris, the same boot image can be used for both 4c and 4m architectures.
Enable the tftp daemon in /etc/inetd.conf. The syntax for this entry is slightly different between Linux and SunOS/Solaris systems. On the Sun systems there's a -s option which enables the daemon in “secure” mode. The Linux tftp daemon does not use this option and, if it is present in the config file, it treats it as a directory name and all tftpd accesses fail. The other difference between Linux and SunOS/Solaris is that most recent versions of Linux come with the inetd.conf file configured with the tcpd logging daemon configured as default.
tftpd dgram udp wait nobody /usr/sbin/tcpd\ in.tftpd /tftpboot
tftpd dgram udp wait root /usr/etc/in.tftpd\ in.tftpd -s /tftpbootSolaris:
tftpd dgram udp wait root /usr/sbin/in.tftpd\ in.tftpd -s /tftpboot
Reinitialize inetd using a kill -HUP <inetd PID> or by rebooting the server.
Ensure that the the client's Ethernet address is in the arp cache, and the rarp cache for Linux systems, on the server.
arp -s 172.17.172.52 08:00:20:03:09:96 rarp -s 172.17.172.52 08:00:20:03:09:96Note the leading zero padding added to the Ethernet address in both cases.
The rarp command might produce this error:
cat: /proc/net/rarp: No such file or directory
It indicates that rarp isn't compiled into the kernel, and that the rarp module hasn't been loaded. Use insmod to load it and rerun the rarp command.
server# insmod /lib/modules/2.*.*/ipv4/rarp.oIf the module isn't present, you'll have to rebuild the kernel with rarp enabled.
The client system usually takes about three minutes to boot into the installation program. The screen changes from the default black-on-white to white-on-black with a much smaller font once the kernel loads. You'll start to see the normal Linux boot messages as devices are probed and identified. At the end of the boot sequence the system drops straight into the install program, and from that point onwards the prompts are pretty much self explanatory. There are still a few things to watch out for, though.
Red Hat install makes virtual consoles available to the user during the whole of the installation process. <ALT>F1 gets you to the main installation screen, <ALT>F2 is a shell, <ALT>F3 displays informational messages from the installation program, <ALT>F4 displays console messages and <ALT>F5 displays messages from the individual package installation programs as they run.
|Non-Linux FOSS: libnotify, OS X Style||Jun 18, 2013|
|Containers—Not Virtual Machines—Are the Future Cloud||Jun 17, 2013|
|Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer||Jun 12, 2013|
|Weechat, Irssi's Little Brother||Jun 11, 2013|
|One Tail Just Isn't Enough||Jun 07, 2013|
|Introduction to MapReduce with Hadoop on Linux||Jun 05, 2013|
- Containers—Not Virtual Machines—Are the Future Cloud
- Non-Linux FOSS: libnotify, OS X Style
- Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer
- Linux Systems Administrator
- Introduction to MapReduce with Hadoop on Linux
- RSS Feeds
- New Products
- Validate an E-Mail Address with PHP, the Right Way
- Weechat, Irssi's Little Brother
- Tech Tip: Really Simple HTTP Server with Python
- Poul-Henning Kamp: welcome to
6 min 25 sec ago
- This has already been done
7 min 25 sec ago
- Reply to comment | Linux Journal
52 min 39 sec ago
- Welcome to 1998
1 hour 41 min ago
- notifier shortcomings
2 hours 4 min ago
3 hours 41 min ago
- Android User
3 hours 43 min ago
- Reply to comment | Linux Journal
5 hours 36 min ago
8 hours 25 min ago
- This is a good post. This
13 hours 38 min ago
Free Webinar: Hadoop
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Some of key questions to be discussed are:
- What is the “typical” Hadoop cluster and what should be installed on the different machine types?
- Why should you consider the typical workload patterns when making your hardware decisions?
- Are all microservers created equal for Hadoop deployments?
- How do I plan for expansion if I require more compute, memory, storage or networking?