Archive for the 'Tech' Category

Big Blog Headers

31 March 2009

Screenshots of other blogs, as seen on my netbook:

Big Blog Header Big Blog Header

I don’t know what value this header ads, but I find it annoying.

Bacon Conservation

7 November 2008

In the distant past, I did the right thing; despite my best efforts, I was no longer able to ignore it.

What we have here is a failure to communicate

Previously, I bought a copy of the Indeo vodec; I started scrubbing my Archive array for content to transcode. Instead, I found copies of stuff I no longer care about. While pruning, I noticed the amount of space used on the Archive array was getting close to 500 gigs; if I could prune a few more hundred gigs, I could break up the RAID-5 array and make a couple striped mirrors. I discarded this plan, as it wouldn’t create any more redundancy; two drive failures out of four could still cause data loss. While transcoding, I noticed a lot of drive activity I wasn’t expecting.

Then everything went sideways

An awful noise started emitting from my computer. Previously, I had blamed my craptastic UPS for this noise; however, I’d taken it out of the loop. I did an orderly reboot; on reboot, the noise stopped. I thought about digging around in BIOS to see if there was a temperature alarm, but I missed the window in which I could have hit F2; instead, I was greeted with ARRAY FAILED and a countdown.

Previously, I had shut the computer down, turned it on to no noise, and walked away while it rebooted. Long enough, apparently, to miss the ARRAY FAILED message.

If only I’d paid attention

Over the last month or so, I noticed a lot of drive array activity at strange times. I’d check for IO in taskmanager, but there would be no processes with a history of IO. I figured this was a background task; in a way, it was. The RAID card was moving sectors around to handle failures; since the IO was masked at the hardware level, it didn’t show up.

The array was rebuilt at least four times, taking five hours each time. For some reason, it only alarmed audibly twice. The first time I blamed my UPS (the noise was the same) and replaced it. This time, I knew it wasn’t the UPS.

Recovery

I’d like to thank the bright bulb at Seagate that put serial number stickers on the front of drives; this made finding the failed drive much easier. After the drive was replaced, everything rebuilt and I’ve been A-OK for a week. The RMA replacement drive should be here next week.

I enabled the “email on event” tool in the driver for the RAID card. Had I enabled this before, I would have had plenty of warning about SMART errors before the failure; instead, I opted to fly blind.

The one right thing I did when I set everything up? I enabled the audible alarm on my RAID controller. Had I not done that, I probably wouldn’t have noticed until a second drive failed and I lost everything.

Vista CPU upgrade mystery

22 July 2008

My CPU died; I had to peel off $80 for a replacement. After I installed the replacement, I started getting these errors:

Product: Windows Media Player Network Sharing Service
Problem: Stopped working
Date: 7/22/2008 7:02 PM
Status: Not Reported

Problem signature
Problem Event Name:    APPCRASH
Application Name:    wmpnetwk.exe
Application Version:    11.0.6001.7000
Application Timestamp:    47919370
Fault Module Name:    Indiv01.key
Fault Module Version:    11.0.6000.6324
Fault Module Timestamp:    47e31c1d
Exception Code:    c0000005
Exception Offset:    00107422
OS Version:    6.0.6001.2.1.0.256.1
Locale ID:    1033
Additional Information 1:    375c
Additional Information 2:    b609c209d0d5efeb3cfcb425b7f84343
Additional Information 3:    375c
Additional Information 4:    b609c209d0d5efeb3cfcb425b7f84343


Product: Windows Media Center Receiver Service
Problem: Stopped working
Date: 7/22/2008 7:02 PM
Status: Not ReportedProblem signature
Problem Event Name:    APPCRASH
Application Name:    ehRecvr.exe
Application Version:    6.0.6001.18000
Application Timestamp:    47919381
Fault Module Name:    Indiv01.key
Fault Module Version:    11.0.6000.6324
Fault Module Timestamp:    47e31c1d
Exception Code:    c0000005
Exception Offset:    0010275e
OS Version:    6.0.6001.2.1.0.256.1
Locale ID:    1033
Additional Information 1:    fd00
Additional Information 2:    ea6f5fe8924aaa756324d57f87834160
Additional Information 3:    fd00
Additional Information 4:    ea6f5fe8924aaa756324d57f87834160

Product: Digital Cable device registration application.
Problem: Stopped working
Date: 7/22/2008 7:02 PM
Status: Report SentProblem signature
Problem Event Name:    APPCRASH
Application Name:    ehPrivJob.exe
Application Version:    6.0.6001.18000
Application Timestamp:    479192da
Fault Module Name:    Indiv01.key
Fault Module Version:    11.0.6000.6324
Fault Module Timestamp:    47e31c1d
Exception Code:    c0000005
Exception Offset:    00111e04
OS Version:    6.0.6001.2.1.0.256.1
Locale ID:    1033
Additional Information 1:    d947
Additional Information 2:    9e34b879961a9cb241047333b26275f7
Additional Information 3:    59fd
Additional Information 4:    ea834e34771542fd8e86caf9cf1343ce

Extra information about the problem
Bucket ID:    855048358

I found the solution after a lot of searching. In a nutshell, a new CPU requires a new DRM store. To effect this, start an elevated cmd prompt and:

Microsoft Windows [Version 6.0.6001]
Copyright (c) 2006 Microsoft Corporation. All rights reserved.C:\Windows\system32>cd \ProgramData\Microsoft\Windows\DRM

C:\ProgramData\Microsoft\Windows\DRM>rd /s/q .
The process cannot access the file because it is being used by another process.

C:\ProgramData\Microsoft\Windows\DRM>

Warning: if you needed any of that data, it is gone. Hopefully the Vista AppCompat team can look into publishing a better solution more widely.

Airlines, telephony, and power

28 June 2008

Four links to Chicago Boyz in one day; such is the life of a shill. Carl has a great post up on, as he calls it, “the age of unreliability“. One link he missed, I think, is how much infrastructure could have been built up in the last seventy years if socialism wasn’t fucking our economy over.

No network provider accepted the given network path.

19 June 2008

2003 SP1 and 2008 have a new policy in place which prevents you from reaching some servers via UNC paths when you can otherwise ping them. To disable this policy, from a cmd window running as administrator, type:

REG ADD HKLM\SYSTEM\CurrentControlSet\Control\Lsa /V DisableLoopbackCheck /T REG_DWORD /F /D 1

IT Pro has more details.

Blackjack and Blackjack II hard reset

18 April 2008
  • pull battery
  • replace battery
  • hold DPAD-UP
  • power on
  • follow instructions

ASUS EEE PC customer care

21 March 2008

My EEE PC died three weeks ago. I finally got around to calling ASUS and getting an RMA. Total time from Ry to ASUS to Ry was about a week; turns out, instead of repairing my laptop, they just sent me a new one. I’m happy; an extra battery and charger is nice. I’d like to know what killed the SSD, though; the lady on the phone seemed to think it was the fan dying and the SSD getting baked.

Hopefully ASUS #2 doesn’t have the same fate in store.

Flo Rida: Apple fan

11 March 2008

Q: Tell me about ringtones. How do those royalty checks look?
Rida: Oh man, oh man. It’s a blessing. … I got two MacPros.
Q: What kind of technology do you see a need for? What does someone need to come up with to make your life easier?
Rida: I remember one time, I accidentally dropped my first iPod into the toilet. If they can make them waterproof, that’d be great.

Rising rap star doesn’t need RIAA

Aluminum foam

30 July 2007

Aluminum FoamAluminum Foam
Materials science is amazing stuff.

Tech stuff

9 April 2007

If you own a Zune (or an Xbox 360), a gentleman has released a tool called Encode360. It does all of the stuff you would expect a modern encoding tool to do, including watching a directory for files for transcoding.

In a subject near and dear to my heart, Drobo has released a $700 home storage robot. Kickass. As soon as these are shipping, I’ll buy a couple, stuff them with drives, and party on.