October 2008 ~ pimp-my-rig reloaded

FAQ: PERL CPAN Error "Undefined subroutine &Compress::Zlib.."

root@host# perl -MCPAN -e shell

cpan shell -- CPAN exploration and modules installation (v1.9205)
ReadLine support enabled

cpan[1]> install Bundle::CPAN
...
...

In order to cut the story short we encountered the error below:

Undefined subroutine &Compress::Zlib::gzopen called at /usr/lib/perl/5.8.8/CPAN.pm line XXXX.

About to give up hope in this automated and supposedly seamless installation, I decided to give it a manual push. Quitting the cpan shell, I changed directory to where the source files of Compress::Zlib were located.

root@host# cd /root/.cpan/build/Compress-Zlib-2.015
root@host# perl Makefile.PL -xs
Writing Makefile for Compress::Zlib
root@host# make test
PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0, 'blib/lib', 'blib/arch')" t/*.t
t/000prereq.....ok
t/01version.....ok
t/03zlib-v1.....ok
t/05examples....ok
t/06gzsetp......ok
t/08encoding....ok
t/14gzopen......ok
t/99pod.........ok
All tests successful.
Files=8, Tests=766, 3 wallclock secs ( 0.02 usr 0.21 sys + 1.76 cusr 0.67 csys = 2.66 CPU)
Result: PASS

root@host#

After seeing the result of "make test" above, I had breathed a sigh of relief -- Result: PASS. Manually kicking the installation of Compress::Zlib did solve the problem. After execution of "make install", we went back to the automated install of the Bundle::CPAN module and everything went well.

I hope the above solution works for you as well.

FAQ: ACPI Multiprocessor HAL Upgrade Script

dillagrfaq, windows

Rather than replying to separate mails with the script attached, I have decided to post the code of the script here instead.

Kindly refer to below for the HAL upgrade script:

@echo off
@title "Upgrading to ACPI Multi-Processor HAL.."
cls

echo ====================================================
echo Upgrading to ACPI Multi-Processor HAL..
echo ====================================================
echo.
echo please wait..

devcon sethwid @ROOT\PCI_HAL\0000 := !E_ISA_UP !ACPIPIC_UP !ACPIAPIC_UP !ACPIAPIC_MP !MPS_UP !MPS_MP !SGI_MPS_MP !SYSPRO_MP !SGI_MPS_MP > nul
devcon sethwid @ROOT\ACPI_HAL\0000 := !E_ISA_UP !ACPIPIC_UP !ACPIAPIC_UP !ACPIAPIC_MP !MPS_UP !MPS_MP !SGI_MPS_MP !SYSPRO_MP !SGI_MPS_MP > nul
devcon sethwid @ROOT\PCI_HAL\0000 := +ACPIAPIC_MP > nul
devcon sethwid @ROOT\ACPI_HAL\0000 := +ACPIAPIC_MP > nul
devcon update %windir%\inf\hal.inf ACPIAPIC_MP > nul

echo.
echo ====================================================
echo Script Completed: press any key to reboot..
echo ====================================================

pause > nul

devcon reboot

Copy and paste the above code into your favorite text editor and save as halupgrade.cmd. Please note that lines starting in "devcon" (the five consecutive lines of code) should end in "nul".

I hope the above helps. Share with me your success story in using the script.

FAQ: Disable Port 135 (Disable DCOM)

dillagrfaq

"There’s No Such Thing As A Silly Question" -- does the cliche sound familiar? In this part of pimp-my-rig reloaded, technical questions are answered. Mail them to me and I will post the answers here. If you have a better answer, by all means share it with us.

FAQ: One of the questions thrown at me was about closing down port 135. This of course entails disabling the corresponding application using the port, which, in this case is DCOM. Thus this procedure is also for disabling DCOM.

NOTE: Remember, before making any changes backup the registry or export the branch for safekeeping. Simply merge the backup in case things don't go as smoothly.

[1] Start by launching the registry editor.

Start » Run » regedit.

[2] Navigate over to key: HKEY_LOCAL_MACHINE \ Software \ Microsoft \ OLE

At the right column, locate the value "EnableDCOM" and modify the value to "N".

[3] Navigate to this registry key: HKEY_LOCAL_MACHINE \ SOFTWARE \ Microsoft \ RPC

Right click on & Modify the value named "DCOM Protocols" Under the key "Value Data", you will see values like below (or something similar). These values keep port 135 open. Highlight everything listed and delete all existing data. Doing so gives "DCOM Protocols" blank data which will in turn close down port 135.

[4] It is now safe to disable the services related or tied to DCOM. Open Control Panel » Administrative Tools » Services. Disable the following services since DCOM has been disabled:

- COM+ Event System
- COM+ System Application
- System Event Notification

[5] Restart the computer after changes have been made. To verify, when your computer has restarted open a command terminal.

Type "netstat -an" and for certain you will no longer see port 135 -- meaning it has been closed. Hope this has helped you in finally closing down port 135, thereby eliminating a possible vulnerability.

In my experience, this has no impact on office applications or internet connectivity. If running other applications, consult with the vendor for requirements relating to port 135 (or DCOM) before closing it down.

Check with us again soon as we will outline how to close down port 137 (netbios-ns), port 138 (netbios-dgm) and port 139 (netbios-ssn).

REVIEW: GELID Wing12 Silent Fan

dillagrreview

As complicated as making a choice of a heatsink, is the choice for the fan to couple that heatsink. Although generally, some heatsinks are already pre-packaged with fans, that doesn't limit the end-user from tweaking with the stock fan or even upgrading to a better one. When deciding, the major categories of prime concern being noise levels and the volume of air the fan can circulate (in cubic feet per minute or CFM).

Coming up with the choice of fans is even harder due to most hardware sites featuring heatsink performance rather than fans. Not as worse as it may seem, but the truth is -- fans are overlooked. Let us change that impression a bit by putting the GELID Wing12 Silent Fan in the limelight.

Fans do contribute to the system as a whole. And probably the most important aspect when making the choice is striking the balance between noise and performance. More often than not, performance has to be acceptable but at the cost of higher noise levels.

But just as heatsinks have evolved into what they are now, fans do too. When stumbling into the term "hydrowave bearing", what comes into mind is Panaflo. Panaflo fans have built a reputation due to that technology embedded in its fans and the proven performance it delivers. GELID attempts to do the same with its innovative nanoflux bearing (or NFB).

According to its product label:

NanoFlux Bearing (NFB) - Patented GELID's Nanoflux bearing incorporates a clever magnetic design, very high precision manufacturing and runs friction-free and operates noiseless. Electromagnetic fields are exploited to keep solid parts from touching. The load is carried by a magnetic field. The nanotechnology composite ensures lowest abrasion and extreme durability (MTTF 100,000 hours).

Compared to sleave and ball bearings the high precision self-lubricated material of GELID's Nanoflux bearing eliminates oil leakage. Additionally the impeller can be detached and this the fan blades can be cleaned easily and is even waterproof.

The fan is conceptualized with absolute silence in mind. Although the featured ratings of 0.22A 1500 RPM already indicate silent operations for a 120mm fan, it is further packaged with a fan speed control and rubber gromet fan mounts to absorb vibration. However minute, this attention to detail is vital.

It is not everyday you get to see a waterproof fan and the Wing12 is one. On the aethetics department, the fan blades are UV reactive. So if you got yourself a windowed case with UV cathode lamps this fan will add more bling bling to the rig. Judging from the look of the fan, the advertised high quality material composition is very evident.

Also, the fan impeller could easily be detached and replaced with an impeller of another color. The detachable impeller makes possible easier cleaning of the fins and the fan cage.

To test its performance, the Wing12 silent fan was mounted on a Thermalright Ultra 120 Extreme (TRUE120) with an Intel Core 2 Duo E6600 overclocked to around 3GHz. With the same set-up, we pitted comparator fans against its performance.

To load the processor cores, the multi-threaded WinRAR benchmark was ran. And with only the heatsink fan spinning, temperature readouts were taken. See below for the results.

The fans performed according to their CFM ratings. With FM121 providing the best output of around 110CFM but at disturbing noise levels while the two silent competitors follow suit. GELID Wing12 proved to be slightly better than the AF12025. Although both fans were silent, when in operation the Wing12 was relatively quiet compared to the AF12025.

The Wing12 strikes the balance between performance and silent operations. In my opinion, the bundled fan speed control mechanism is no longer required as the fan is silent as it is. Without the bundled fan speed control the price could probably be shaved further making the fan more affordable. Perhaps another variant which has a higher speed and CFM rating can make use of that accessory.

It is interesting to see evolution of technology in the area of fans. GELID's Nanoflux bearing is a "newbie" but is up to the task and is certainly a step in the right direction. Factor in that technology together with the build quality, aesthetics, replaceable impeller, the balance between performance and silent operation -- the Wing12 Silent Fan has them. Definitely, the Wing12 silent fan is a powerful addition to a rig aimed at silence. It will, no doubt, pimp your rig!

Thanks to GELID solutions for sending us review samples of the Wing12 and Wing9 low speed silent fans.

REVIEW: WD Caviar 640GB WD6400AAKS SATA II Drive

dillagrreview

In the previous review (Seagate ST3640323AS 640GB), we found out that the cache and platter areal density contributes or influences the overall performance of the drive. And, from the results, it is quite clear that the trend is a direct proportion.

What we will see in this review prompted us to correct that view and state the above statement must be true for two-platter drives. Or this may perhaps be a general rule but with exceptions, of course. What does this mean?

First, please allow me to give a background of what happened. We were using the ST3640323AS 640GB as primary (boot) drive, standalone. After launching several applications, the PC would seem to freeze with the HDD activity LED lit up. Eventually the system will recover after random lengths of "freezing". Other times the drive would "disappear" and go undetected by the BIOS.

After hours of troubleshooting, BIOS flashing, and re-installing Windows, and days of waiting for replacement for the brand new ST3640323AS that we thought to be defective.. we soon discover that there is an incompatibility between the ST3640323AS and the Asus Maximus II Formula. After going through three ST3640323AS replacements, the decision was to replace the drive with a similarly-sized WDC 640GB WD6400AAKS.

(For purposes of traceability, the firmware on the ST3640323AS drives is SD13.)

To gain better confidence with the health of the drive, we ran the same check and found it to be in perfect shape with the obvious conclusion that there must be a compatibility issue with the chipset and the 640GB ST3640323AS.

We ran the benchmark of HD Tune to see if the drive lives up to its standards. And indeed it does!

There are several facts gleaned off this experience and the empirical observations gathered. Not only is the Western Digital Caviar 640GB WD6400AAKS faster than its 1TB next generation counterpart, with higher areal density, it is also incompatibility-free (for the lack of a better term). This seems to be a logical explaination to the previous puzzle on the price premium of WD drives over its counterparts.

Comparing apples to oranges is unfair. Three (3) platter disks are not the same as two (2) platter ones but from the point of view of areal density, bigger does not always translate to faster. It may be too, that the firmware of the 1TB WD1001FALS needs more tweaking to perform better. This line of reasoning seems to be more applicable in this case.

Regardless of the reasons behind, this fact does not escape our notice: the WDC 640GB WD6400AAKS is a fast drive that will pimp your rig!

Acknowledgments, again, to my good friend Xavier Zulueta for the review of the WD6400AAKS.

FAQ: Tame the Time of a Virtual Machine?

dillagrfaq, vmware

"There’s No Such Thing As A Silly Question" -- does the cliche sound familiar? In this part of pimp-my-rig reloaded, technical questions are answered. Mail them to me and I will post the answers here. If you have a better answer, by all means share it with us.

FAQ: The current concept of centralization is being countered by the evolution of virtual machines (VM). Specific functions are decentralized into containers that can be decommissioned just as fast as they were commissioned. However, one can run into problems as the VM seems to have a mind of its own when it comes to time. How then can you tame the VM's concept of time?

Virtual machines are convinient. Given enough resources, VM's for testing purposes can come and go. However, for time sensitive applications the VMs seems to have its own world. And it is important to tame the beast.

For Linux or Solaris x86:

1. Configure ntp client and enable to start automatically on boot.

2. Install vmware-tools on the client and enable to run on boot as well.

3. Still on the client, add clock=pit in the line containing kernel of the grub configuration file (menu.lst). On some distributions, this should be clocksource=pit instead of clock=pit.

4. On the VM server, edit the client vmware config file (SOME_HOST.vmx) on the vmware server. Change "tools.syncTime" value from "FALSE" to "TRUE".

5. Reboot the client after making the above changes.

For Windows:

1. Install vm-tools on the client and enable the service to automatically start. A reboot is required after the install.

2. The vm-tools icon will appear on the system tray of the VM client. Right click on this icon and launch vm-tools.

3. Under the Options tab, tick "Time synchronization between the virtual machine and the host operating system." (see below). Click OK to accept the changes.

The above will keep the VM time in sync with the host OS.

REVIEW: WD Caviar Black 1TB WD1001FALS SATA II Drive

dillagrreview

A few years back, having a hard drive that boasts 250GB of capacity is the bomb. To conform to uber standards computers need to have hundreds of gigabytes of storage. Nowadays the mention of gigabytes is a common thing, nothing out of the ordinary. Gigabyte is out.. Terabyte is in.

What we have for review is the Western Digital Caviar Black 1TB WD1001FALS desktop hard drive. It has a generous 32MB of cache which, as you may notice, is becoming quite common to desktop drives as well. How does this drive perform? Please allow us to share our experience.

In our review of the Seagate ST3640323AS 640GB, we found out that the cache and platter areal density contributes or influences the overall performance of the drive. The WD1001FALS has three (3) 334GB platters that allows the drive to reach the terabyte mark. Up 14GB from the quick Caviar SE16 product line.

The drive also features WD's StableTrac which according to the Western Digital website is:

StableTrac - The motor shaft is secured at both ends to reduce system-induced vibration and stabilize platters for accurate tracking, during read and write operations.

The WD1001FALS also has a dual processor system to maximize performance. See below how it performs under HD Tune.

It is good to see drives hitting average transfer rates above 100MB/s. The WD Caviar Black 1TB WD1001FALS goes over the watermark with ease, averaging close to 110MB/s. Judging from the graph, its performance is almost uniform from the beginning of the disk to the outer end.

The drive is also quite quick, which is apparent from the lower access times in the benchmark.

And one of the many aspects common to all desktop drives now is the lengthy five (5) year warranty. It seems every hard drive manufacturer is ever growing confident with the quality of their hard drives. True enough in the last few years I have experienced DOA drives but I have yet to experienced a drive give out mid-life. DOA means manufacturing or transport induced defects. On the other hand mid-life death is more likely due to wear and is expensive -- both to the end-user and to the manufacturer.

With desktop grade drives offering fast transfer speeds, healthy amounts of cache and about the same five-year warranty the aspect that would separate each from its competitors is its price. But in the Philippines the WD1001FALS sells for about PHP 1000 more than its 1TB competitors.

There must be something in the drive that warrants a price premium but it still stands as a puzzle to us what that is. If you think paying about $20 more for the same size, then go for it.

As said in the beginning of this review, gigabyte is out terabyte is in. The WD Caviar Black 1TB WD1001FALS has just that, and it will definitely pimp your rig!

Acknowledgments to my good friend Xavier Zulueta for the review of the WD1001FALS.

HOW-TO: Disable USB Storage Devices

dillagrhow-to, security, usb

IT infrastructure engineers install firewalls to protect the intra-net from outside. Mail filters on SMTP servers are there to protect users from SPAM. Anti-virus software are deployed to further boost protection on the workstation itself. Regardless of whatever software or hardware is put in place, it serves to beef up security.

One of the many open portals or backdoors that may compromise the infrastructure as a whole is the USB port. Why?.. because USB portable storage devices are probably the next best thing since sliced bread in the storage world -- portable, handly, concealable, you name it. As many as its advantages are its disadvantages. To secure the infrastructure, the access to the USB port needs to be restricted. But how?

A year ago 1GB seemed to be a lot, but 2GB is even frowned upon now. This topic is a bit old.. But still the threat is out there in the open. And the issue is still a hot topic for debate. There is a lengthy instruction on how this can be done but there is no concrete solution that permanently works. Even microsoft has published a lengthy procedure that outlines how this is done.

The procedure outlined in the microsoft KB article, suggests the denial of permissions on the two files responsible for the installation of drivers and services for USB storage functionality.

%SystemRoot%\Inf\Usbstor.pnf
%SystemRoot%\Inf\Usbstor.inf

The deny permission is important especially if the users are non-administrators. The mentioned files could also be renamed, but it needs to be noted so that the procedure can be reversed if need be.

Aside from the above, the registry needs to be checked for this key:

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\USBSTOR

The value for the data key "Start" should be set to 4 (or Disabled) from 3 (or Automatic). The value determines whether the driver is loaded or not when Windows starts. A reboot will be required after the changes.

But for a person who wants a click and go solution, one may want to try another utility to achieve the same purpose. You may download the utility from the author's website.

The click and go solution can also quickly reverse the procedure to enable USB storage devices. The only thing it cannot do is work around the required reboot.

Tighter security always sacrifices functionality. As usual, the goal is to strike a balance between the two. But when it comes to the balance and trade-offs, it is always a grey area.