• John
  • Felde
  • University of Maryland
  • USA

Latest Posts

  • USLHC
  • USLHC
  • USA

  • James
  • Doherty
  • Open University
  • United Kingdom

Latest Posts

  • Andrea
  • Signori
  • Nikhef
  • Netherlands

Latest Posts

  • CERN
  • Geneva
  • Switzerland

Latest Posts

  • Aidan
  • Randle-Conde
  • Université Libre de Bruxelles
  • Belgium

Latest Posts

  • TRIUMF
  • Vancouver, BC
  • Canada

Latest Posts

  • Laura
  • Gladstone
  • MIT
  • USA

Latest Posts

  • Steven
  • Goldfarb
  • University of Michigan

Latest Posts

  • Fermilab
  • Batavia, IL
  • USA

Latest Posts

  • Seth
  • Zenz
  • Imperial College London
  • UK

Latest Posts

  • Nhan
  • Tran
  • Fermilab
  • USA

Latest Posts

  • Alex
  • Millar
  • University of Melbourne
  • Australia

Latest Posts

  • Ken
  • Bloom
  • USLHC
  • USA

Latest Posts

Homer Wolfe | The Ohio State University | USA

Read Bio

Kerberos at FNAL with Ubuntu 10.4 Lucid Lynx

Sunday, May 2nd, 2010

Sorry for the overly technical post, but I just wasted a half hour googling for something, and wanted to save others who log into FNAL some time.

Ubuntu 10.4 was just released last Thursday, and I know I’m not the only person at FNAL who’s using it on a personal machine, since I just got an email from a colleague of mine asking for help with the same problem I was having. If you try to get a kerberos ticket using the krb5.conf file given by FNAL computing division, you’re greeted with the error message:

kinit: No supported encryption types (config file error?) while getting initial credentials

This is because FNAL’s kerberos servers only uses single-DES encryption, which is considered “weak” by kerberos version 1.8.1 and up. It’s disabled implicitly by default. To work around this, open your /etc/krb5.conf file, and look for the section headed by the tag [libdefaults]. Add the line

allow_weak_crypto = true

And it should work like it did with older versions of krb5.

Share

Do satire news papers have “Letters to the Editor” sections?

Wednesday, November 4th, 2009

Pardon me for taking something too seriously:

A lot of my friends are passing around this article from The Onion.  It is really funny, as usual, but there’s an annoying error in it.   Yes, I know the whole article is intended as a joke, but not every untrue thing is funny.  I’m talking about a mistake with units of measurement, which is pretty much like a grammatical error, except that it actually matters, as I’ll explain below.  The article is about congress approving a monument which is a tribute to human folly.  Its got a big energy-wasting light at the top, and to make it sound like its really wasting a huge amount of energy, they wrote that it will “use a staggering 12 gigawatts of power per second.”   I’m not saying the number it too big:  Nuclear power plants produce about a gigawatt of power, so yeah, the number 12 is on the funny side of big.  What I’m saying is that a “gigawatt per second” isn’t what they wanted.

Power is the rate of energy usage, or transferral or whatever.  A 60-watt rated lightbulb consumes about 60-watts of power.   You could also say that the same bulb  consumes about 60 Joules of energy per second,  but a watt per second would describe something weird like the rate that the bulb’s energy consumption would change over time.   The rate of the rate of energy use.  Thats wrong in this context, but not funny-wrong.

Why am I so annoyed by this?  Well, when I was learning physics in school, I realized that if I kept the units of a quantity right, I could solve lots of problems without ever memorizing equations.  Its not just a trick or a mnemonic.  It basically gives structure to how one thinks about a problem, and its pretty powerful.  This type of thing is called dimensional analysis, and real physicists use it to guide them in research.    I have an awful memory, and can’t memorize equations or anything, but dimensional analysis basically got me through the Subject GRE and the Qualifier Exams.   If you want to learn physics, or if you just want something to help you when think about any kind of quantitative stuff, dimensional analysis is a big help.

Power describes a change in energy per change in time.  There.   I’m done.  Go read the article.  Its funny.

PS: Thanks to my friend Tim for pointing this out.

Share

Back at the Home Institute.

Tuesday, September 29th, 2009

I’m in CMH airport in Columbus now, waiting for my plane back to Chicago, Fermilab, and home.

I really should have cut this closer, since I have an hour to go, and I left 15 minutes early from a really great colloquium by Sean Carroll at OSU.  He was talking about topics from his yet-unreleased book, From Eternity To Here,  specifically the arrow of time.  In honor of this, I’m writing this post chronologically backwards.

The talk was fun, and you can see the slides from it here, although you miss the humor of him reading scrawled hate mail from a 10 year old out loud in a serious tone.  The author of said hate mail was irate about the possibility of his universe fluctuating in and out of the vacuum or some such.  In general, the talk was full of historical anecdotes and broad conceptual strokes, which I think is a great tack for a talk to a physics department.

The earlier part of today I spent working with the two grad students from my group who are working on CDF, and running an early stage meeting on developing software for a Higgs search I work on.  Right now, there are a couple of distinct analyses going on at CDF that look for a Higgs bosons which are produced in the same collisions as a W boson.  These separate searches share a lot of the same methods, but they intentionally differ in the way that they estimate the likelihood that a given event is from a Higgs instead of other, less exciting processes.  The problem is that these separate studies were developed by separate people using separate code, so that the other steps of the search are similar, but not really identical.  Some of the differences are trivial, some are just matters of taste, and some of them are subtle and generate a lot of debate.  They are all pretty robust and reliable analyses, but it would save a lot of effort and make things like combination studies go way smoother if we standardized some of this.

Last night I went out in Columbus with a few of the grad students from my group and some of their friends.  It really made me miss being in a college town, since the Tevatron is really out in the suburbs.

Monday was the whole reason I made this trip:  The Department of Energy Review.  My salary, and a large fraction of High Energy Physics in this country is funded by the DOE.  My advisors were given a grant to do particular research, including a chunk for postdocs like myself, and, once a fiscal year, they send someone to review the progress of a group and make sure they are meeting the goals of the grant.  The reviewer in question, so I’m told, is a physicist originally from SLAC.  I can’t confirm this myself, since I didn’t speak to him except during my presentation.  The presentation was actually pretty fun.  I work for a good group who has a long history at CDF and top/Higgs searches, so the presentations were full of accomplishments and big hopes for improving our analyses.  No telling how the funding will go, since I’m pretty sure the DOE budget itself isn’t really settled.

While I work for two professors at OSU, I spend most of my time at Fermilab keeping our part of CDF running, and working on analyses.  I’ve only had this postdoc for eight months are so, and, not counting the interview, I think I’ve been in Columbus only four times.  Luckily, there’s a good set of flights that I normally come in 8am one day, work till 7pm or so, crash at the closest hotel, back in the office at 8am, and fly out at 7pm or so the following day.   Its a bit draning, but it gets things moving the way that video conferencing whishes it could.

Share

The Tevatron is almost up again!

Monday, September 14th, 2009

For the last 10 weeks or so, the Tevatron has been off for scheduled maintenance.  During this time I could focus more on data analysis and I got paged a lot less, but I really missed taking data.  Accelerator Division has been doing machine studies the last few days, checking out their tune-ups and repairs, and today they went through most of the steps to making the proton-antiproton collisions which CDF and D0 observe and record.  I should mention that they weren’t scheduled to make collisions until the 19th, but it seems things were going ahead of schedule.  At 4pm or so, today, when the Tevatron main control room declared “Shot Setup” I hung out in the CDF control room for a bit, hoping to be around when the first collisions happened.  This was partially because a grad student from my research group is on shift for the first time, partially because I figured I’d just get paged when the run started anyway.  Really, I just miss the feeling of being where the action starts, just a few floors above the collisions, and surrounded by all the DAQ computers, crunching away.  Looks like there won’t be any collisions tonight, but we’re still ahead of schedule.

You can follow the progress of the Tevatron on twitter here.

which is basically as good and fresh a source of info as the official channels.

Share

First Post!

Wednesday, September 9th, 2009

Greetings, Internet!

This is my first post with Quantum Diaries. Its 6pm local, and I’m leaving the office right now, but I decided that it would be better to have a minimal-content post than none at all.

Share