Best of Technical Support

Our experts answer your technical questions.
Choosing the Right Commands

I'm trying to configure my Red Hat 6.0 system to allow clients to access CD-ROM images from my Linux server hard drives. After looking at various file systems such as Samba and NFS and commands such as MAKEDEV, vnconfig, mount, smbmount etc., I'm getting confused as to which combinations of commands to use. —Mark J. Foucht, mark.foucht@eds.com

Red Hat 6.0 uses knfsd, which works somewhat differently compared to the old userland NFS server. One big difference is that you have to export each file system mounted in order for clients to see them (with the old server, you could just export= /, and clients would have a view on all your file systems).

In your case, if your CD-ROM is mounted under /mnt/cdrom, put the following in your /etc/exports file:

/mnt/cdrom (ro)

Then, type the following to migrate the entry to /var/lib/nfs:

/xtab
moremagic:~# exportfs -av
exporting :/mnt/cdrom

To see if it worked, type:

moremagic:~# showmount -e localhost
Export list for localhost:
/mnt/cdrom (everyone)

To mount from another machine, type:

mkdir /mnt/remotecd
mount remotemachinename:/mnt/cdrom /mnt/remotecd

—Marc Merlin, merlin@varesearch.com

If your CD-ROM will be used by Windows computers, you should use Samba. Here is the entry you can add to your /etc/smb.conf file:

[CDROM]
comment = CDROM
path = /mnt/cdrom
read only = yes
guest ok = yes
case sensitive = no
mangle case = yes
preserve case = yes

You should restart Samba after modifying the file. Just type as root:

/etc/rc.d/init.d/smb restart

If you want to make it accessible to NFS users (UNIX computers), you should add the line /mnt/cdrom to your /etc/exports and restart your NFS daemon by using /etc/rc.d/init.d/nfs restart. —Pierre Ficheux, pficheux@com1.fr

Memory Error, Parallel Computing

I am currently running Caldera Openlinux 1.3 on a Compaq Presarion CDS 526 (486 66MHz). Believe it or not, I had no trouble getting it loaded on my machine. I do have a problem with my RAM memory. When I do the free command, it shows I have only 15MB of memory, when I actually have 36MB. Why is this? Is it a problem that has been corrected in a more current kernel, or is it more of a hardware problem?

My next question concerns the world of parallel computing. I have a new computer on order (P3 500MHz), and when I get it, I will be installing Linux on it as well as the one mentioned above. I am interested in hobbying in the world of parallel computing, and I wondered if it would do any good trying to run parallel with a 500MHz machine and a 66MHz machine, or will the whole thing run slower? Thanks for your help. —John, johnwtaylor2@hotmail.com

This 36MB you've mentioned is a rather “non-standard” amount of memory. Please use a dmesg command to see how much memory it finds during boot time. —Mario Bittencourt, mneto@argo.com.br

There is a kernel option for limiting the memory to 16MB; maybe it is activated in your current kernel. You should recompile a new kernel without this option, in “General Setup”: Limit memory to low 16MB (CONFIG_MAX_16M) [N/y/?] N —Pierre Ficheux, pficheux@com1.fr

On your second question, it will depend on how you do it. Think of a job jar, representing a problem decomposed into independent jobs. Each CPU grabs a job out of the jar when it's finished with the previous job. With good choices of job sizes, you win. If the jobs are too small, the extra communication and coordination overhead negates the gains from the slower CPU. If too large, the faster CPU will finish first and have to wait for the slower one to finish—and it may end up waiting longer than if it had done all the work itself. You may have to experiment to find good job sizes, though the obvious computation based on the two systems' relative speeds should get you in the right neighborhood.

I'd recommend you start by looking into PVM, the Parallel Virtual Machine system. Find PVM at www.epm.ornl.gov/pvm/pvm_home.html. —Scott Maxwell, s-max@pacbell.net

Correction

______________________

White Paper
Fabric-Based Computing Enables Optimized Hyperscale Data Centers

Today’s modular x86 servers are compute-centric, designed as a least common denominator to support a wide range of IT workloads. Those generic, virtualized IT workloads have much different resource optimization requirements than hyperscale and cloud applications. They have resulted in a “one size fits all” enterprise IT architecture that is not optimized for a specific set of IT workloads, and especially not emerging hyperscale workloads, such as web applications, big data, and object storage. In this report, you will learn how shifting the focus from traditional compute-centric IT architectures to an innovative disaggregated fabric-based architecture can optimize and scale your data center.

Learn More

Sponsored by AMD

White Paper
Red Hat White Paper: Using an Open Source Framework to Catch the Bad Guy

Built-in forensics, incident response, and security with Red Hat Enterprise Linux 6

Every security policy provides guidance and requirements for ensuring adequate protection of information and data, as well as high-level technical and administrative security requirements for a system in a given environment. Traditionally, providing security for a system focuses on the confidentiality of the information on it. However, protecting the data integrity and system and data availability is just as important. For example, when processing United States intelligence information, there are three attributes that require protection: confidentiality, integrity, and availability.

Learn more about catching the bad guy in this free white paper.

Learn More

Sponsored by DLT Solutions