WD Sentinel DX4000 — unacceptable power design

Western Digital has released a new SoHo NAS product, the WD Sentinel DX4000, which is driven by an Intel Atom mainboard and CPU, supports dual gigE (for failover, not trunking), and up to 4 disks in RAID-5. It’s 8x6x8 inches in size for the 4-disk version. The cost? US$1500 retail.

Remember: the product is intended for businesses or high-class end-users.

Now take a look at its AC power connectors: two circular female connectors with a male centre pin. You know, the common “wall wart” and “in-line brick” crap you’d see on consumer-grade products that results in cable clutter (and absolutely impossible to properly deal with when placed in a rack)? No, surely that couldn’t be, especially for a SoHo device of those physical dimensions…

unssh: what purpose does it serve?

Today I was reviewing CVS commits to ports, and I saw this flow across my screen:

 Edit ports/security/unssh/Makefile
  Add delta 1.4 2011.10.18.22.35.18 pav
 Edit ports/security/unssh/distinfo
  Add delta 1.3 2011.10.18.22.35.18 pav
 Delete ports/security/unssh/files/extra-patch-unssh.sh.in

I immediately wondered what unssh was. Apparently it’s a shell script that modifies (removes lines from) your ~/.ssh/known_hosts file when run with the same arguments as ssh.

This immediately made me think “why is this even necessary when OpenSSH has framework to already accomplish this task”? Just edit your ~/.ssh/config and enter the following:

Host gw
StrictHostKeyChecking no

The only difference here is that this will never remove the outdated (“offending”) entry in ~/.ssh/known_hosts and you will continue to see the nasty man-in-the-middle warning every time you SSH to the host “gw”. However, you’ll be able to connect regardless of the warning.

There’s really no harm in ignoring the nastygram when connecting to a machine you know is going to change its SSH identity keys all the time. Really — the person SSH’ing there will already be aware of that (the fact they would go and find/use unssh is proof of that).

Sadly there is no -q equivalent in ssh_config(5), otherwise StrictHostKeyChecking no and Quiet yes would be a great combination for this exact situation.

Dennis Ritchie has passed away

The statement comes from Rob Pike, so I’m inclined to believe it.

• http://boingboing.net/2011/10/12/dennis-ritchie-1941-2011-computer-scientist-unix-co-creator-c-co-inventor.html
• https://plus.google.com/u/0/101960720994009339267/posts/ENuEDDYfvKP
• https://twitter.com/#!/packetslave/status/124270413927809024
• http://www.biobiochile.cl/2011/10/12/muere-dennis-ritchie-padre-del-lenguaje-de-programacion-c-y-el-sistema-operativo-unix.shtml
• http://nodo9.com/2011/10/12/fallecio-dennis-ritchie-cofundador-de-unix-creador-de-el-lenguaje-los-70-aos-de-edad/
• http://en.wikipedia.org/wiki/Dennis_Ritchie

If you’re a programmer and don’t know who Dennis Ritchie is, then you should have your programming license revoked permanently.

RIP, dmr. You will never be forgotten.

Systems monitoring nightmares on FreeBSD

For quite some time (years) I’ve been dealing with monitoring of FreeBSD systems. “Monitoring” in this case does not mean service availability, it means data/statistics acquisition of key parts of the system — things like memory and VM usage, CPU load, pf (firewall) statistics, NIC statistics, disk usage, disk I/O, and so on.

You still have to store the acquired data somewhere/somehow. And that’s where RRDTool, unfortunately and unjustifiably, comes into play. If you’re an administrator (or developer) that has to deal with systems monitoring and statistics data acquisition in an open-source world, there’s a very good chance (~90%) that you’ve had to painfully deal with this software.

Sporadic shutdown adventure — finale

It’s been a while since I’ve written about this issue. Folks might have concluded that re-imaging (reinstalling the OS) my machine fixed the problem.

Well, it didn’t.

Things worked for about 4-5 days, or so I thought. Then suddenly one morning while doing backups (from the local hard disk to a USB-connected hard disk), the machine sporadically shut off and stayed off. I had no 3D applications/games running, so this worried me even more — plus the machine didn’t power back on. Manually powering the system on worked fine, where it came back up and remained up for about 5-10 minutes more before losing power (but this time, starting back up on its own).

Sporadic shutdown adventure — day 2

Day 2 of my super-happy-fun-time adventure with PC hardware.

The new PSU hasn’t arrived yet, but as previously stated I did go about rebuilding the system to some degree (sans hardware replacements). More specifically, the following things were performed:

Sporadic shutdown adventure — day 1

Starting yesterday, my main workstation at home has begun to suffer from random power loss (abrupt system shutdown). The system “reboots” (powers back up) on its own due to my BIOS configuration having a power-on setting of “Last State”.

I wanted to document how I went about troubleshooting this issue and what I did to solve it. At the time of this writing I have no confirmation of what the problem part is, but over time I hope to figure it out.

The problem is directly related to GPU use — more specifically, 3D games. memtestG80 and memtestCL, which are used for testing VRAM/GDDR RAM on video cards (and to some degree stress-testing the GPU itself) works with flying colours for hours, yet within 1-15 minutes of loading any sort of 3D game the system abruptly loses power. memtest86 (for testing system RAM) also runs fine.

Firefox 5 and print-related context menus

Since switching to Firefox many years ago I’ve had to deal with the idiocy that is the lack of “Print” and “Print Preview” context menus. For those not familiar with the term “context menu”, I’m referring to what you see when you right-click somewhere on a web page that isn’t a link; you know, Back, Forward, Reload, Stop, Bookmark This Page, etc… Some of the Firefox developers feel that Control-P is sufficient (possibly the same developers who though removing Control-E to switch text input focus to the Search Bar was an intelligent idea?).

Every time there’s a release I have to go through the annoyingly repetitious process of finding an Addon that addresses the lack of said context menu. With the release of Firefox 5, there are absolutely none which work with it — except one, which I’ll save for last.

GV-N560OC-1GI hardware failure

Back in January I posted a review of Gigabyte’s latest GTX 560 Ti card. My opinions of the card have not changed during the past 4 months, and even now have not changed.

Sadly, however, starting sometime this week my card began to exhibit very strange problems — but only on 2D (or presumed to be 2D) surfaces. It’s not something easily noticeable in a 3D-based game, but playing something like Dwarfs!? made it apparent within seconds. Then I tried Two Worlds (which uses a 2D texture for its sky) and noticed occasional problems there too.

Later, I ran Misha’s Video Memory Tester application which confirmed that the likely source of the problem was bad GDDR5 RAM on the card, or a bad bus/path between the RAM and the GPU itself.

SCFH DSF now open-source

As of 2011/04/29, the SCFH DSF software has been made open-source and is available via github:

http://mosax.sakura.ne.jp/fswiki.cgi?page=SCFH+DSF+Dev

The translated bullet items on the above page (as of 2011/04/29) read as follows, acting as a “to-do” list:

How not to repair hard disks

In my search for some information about hard disk manufacturing processes, I came across this absolute beauty and had a good laugh.

I think one of these days I might have to mimic the above while doing SMART analysis and see what the drive claims is going bad. ;-)

Gigabyte GTX 560 Ti (GV-N560OC-1GI) vs. 9800GT review

Note: Folks reading the below review may also want to read my entry titled GV-N560OC-1GI hardware failure, where the same card after ~3-4 months of occasional gaming (6-7 hours a week) experienced artefacts as a result of GDDR5 RAM that went bad.

I’ve been using an nVidia GeForce 9800GT for quite some time now, and prior to that a 9600GT. When going from my 9600GT to my 9800GT, I was disappointed by the extreme jump in temperature, as well as power use — but the major performance increase made it worthwhile. I was also quite fond of my 9800GT since it was one of the BFG cards which didn’t have a PCIe power connector (it was powered completely off the PCIe bus) and only took up the space of one physical slot.

The reason I stayed with my 9800GT was that none of the newer cards impressed me. Sure, performance-wise they were superior, but I kept reading horror stories about temperatures reaching 90-100C (and people justifying that this is normal), in addition to fans reaching 55-65dB while the card was under load. Why were consumers accepting this garbage?

Gone from Facebook

For those whom I know in person or who otherwise communicate me through Facebook, be aware that I’ve deactivated my account in full. I’d been thinking about doing this for quite some time and decided that today would be a good day to do it.

Why? Too many reasons, but a few stood out from all the others.

At Adobe, HTTP redirection is hard

Well this sure explains why Firefox is claiming the redirection will result in an infinite loop…

$ curl -i -s -S <b>http://get.adobe.com/reader/</b>
HTTP/1.1 301 Moved Permanently
Server: Apache/2.0.63 (Unix)
<b>Location: http://get.adobe.com/reader/</b>
Keep-Alive: timeout=5, max=500
Content-Type: text/html; charset=iso-8859-1
Accept-Ranges: bytes
Connection: Keep-Alive
Date: Thu, 09 Sep 2010 04:47:10 GMT
Age: 248
Content-Length: 236

But every so often, you’ll get back something that works…

$ curl -i -s -S <b>http://get.adobe.com/reader/</b>
HTTP/1.1 302 Moved Temporarily
Date: Thu, 09 Sep 2010 04:47:11 GMT
Server: JRun Web Server
Content-Type: text/html; charset=UTF-8
Cache-Control: private, no-store, no-cache
Content-Language: en-US
Content-Language: en-US
<b>Location: /reader/otherversions/</b>
Set-Cookie: SETTINGS.LOCALE=en%5Fus;domain=.adobe.com;expires=Sat, 01-Sep-2040 04:47:11 GMT;path=/cfusion/
Set-Cookie: READER_HTTPREFERER=;domain=.adobe.com;path=/
Content-Length: 0
Connection: close

At Adobe, web services are hard. I wonder if they’ve been replacing their system administrators with monkeys.

Facebook falling over on its face

Keep trying, guys. You’ll get it right eventually. Web services in 2010 are hard.

$ date
Tue 31 Aug 2010 06:53:37 PDT
$ curl -v -i http://www.facebook.com/
* About to connect() to www.facebook.com port 80 (#0)
*   Trying 69.63.189.26... connected
* Connected to www.facebook.com (69.63.189.26) port 80 (#0)
> GET / HTTP/1.1
> User-Agent: curl/7.20.1 (amd64-portbld-freebsd8.1) libcurl/7.20.1 OpenSSL/0.9.8n zlib/1.2.3
> Host: www.facebook.com
> Accept: */*
>
* Closing connection #0
* Failure when receiving data from the peer
curl: (56) Failure when receiving data from the peer

Hmm, what’s happening on a TCP level?

# tcpdump -p -i em0 -l -n -s 8192 "port 80"
listening on em0, link-type EN10MB (Ethernet), capture size 65535 bytes
06:55:33.380449 IP 192.168.1.51.50324 > 69.63.189.26.80: Flags [S], seq 2661770843, win 65535, options [mss 1460,nop,wscale 3,sackOK,TS val 2499401659 ecr 0], length 0
06:55:33.482251 IP 69.63.189.26.80 > 192.168.1.51.50324: Flags [S.], seq 1390727061, ack 2661770844, win 4380, options [mss 1460,nop,wscale 0,nop,nop,TS val 3362678839 ecr 2499401659,sackOK,eol], length 0
06:55:33.482285 IP 192.168.1.51.50324 > 69.63.189.26.80: Flags [.], ack 1, win 16471, options [nop,nop,TS val 2499401761 ecr 3362678839], length 0
06:55:33.482416 IP 192.168.1.51.50324 > 69.63.189.26.80: Flags [P.], ack 1, win 16471, options [nop,nop,TS val 2499401761 ecr 3362678839], length 148
06:55:33.585188 IP 69.63.189.26.80 > 192.168.1.51.50324: Flags [R.], seq 1, ack 149, win 4528, length 0

Oh, TCP RST + ACK. Nice.