Log in

No account? Create an account

Two Years Abroad — LiveJournal


In the last few days... Jul. 20th, 2008 @ 03:24 pm
1) The motherboard on my primary laptop died, six weeks after the end of a four-year warranty. I cannot renew the warranty -- because Dell only offers warranties up to 5 years, only sells warranty periods in units of one year, and the laptop is four years plus six weeks old.

I bought a new motherboard to install. It's supposed to be on its way by now but I haven't heard.

2) My Mac Mini has repeatedly hung with CPU failures ("cpu #1 [of a dual core] has stopped responding to interrupts.")

3) My Drobo (NAS drive) died, taking every bit of data I own with it. It turns out there was massive filesystem corruption, taking ~10% of my files with it; the rest is being recovered with Disk Warrior.

4) The 120GB external drive I first used to try to make a backup of my laptop died (just a month old! Shipped back under warranty.)

5) One of the 300GB drives making up my Addonics tower has died.

All of this has happened since Thursday.

Don't read this blog entry -- my mojo might be contagious.
Current Mood: stunned and depressed

P.S. Performance of Addonics w/ Software RAID vs a RAID NAS Apr. 19th, 2008 @ 08:09 pm
Again, a USB2-connected (through a PCMCIA card) storage tower vs. a dedicated Intel NAS box.
The Addonics tower is using three drives, while the NAS uses 4 drives; both are RAID-5.

tdb@bow:~$ time dd if=/dev/zero of=/media/AddonicsRaid/zerofile bs=1024 count=1000000
1000000+0 records in
1000000+0 records out
1024000000 bytes (1.0 GB) copied, 198.193 seconds, 5.2 MB/s

real    3m18.889s
user    0m0.632s
sys     0m12.473s
tdb@bow:~$ time dd if=/dev/zero of=/media/NAS/zerofile bs=1024 count=1000000
1000000+0 records in
1000000+0 records out
1024000000 bytes (1.0 GB) copied, 1334.64 seconds, 767 kB/s

real    22m15.187s
user    0m1.176s
sys     0m51.887s

The NAS drive has gigabit ethernet but the server only has 100M, so that's a wash. For comparison, from another server to the NAS drive using wired gigabit, not much of an improvement due to the networking:
arrow:~ tdb$ time dd if=/dev/zero of=/Volumes/public//zerofile2 bs=1024 count=1000000
1000000+0 records in
1000000+0 records out
1024000000 bytes transferred in 1111.925912 secs (920925 bytes/sec)

real    18m31.995s
user    0m4.679s
sys     0m57.112s
arrow:~ tdb$

Update: For completeness' sake, I did the same test from the other server to the shared Addonics Raid, and was surprised:

arrow:~ tdb$ time dd if=/dev/zero of=/Volumes/AddonicsRaid/zerofile4 bs=1024 count=1000000
1000000+0 records in
1000000+0 records out
1024000000 bytes transferred in 496.690487 secs (2061646 bytes/sec)

real    8m16.797s
user    0m4.571s
sys     0m54.870s
arrow:~ tdb$

So the complete matrix:

(times in minutes)
to Addonics tower
w/Software RAID5
to Intel NAS
w/GB ethernet
BOW, USB to Addonics, 100Mb net
ARROW, GB ethernet

Not what I expected to see at all.

My experience using Linux software RAID on my Addonics storage tower. Apr. 18th, 2008 @ 08:22 pm
Note: During this process, I emailed Addonics Tech Support for assistance, and in addition to their (quick) email response I received a phone call the next day to discuss the situation. Apparently this was a configuration they had not encountered before and so I offered to write up the results. (If you are not here for that technical information, now would be a good time to leave. :) )

  Addonics JBOD USB tower with four drives (2x300GB, 1x320GB, 1x180GB).
  IBM Thinkpad Pentium 3-M laptop with USB 2.0 served through a PCMCIA card.

  Xubuntu 7.10 with all updates as of 4/2008.

  RAID 5 with the first three drives (around 300GB each.)



Attempt 1: At first I wasn't aware that there was a lot of initializing to do for the array, and attempted to continue to mkfs as soon as the mdadm command completed. The mkfs repeatedly failed:

root@bow:~# mkfs -t xfs /dev/md0
mkfs.xfs: pwrite64 failed: Input/output error

...or ...

root@bow:~# mkfs -t ext3 /dev/md0
Warning: could not erase sector 2: Attempt to write block from filesystem resulted in short write
Warning: could not read block 0: Attempt to read block from filesystem resulted in short read
Warning: could not erase sector 0: Attempt to write block from filesystem resulted in short write
ext2fs_update_bb_inode: Illegal triply indirect block found while setting bad block inode

Attempt 2: Around this time I noticed that the CPU meter was still pegged even though I wasn't doing anything. Seeing mdadm still running made me look up mdadm's man page, which directed me to /proc/mdstat, which indicated the initialization was still proceeding:

tdb@bow:~$ cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md0 : active raid5 sdb1[3] sdc1[1] sdd1[0]
      586067072 blocks level 5, 64k chunk, algorithm 2 [3/2] [UU_]
      [===>.................]  recovery = 16.4% (48195320/293033536) finish=1010.3min speed=4035K/sec

unused devices: 

Eventually I noticed that the recovery percentage never went above around 2% (unlike the above sample, taken later) before getting back to a completed state:
tdb@bow:~$ cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md0 : active raid5 sdb1[2] sdc1[1] sdd1[0]
      586067072 blocks level 5, 64k chunk, algorithm 2 [3/3] [UUU]

unused devices: 

This process took about 10-15 minutes. Since there were no error messages in /proc/mdstat or dmesg, I assumed that the work was all front-loaded and the "other" 98% of the work was trivial, like setting up a partition table, rather than calculating and writing parity blocks. However, the errors from mkfs didn't change, and the result was still unmountable. I also tried altering the partition type from fd (Linux raid automount) to 83 (plain Linux) but there was no effect.

Attempt 3: At this point I was starting to wonder if one of the drives had gone bad, although I wasn't sold on the idea because they had worked right up to the start of this process. Still, I repartitioned the drives to start at cylinder 100 (leaving the first 100 cylinders blank on each disk) to skip past any initial problem spots on the drives. There were still no changes to the error messages, indicating the problem did not shift geometry with the partition change, which in turn suggested that the problem was probably not physical.

Attempt 4: I emailed Addonics tech support. Though they had not previously encountered my configuration, within that conversation there was a clue that the creation of a RAID 5 array takes a long time ("3 500GB drives takes as much as 24 hours.") I went back to thinking of the mdadm command as the culprit, wondering why my initialization step terminated at 2%. Again, attempting to force some kind of change in geometry, I recreated the RAID array with the drives in reverse order: sdd1, sdc1, sdb1.

This worked. Completely. I was rather gobsmacked. I was able to create an XFS filesystem and mount it.

I have since repartitioned the drives back to use the complete drive geometry (20gb is wasted on one drive, shucky-darn) and everything is still working fine. I also note that the partition type (83 or fd) doesn't seem to affect the creation of the array, although it may have other consequences for mdadm and related tools (don't know yet.)
tdb@bow:~$ mount
/dev/md0 on /media/AddonicsRaid type xfs (rw,noexec,nosuid,nodev)
tdb@bow:~$ df -h /media/AddonicsRaid
Filesystem            Size  Used Avail Use% Mounted on
/dev/md0              559G  1.5M  559G   1% /media/AddonicsRaid
tdb@bow:~$ sudo mdadm --query --detail /dev/md0
        Version : 00.90.03
  Creation Time : Fri Apr 18 10:56:23 2008
     Raid Level : raid5
     Array Size : 586067072 (558.92 GiB 600.13 GB)
  Used Dev Size : 293033536 (279.46 GiB 300.07 GB)
   Raid Devices : 3
  Total Devices : 3
Preferred Minor : 0
    Persistence : Superblock is persistent

    Update Time : Sat Apr 19 08:44:38 2008
          State : clean
 Active Devices : 3
Working Devices : 3
 Failed Devices : 0
  Spare Devices : 0

         Layout : left-symmetric
     Chunk Size : 64K

           UUID : b3c7a2c5:d065d41c:80d35bfa:21d4c790 (local to host bow)
         Events : 0.4

    Number   Major   Minor   RaidDevice State
       0       8       49        0      active sync   /dev/sdd1
       1       8       33        1      active sync   /dev/sdc1
       2       8       17        2      active sync   /dev/sdb1
tdb@bow:~$ touch /media/AddonicsRaid/testit
tdb@bow:~$ ls /media/AddonicsRaid

Et voila! A new use for those old drives. See next entry for relative performance measurements...
Current Mood: accomplished

Is it 1984 yet? Mar. 26th, 2008 @ 11:15 pm
[full article from the Seattle Times is here]

The unsettling thing about living in a surveillance society isn't just that you're being watched. It's that you have no idea.

That's what struck me about a story told last week by a border agent at a meeting of 200 San Juan Islanders. He was there to explain why the federal government is doing citizenship checks on domestic ferry runs.But near the end, while trying to convince the skeptical audience that the point is to root out terrorists, not fish for wrongdoing among the citizenry, deputy chief Joe Giuliano let loose with a tale straight out of "Dr. Strangelove."

It turns out the feds have been monitoring Interstate 5 for nuclear "dirty bombs." They do it with radiation detectors so sensitive it led to the following incident.

"Vehicle goes by at 70 miles per hour," Giuliano told the crowd. "Agent is in the median, a good 80 feet away from the traffic. Signal went off and identified an isotope [in the passing car]."

The agent raced after the car, pulling it over not far from the monitoring spot (near the Bow-Edison exit, 18 miles south of Bellingham). The agent questioned the driver, then did a cursory search of the car, Giuliano said.

Did he find a nuke?

"Turned out to be a cat with cancer that had undergone a radiological treatment three days earlier," Giuliano said.

He added: "That's the type of technology we have that's going on in the background. You don't see it. If I hadn't told you about it, you'd never know it was there."

The Importance of Weather Forecasts Mar. 22nd, 2008 @ 09:56 pm
... or, How Not to Discover the End of Ice Fishing Season!

(via The Daily Irrelevant)
Current Mood: laughing

Aaaaaaaand we're back! Nov. 19th, 2007 @ 07:51 pm
Yes I am back home -- sharing living quarters with Laurie, eating like a pig, and consulting to start building up the bank account again. My life in a nutshell. :) Perhaps at some other time I will expand on that, but everything else is trivial (i.e. "it's all small stuff". :) )

The reason for reactivation of this blog: my thesis project has rather come to a grinding halt... many reasons but the end result is the same, and I am looking for a new topic. One which shows promise is something to do with the One Laptop per Child project -- which for those of you who may not have noticed, you can get one of your own, right now, for one more week. If you want one and don't live in the U.S. or Canada, let me know, and I'll order it for you and forward it. (Price with shipping to me is $423 ("Give one get one", you're really buying two of them; one goes to you and one goes to a kid somewhere) and I'll forward it for free.)

Anyhoo, I've got mine on order (so does Laurie :) ) and I've also got simulators up and running in VMware. Cool stuff, very different.

I'm posting this primarily to track some oddities about what I'm seeing so that I can track them down again later.

  1. Some installations of a new VMware virtual machine result in a repeatedly crashing X server.

    The failure occurs (obviously) during X startup; the screen turns white, the X cursor appears... but about at the point where a working copy's cursor changes to the large arrow, the failing ones crash.

    I've made about ten copies of this install image. Half of them worked, half of them didn't. The only fix I've found is to delete the image and recopy from the template.

  2. I have four simulations running. One of them has created a shared document, two others are participating in the writing. In the fourth simulation's Neighborhood view I should see a document with three (or two, if the document counts as one itself) XOs clustered around it. I see the document but never the never see that many, only zero or one. This does not appear to have anything to do with whether one or the other has designated the viewer or the document owner as a "friend".

  3. When one simulation attempts to connect to a shared document, sometimes it gets the correct document, but other times it gets a blank new document with the shared document's XO colors.

  4. Something closes with a hard crash/coredump during shutdown/reboot. Can't tell what's doing it, though, the diagnostics are more than a screen long.

I don't know if these are artifacts of the fact that I'm running in a VMware simulation or not. We'll find out when the machine comes...

Other bits:
  • I still haven't figured out what "groups" are.

  • I want to find out more about the school servers.

  • OLPC has a job listing for a "Back End Engineer" that leads with, "Are you interested in problems of scale?" :D I've sent them a resume.

Why I Hate My Net Connection: Jun. 11th, 2007 @ 02:18 am

'Nuff said.
Current Mood: irritated

Annnnnnnd IT'S OFFICIAL... May. 2nd, 2007 @ 11:15 pm
I have earned all the credits and passed all the courses I need for the degree, other than the thesis itself!

[insert sound of Kermit the Frog cheering...]
Thank you, thank you... and I even have one extra credit (about 1/6th of a typical class credit) left over...

Received the last outstanding grade today. In total, a credit-weighted average grade of 7.8. I am very happy to report that of the 14 classes I ended up taking, I anticipate 9 having the strong possibility of being directly useful in my ideal job (whatever that is,) which is an even better rate than I anticipated -- vs. 50% -- and far higher than if I had taken any of the distributed computing degree tracks I saw in the U.S.

So I guess all this means that my time in Amsterdam is winding down. Like most transitions, this is both good and bad, but overall I'm looking forward to getting back.

Sorry for the long hiatus here but, while lots of things have been going on, none of them have been particularly new or noteworthy...

- First, of course, there are the typical winter doldrums which seem to be exacerbated by a near-total lack of sunlight for months on end...
- I took a visit back to NH in February just because I was feeling a bit homesick...
- Continuing struggles with Moxie's health, although I think she's turned the corner and is gaining a bit of weight back...

The exception to the rule: My thesis has started up! I am indeed doing the NEW-TIES related project in Scotland, and will be back and forth there a couple of times -- I've already gone once, but no pictures to speak of. I'll be staying for a weekend at some point and hope to get some shots then. In the meantime, for those interested in following the thesis in detail, there is a new blog at http://pdcs-ttg.livejournal.com/ which I am using to document events, milestones, decisions, etc. for posterior's sake.

But worth posting are these videos I took of Queen's Day celebrations two days ago -- I went out and bought some orange hair and saw some funny stuff...

First, a demonstration of the great respect and affection which the Dutch hold for their royalty:

And second, traffic jam on the canals!

Current Mood: Pleased with the milestone

From Batty to Evil Jan. 10th, 2007 @ 07:11 pm
I shouldn't be posting this because I have an exam tomorrow but I already know I'll be up late anyway... There's a news story I've been following because of some issues of personal interest to me, and I ran across some comments which were so jaw-dropping I couldn't resist exposing them to the light of day.

There's a deaf kid in New York who has gotten into a legal fight with his school about whether he can bring his new service dog to school with him (original story here.) Most of you know that I'm deaf in one ear and some of you know that my parents raise service dogs, so I'm pretty familiar with most of the issues involved.

I can understand why the school is unhappy about the dog -- they are worried about it being a disruption and a liability (both of which are ludicrous, and only expose their ignorance of service dogs - the mother, when asked about the dog's behavior during the heated argument, said, "He behaved himself impeccably, which is more than can be said for the adults, including myself, I might add.") and they may also be worried about the perception by others that this kid is getting special treatment by bringing a "pet". All of this should be fixable simply with everybody getting together and getting educated on the subject. There are other logistical issues around this -- what if some other kid is allergic, and apparently the school was asking for some kind of bureaucratic review which the mother didn't pay any attention to -- but the bottom line is simple: the dog needs to be with his person, and the law says the dog can be there. That's all, there ain't no more.

But I ran across a reference to this story on FreeRepublic today (the lunatic fringe right-wing site) and the comments on their story range from reasonably aware to "ignorant and proud of it" to downright evil.

A lot of people with hearing problems would really get angry to be called disabled. Kid has a pet he wants to bring to school, get over it kid.

I'm getting the feeling this mother is fishing for a lawsuit against the school (using current state and federal special education laws).

He is handicapped beyond mainstream and should be in a schools for the deaf. His mother refuses to admit she birthed a defective.

Admittedly this is a one-sided sampling, but it's the majority side.

I just wanted to take this moment out to thank my parents for raising me right. :)
Current Mood: busy

Mouth, meet money. Money, meet mouth. Dec. 16th, 2006 @ 11:10 pm
I've been in Amsterdam for a year and a half now, and in terms of Dutch food I've primarily been exposed to snackish stuff:

  • french fries with mayonnaise-like sauce

  • the "cheese souffle", cheese wrapped in a packet of something which is sort of halfway between pasta and dough, then breaded and fried

  • and of course herring -- pretty good smoked, but that's hard to find in town and I can't get around the pickled ones which are everywhere.

  • and recently, poffertjes, basically tiny, slightly chewy pancakes of no particular taste other than they're served with a big hunk of butter and covered with powdered suger.

But I'd never really had any of the few "traditional" Dutch main courses. I've been rather put off, not only by the quality of the food I have had, but by their descriptions (as in here for stamppot specifically, or here for a general food review I posted this time last year.) And to give it a really fair test, I should go to a place recommended by a Dutch person, right?

Well, I broached the subject with a Dutch friend and he recommended a place called "De Keuken" (The Kitchen) at Spuistraat 4 near Central Station. So some classmates and I checked the place out. I, of course, neglected to bring a camera, but not everybody was as silly as that (thanks Ann!), so I give you stamppot:

click for larger image

  • Cook and mash some potatoes (no milk, just mash them up, and leave them chunky)

  • Mix in some chopped lettuce (raw, as opposed to the cooked cabbage I saw in the prior description)

  • Sprinkle with pieces of bacon (lots -- this is the only part of the dish with any flavor)

  • Add a hunk of sausage -- not spicy, not smoked, just sort of plain

  • Cover it all with a little thin brown gravy

Presto, stamppot.

click for larger image

See? Not as bad as I thought it might have been...

You may notice the two soda bottles behind me? 0.2L each -- together they're just slightly larger than one 12 ounce soda can. I could have gone through a lot of those... but they're US$1.75 apiece, so I restrained myself.

So all in all it was a good night, and the Dutch food was not as bad as I thought it might be. But "high cuisine" it ain't.
Current Mood: chipper
Top of Page Powered by LiveJournal.com