« July 2001 | Main | September 2001 »

August 2001 Archives

August 5, 2001

Magna cum laude, summa cum laude, the radio's too laude

OSCAR 1.1 has been released. Woo hoo!

Here's an embarrassing note: during the OSCAR teleconference this past week, we were plagued with all kinds of audio troubles with Intel's teleconferencing system. People would drop in and out, echos would abound, etc. But we still managed to have a reasonable conference.

The conference is normally scheduled for an hour. At the end of the hour, though, we weren't quite done. An automated announcement said, "To extend your conference for 15 more minutes, hit *9." Everyone agreed that we should continue to finish up the pending details, so I hit *9.

EEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE-BEEEEEEEEEEEEEEEEEEEEEEEEEEEP!

"Oh great," I thought, "More problems with Intel's #$#@%
teleconferencing system."

A split second later, I realized that that was my fax machine --
it's programmed to pick up if you hit *9 (handy when you only have one telephone line; you can answer from any phone in the house and make the fax machine pick up if it's an incoming fax, not a person).

I had to race down the hall and rip out the phone cord from my fax machine, and then come back and tell everyone what happened. How embarrassing. :-)


We also released LAM 6.6b1 that includes, among other things, a first cut at Myrinet support. It lacks some optimizations (doesn't pin user memory that is already pinned, doesn't use shmem for communications on the same node), but those will likely have to wait until post-dissertation.


My Linksys router box came. I got it setup nicely, such that it does selective IP forwarding to my back-end boxen. I found a handy feature in OpenSSH that allows it to listen on multiple ports for incoming connection. i.e., I don't have to muck around and have two different OpenSSH servers running , each sitting on a different point
-- OpenSSH allows this behavior just by editing a single config file and listing multiple ports. How cool is that?

Why is this important? shh normally accepts connections on port 22. With my DSL connection, I only have one fixed IP address. But I have two unix machines on my backend LAN that are generally on 24/7. I would like the ability to ssh directly to both of them from the greater internet. But there's only one port 22. So my linksys box forwards all incoming port 22 requests to one machine. But what about the other? This means that I have to pick some other port.

The bummer about my linksys router is that it will only IP forward on the same port -- so I can forward port addr1:port to addr2:port. I cannot forward addr1:port1 to addr2:port2. Bummer. So if I have two incoming ssh ports on my router, the second (non-standard) port has to be forwarded to the same port on a backend machine. This is where OpenSSH's feature comes in handy -- not only does it listen on port 22 for normal ssh connections (e.g., for connections from my internal LAN), it also listens on port N for connections from the greater internet. Cool!

I was short a cat 5 cable, though -- had to run out to Best Buy to get one.


ARRGGHH!!! It seems that I deleted all my pine mail for July 2001. How the heck did I do that? It must have been in the monthly archive on August 1. Doh. :-(

So I initiated a report with suggest@darwin.helios.nd.edu, and they actually restored the last backup (from July 31) within a few hours, and salvaged it all. Amazing.


Windoze sucks.

Oh, I'm sorry -- have I said that before?

I'll say it again: Windoze sucks.

The following describes a windoze "gotcha" that bit me on Friday. I know that most of the jjc readers have nothing to do with windoze; I describe it here mainly because a) I'll remember it this way, and b) you never know when you (jjc reader) may need to have a few 'doze sysadmin tricks up your sleeve.

My church just bought two new computers for staff members to replace some really aging computers (the old ones we so bad that they would swap almost continually, making any amount of work extremely hard to do). They were Gateway PIII 1GHZ machines (you really can't get much lower than that these days without going into Celeron country, which I highly recommend against!) with Windoze 2k. This now makes three w2k machines; the rest are all w98 and w95.

My church actually has a little LAN setup in their offices (I've described it in previous journal entries) with about a dozen machines on it. They do a few windoze shares to share some directories between machines for various databases and whatnot.

I had intended to spend 2-3 hours installing the two new computers, copying over the data from the old computers, installing the extra software that they needed, training the staff members in the differences with w2k, etc. I should have known better. <sigh>

Setting up the computers was easy enough; transferring the data, installing the extra software that they required, etc., wasn't too bad because the staff is actually fairly organized, and had all their data files in one place, etc. Yay for smart users! :-)

One weird thing, though, Printshop -- I think it was a fairly old version -- wouldn't work for regular users (i.e., not the "administrator") unless I installed it as the user. i.e., when I installed it as "administrator", it would give amorphous errors when a regular user tried to run it. I assume that this was because of permissions issues (I only had the CD case, not the original box, so I don't know it was supposed to support NT/2k or not -- I suspect not). Whatever. I temporarily bumped up the user's access level, installed the software, reduced the user's access level to its original state, and then all was well. <sigh>

But that wasn't too big of a deal; it only took an extra 15 minutes or so to figure out.

One of the two new machines was replacing a machine that previously shared one of its directories to the rest of the LAN. This is where my troubles really began.

There is no NT domain on the LAN -- those cost many thousands of
$$$! (before you scream "use Linux/Samba!", read the rest of this entry) So instead they just share a Windoze workgroup. It works well enough; we're talking about a staff that mainly does word processing, some spreadsheets, and a few databases --
nothing really fancy.

I setup the sharing on the w2k machine and then went to a w98 machine to try to mount the share (you know, check that it actually works. Sometimes this is a novel concept to IT support staff :-).

It asked for a password. WTF? It never required a password before (i.e., when the w98 box was the sharer). Not understanding why it was asking for a password, I tried a couple of obvious passwords that I thought it might be, all with no joy. Weird.

I went back and double checked all the sharing settings (permissions, etc.) on the w2k box, but everything looked fine. I went back and forth for quite a while, but could never get the w98 box to mount the share properly. Weird.

I called Johnny to see if he could help (it was about 6pm by this time). I described what I had done to him and he said that it sounded essentially correct. He was actually in a bookstore at that point, so he went over and pulled out a w2k book and looked it up, and indeed, I had everything setup the way that I should. Johnny had to run, so I continued on by myself. Unfortunately, this machine was a rather business-critical machine (more specifically, the share that it provides is rather business-critical), and I had to get it working. Bonk.

After much trial and error, I finally figured out what was going on:

  • The first important factor was that there is no NT domain. As such, there is no global authentication across all the machines. Indeed, there are only two accounts on each of the three w2k boxen: administrator and the user who sits down at that machine. This is an important fact.

  • w98 and w95 machines have no real concept of a user, so this had never mattered before. i.e., w9x sharer permissions are not based on the concept of a user.

  • When the w9x boxen tries to mount the share from the w2k box, it uses the username that the user "logged in" with (you know the "login" window that you can set w9x up with -- although you can hit ESC and skip it...). However, given that there is no global authentication on this LAN, that user will not exist on the sharing (w2k) machine.

  • In this situation, if the sharing machine is a w2k box, it will designate the share request as if it were coming from the "guest" account.

  • The "guest" account is initially set to "disabled" on w2k (which, although frustrating for me the other day, is actually a Good Idea). So I had to enable the guest account and assign a password to it. I then entered that password on the w98 box that was trying to mount the share, and it worked.

Woof. Stepping back, it all actually does make sense, but there were precious few clues for the uninitiated during the process to figure out what was going on. It would have helped immensely if the w98 box had shown the username that it was asking the password for. That would have tipped me off immediately. But it doesn't -- it just asks for a password.

Of course, there were at least two other alternatives that I could have done to solve this problem, but neither were attractive:

  • Setup a Linux box with samba as a primary NT domain controller, make all the windoze machines be clients in the domain, and then have all authentication centrally handled. The problem with this is that I'm not going to be in this parish forever, and I don't want to set them up with technology that they don't know how to maintain that they rely on for day-to-day business, and then leave them stranded when/if I move away from Louisville. Maybe someday, if it turns out that I'm going to be in Louisville for quite a long time. But not today.

  • I could have moved the share to a different machine (w98) and avoided all these problems, but a) someday all the machines in that office will be w2k and the problem will arise anyway, and b) there are actually political issues involved, so the share had to stay on that machine. :-)

So all in all, I'm not actually all that thrilled with the solution from a security standpoint. I had to enable the guest account to anonymously export the share. Granted, this is effectively no different than shares from a w9x box, so it is arguably no less secure that it was previously, but it still bugs me. And since there is no central authentication, I don't want to get in the business of maintaining separate accounts on all machines for every user -- that's an N^2 problem.

Grumble. Perhaps linux/samba is in their future someday, since there's no way that they could afford a real PDC license. Grumble.

August 9, 2001

This office isn't big enough for two weird guys!

I think that insects need to form a "no fly zone" and/or an air traffic control system over human automobile highways.

We'd all be much happier if they just organized and did this; my car wouldn't get nearly as dirty, and less bugs would die.

Heck, if nothing else, why doesn't Darwinism produce smarter bugs that instinctively know to stay away from highways?


Another question -- why does fast food taste so horrible when it's cold? That is, it tastes 37% worse than normal (non-fast food) does when it's cold. Why the disparity?

Does it always taste bad, but when it's hot, we're so concerned with not burning our mouths that it goes down so fast that we don't notice?


I tried switching to mozilla 0.9.3 'cause they claim it's more stable than netscape 4.77. Although it hasn't crashed on any of the things that mozilla has previously crashed on (SSL pages at USAA, LSC page, LAM/MPI page, etc.), it's still not 100% stable. I find myself switching back to Netscape periodically because mozilla won't load or render a page correctly. We'll see how it goes. I deleted oodles of cookies, and am denying cookies to advertising sites left and right, which is nice.

Here's a few bugs that I've noticed:

  • seems to have problems w/ typing in urls
  • doesn't bring up entering/exiting ssl notice until after page is
    fully loaded/displayed
  • doesn't always print properly, eg:
    http://www.geocrawler.com/lists/3/SourceForge/13443/100/6111455/
  • cache behavior seems funny; sometimes it loads, sometimes it doesn't
    get changes
  • sometimes it stalls resolving IP names, even though "dig" on the
    name responds immediately

It has a neat feature to block images from a given web server. My excite pages now have remarkably few advertising banners on them. Cool! (One has to be selective, though -- you can't just block all banners, because some of them actually come from servers were you do want to receive other images. Slashdot does this, for example).


Tracy and I watched the DVD of "The Emperor's New Groove" the other day. A fun little movie; it's all about Llamas. Being a Llama myself, I laughed a good deal at the llama jokes.


I got a new laptop, but then had to leave it at IU. Doh. :-( It seems that they hadn't gotten the paperwork for me to take the laptop home straightened out yet. Hopefully, it'll be worked out when I'm up there next week.

I configured it all up; it's great. 900Mhz PIII, 14.1 inch SXGA TFT (1400x1050 -- wow), 128MB RAM, 20GB. It came w/ 'doze ME, which I promptly erased (am I eligible for the MS refund?). I loaded up Mandrake 8.0 on it, as well as VMware, in which I loaded Windows 2k and Office 2k (bought at the IU bookstore for $5 and $10, respectively).

Some random things learned:

  • sendmail needs the hostname in /etc/hosts, and mandrake didn't put
    it in there (weird)
  • mandrake automatically correctly identified the "dma=1" switch for
    hdparm
  • 1400x1050 resolution looks great (had to do a web search and find
    someone who figured out the right XF86Config stuff)
  • vmware bridged networking doesn't play nicely with wireless. Doh.
    :-(
  • resierfs doesn't allow files greater than 2GB (e.g., vmware virtual
    disks). Doh. :-(


I finally caught up w/ Eileen while I was at IU (she and I were both at ND in grad school together; I probably only met her a handful of times at ND, but we have a bunch of mutual friends and common history at ND). We had a great dinner and corresponding conversation. I helped her pick out a new Dell laptop for work (she's faculty at IU in the Latin Studies department), and inadvertently convinced her to buy a Palm m100.

Her new m100 arrived today (Friday), and she's been playing with it. I've already sent her the , and strongly recommended the parens calculator and DateBk4 apps.

August 12, 2001

Dave tells me that there's lots of company policies that apply just to me

Scenes from the upcoming blockbuster movie, "Pushing Bugs", a gripping story of the stress and tension in the lives of Bug Traffic Controllers trying to provide safe passage of insects across human automobile highways:

BTC: "Junebug 357, you are clear to cross I-265 southbound lanes at mile marker 123.niner. Proceed immediately to cross at minimum safe altitude of niner feet to avoid approaching oncoming small Japanese vehicle, over."
JB357: "Roger."
...seconds later...
BTC: "JUNEBUG 357! Climb immediately to altitude of 25 feet; unscheduled incoming tractor trailer approaching!!"
JB357: "I see him! Climb...<splat>"
BTC: "Junebug 357! Are you there? Junebug 357! Report!!"


I've gotten oodles of hits from Code Red on my web server -- 141 of the CRI variety, and 669 of the CRII-and-beyond variety. I think the funniest two CRII hits that I've gotten are from:

 msgr-cs30.msgr.hotmail.com msgr-cs31.msgr.hotmail.com 

I kid you not; this is straight from my web logs. I had read articles that MS had some of their hotmail servers infected, but to see proof of it right in my own web logs is just too darn funny. :-)


Saw Gone in 60 Seconds tonight on DVD. I thoroughly enjoyed it; it was much better than I thought it would be. I wouldn't call it the greatest movie of all time, but I would say that it is definitely worth seeing. Good action, good humor, good psychos, and good character development. I've always like Nicholas Cage. I was surprised to see Angelina Joline in it, too (took me a little while to recognize her -- her hair is bleached, much longer than Tomb Raider, and styled entirely differently, sorta dreadlocky).

I give it 30 minutes.


It amazes me how other kinds of scientists and engineers use clearly sub-optimal methods when it comes to computers. This is sounds like an unjust elitist view, but hear me out anyway.

In looking for a good demo application for my dissertation code, I heard about some DNA sequencing code that uses MPI in a manager/worker model. Sounds about perfect. I download the latest copy and had a look at it. One thing that struck me right away (and this is definitely elitist) is that you have to manually edit the Makefile. Yuk.

So there's no configure script; I can deal with that. They have "MPI defaults" for you in the Makefile, but they are a) based on MPICH (that's like rubbing a cat the wrong way :-), and b) using the most difficult method rather than just the mpicc wrapper scripts that MPICH (and LAM!) provides.

Sidenote: I've seen lots of MPI projects that do this... why do they do that?

And they make other MPICH assumptions -- such as assuming that MPI will give the same command line arguments to all executables, even if it's an MPMD model. That one threw me for a while -- I wasn't expecting that (nor did I know that MPICH did that).

The overall design of the program itself is actually pretty clever, yet complicated -- it seems to be clearly written by some scientists/engineers who want to "get it working", and you gotta respect that. But it shows some naivete in its overcomplication (IMHO, mind you) -- I think that the overall model could be less complex. For example, there's a fairly elaborate scheme in place to get the data back and forth from the manager to the workers. But it involves four different binaries (three required binaries, and an extra optional "monitor" process).

Even though the overall program seems to work ok, it's not optimal. Yes, you get speedup running on multiple processors, but the speedup is less than linear. And it seems fairly complicated for a manager/worker setup. There's a separate "foreman" executable, for example, that relays work between the manager and the workers. More than that, though, the foreman spins on non-blocking probes. This eats up CPU cycles like nobody's business.

All that being said, this is actually quite advanced for non computer scientists (and for many computer scientists, as well!). It's probably on the forefront of its field in DNA sequencing codes. Note, however, that I make these observations about many computing projects that I have seen -- not just the one DNA code that I cited above; the DNA code is just the most recent example.

So what does this tell us? I guess that that's part of our jobs as computer scientists -- to make tools that other scientists/engineers can use to build complex systems. Tools that suck less than most current tools. Indeed, Lummy had a good observation recently:

One interesting insight I gathered while at the Livermore meeting last month was along these lines. There is a real reluctance in the scientific computing community to use C++ -- especially advanced C++ -- in scientific codes. The reason is not that the people are dim or lazy. Rather, the intellectual capacity taken up by (advanced) C++ leaves room for very little else. These guys also have to be experts in numerical analysis and their application area. Also being an expert in C++ is not really feasible.

The tools that are available are generally powerful, but most of them suck for various reasons (a typical example that many readers can probably associate with is how MS Windows periodically "freezes", or dies the Blue Screen of Death). Indeed, using the current generation of tools to their fullest power requires significant expenditures in terms of time and learning -- something that most people just don't have the resources to spend. How to reconcile the use of building complex systems without either rolling individual solutions or spending huge amounts of time learning complex tools?

An idea that I have been telling fellow scientists and engineers about for quite some time is what I euphamistically call "C+". It's not C, and it's not C++ -- it's somewhere in the middle.

Most scientists/engineers know C but are scared of C++ for exactly the reasons that Lummy cited above. I'm a big fan of essentially writing C code, but using a small number of the advanced tools in C++ such as: std::string, passing by reference, the STL (maps, vectors, and lists), and basic object usage (to get guaranteed initialization/destruction). You don't need to go into full-blown object-oriented design or use all the whacky, bleeding edge features of C++; the simple tools that I listed above are extremely powerful and provide oodles more functionality than you get in vanilla C. They actually save time when programming, and allow for elegant solutions to programming problems. The learning curve on these tools is actually quite low. These are good examples of software that suck a good deal less than most other tools.

As many have noted, the current state of technology in software is really in its infancy. Consider what used to run your computer 5 and 10 years ago. It was vastly different -- there are a few essential concepts that have remained constant, but software itself has changed dramatically over the last 10 years. What will it do in the next 10, 20, 50 years?

Indeed, computers are made for the kind of rote, menial tasks that software tools are supposed to provide for us. So why do I have to spend so much time writing configure.ac for my portable unix program? Why do I have to spend so much time making an iron-clad robust build system for my portable software? Why do I have to spend so much time configuring my computer before it's safe to be put on a network? All of these kinds of things should be able to be automated for me -- it's the software that needs to be able to handle these things.

So that's what I see my job as: to make software that sucks less. It's challenging and exciting -- to be able to give someone power to do things that they have never before been able to do. To actually be able to increase productivity of others just by providing competent tools; that's neat stuff.

----

Ok, I'm a geek. But I've always admitted it. Hell, I love being a geek. But even more than that -- I don't just thoroughly enjoy it being a geek, I revel in it. :-)

August 16, 2001

Everything's kinda backed up in le-kitchen, Dave

I still didn't get to take my new laptop home from IU; paperwork is still progressing. :-(


Some quickies:

  • I love my DSL. I streamed some MP3s to my new laptop speakers in Bloomies from my home server.

  • I love the high speed ramp from I64 to I264 in Looieville. I don't even have to slow down from 75mph going from one highway to the next.

  • Tucson plows on. I've been doing some interesting development with it over the past 2 weeks. More details pending.

  • Here's an odd fact that surprised me: I actually have 13 Depeche Mode CDs (although 2 of those 13 are the second CD in a 2 disc set, and 1 is a single CD). This makes DM the artist that I have the most CDs of, by far. The next closest that I have is Peter Gabriel with 6, and Sarah McLachlan, Tori Amos, and U2 with 5 (I think U2's got a new one out... hmmm...). Why the heck do I have so much SM and TA? However, I have a whopping 47 "various" compilation and soundtrack records; I think that that is what I have been spending the majority of my CD money in recent times.

  • I hear through the grapevine that Louii is getting married; the rumor is that she is marrying someone she met off the internet.

  • I converted my coderedwarn.pl script to send the vigilante script. How funny is that? I disabled it after a little while, just because it seems like a risky thing to do. But it's funny as hell...

  • I discovered by accident that pine has a threading mechanism. Who knew? Certainly not me. It's not quite nice (although perhaps this is just the nature of threaded indexes) in that a new message may appear anywhere in your index, possibly even in a previous screen. So you might not see new messages if they get threaded above the current screenfull of messages in your index. Additionally, there's no indenting (I think I've seen mutt and wanderlust do indenting), so other than looking at the messages above/below the current message, there's no good way to know where in the thread the message is.

  • Much consultation on the organization and design of OSCAR 2.0. We finally decided to have binaries not reside in CVS, but reside somewhere via HTTP instead (OSCAR's sourceforge page, for the moment). Stealing an idea from one of the other OSCAR developers, I setup an automake configuration that will do a wget for any files that you are missing after you do a cvs update. Nifty. So we still have version control -- sort of. The Makefile.am's are version controlled under CVS, and they list the binary files that were associated with each version. Not perfect, but it's better than nothing. And it's better than putting binary files in CVS.

August 18, 2001

Jeff's Journal

I broke off the cover to my palm m100 a few days ago.

Doh!

So I ordered a PlamGlove cover. Should be here in a few days.



I still get oodles of phone calls for a Mr. Rogers. I think he must have owned the phone number before us. Very annoying.

If you ever see him, please tell him to stop having people call me.


They tell me that I'll be able to take home my new laptop and monitor from IU this week. Woo hoo!!

This is a good birthday present for me. :-)


To end further controversy in the US and abroad, I have decided to donate some of my stem cells for scientific research. All those scientists are saying that they need infant stem cells; people are constantly telling me that I have the mentality of an infant. Hence, I've must have what those scientists need.


Brian has none of his MP3s down at IU yet, so he's streaming from my DSL. Too funny! I love my DSL. :-)


I found and old book of "Lord of the Rings" recently. I couldn't find any copies of any of the other books, and I didn't really want to read that old book because it's kinda fragile. So I bought the entire Hobbit/Lord of the Rings series yesterday, and started reading The Hobbit. I remember the basic story, but it's been years since I've read these books.

I only realized after the fact that the movies are coming out in the not-too-distant future.


It turns out that my long-held suspicions were correct -- the dual NIC/IP forwarding setup on my former router linux box was killing the network periodically. Periodically, when I flooded too much traffic through the machine, it would just freeze up and be completely dead. This usually happened when, for example, I would buy a new CD, rip it into MP3s, and scp them to my server.

If I scp'ed 2-3 songs at a time, with a spacing of at least 15-30 minutes between each scp, I was fine. Anything more than that would risk a freezeup.

Well, now my server is no longer my router (got my cool linksys router for that now), my server only has one NIC setup and functioning. I just bought 3 CDs yesterday and ripped them to MP3s. I tried an experiment and scp'ed them to my server all at once --
worked without a problem. So I don't know exactly what it was in the dual NIC/IP forwarding that cause it to barf (heck, it has a pretty old kernel -- 2.2.something), so it's quite possible that whatever the problem is has been fixed/replaced in the 2.4 series (they re-wrote much of that stuff in 2.4.x).

Here's the CDs that I bought:

  • The Crystal Method, The Crystal Method: Pretty cool stuff. I had several Crystal Method songs on various mix CDs that I already owned, and bought this CD on the fact that I liked those songs. Their songs are a pretty good, not total spazoid, but have good energy and techno beat. If I had to give it a genre name, I'd have to call it "mellow techno". Cool stuff. It's good coding music.

  • Fatboy Slim, A Break from the Norm: A little disappointing. It's samples from older songs, and generally slower stuff. Not really what I was expecting, but I guess it's still ok. It won't make it onto my high-rotation list, but I'll probably still listen to it now and again.

  • U2: All that You Can't Leave Behind: U2 is just getting more and more pop-ish. It's ok, but it won't get played much.


A friend of mine recently unexpectedly became unemployed. Please keep this person in your thoughts and prayers.

August 25, 2001

Chock Full of Notes

Some longer ones, some shorter ones, and some ones in the middle.


I saw Planet of the Apes tonight. Good flick, a little lacking in character development. Only thing in common with the original was the fact that it was a planet rules by apes and gorillas; most of the story was different. They didn't even make much of the class difference between the warrior class (gorillas) and the ruling class (apes); it was a much stronger theme in the original movies. I give it 10 minutes.

I also saw a preview of The Lord of the Rings. It's a three year project! They're doing all three books -- Lord of the Rings, The Two Towers, and the Return of the King. LOR is scheduled to be out this December, TTT is supposed to be out December of 2002, and RotK is supposed to be out December of 2003.

Random note: the subtitle for The Hobbit is There and Back Again.


I got my PalmGlove today. I'm a little disappointed in that it has a snap cover instead of a velcro cover. It's also not quite as sleek as I thought it would be; I probably makes the pilot about twice as thick. But it's padded, and it should protect the pilot (which, as has been pointed out to me by several people, is something that I need). We'll see how it goes.


I had a busy 2 days at IU earlier this week. But I was able to take laptop home. Woo hoo!

On Monday, I had dinner w/ everyone, including Katie and Todd's friend, at a local brew pub. It was good company and conversation --
much fun was had by all.

We're setting up milliways to be our main server at IU. We came to good agreements w/ Rob in the IU/CS admin department about what they will do for us and what we will do for ourselves. It's very helpful to be able to speak sysadmin; we should be able to get along well with those guys.

We have to have a good portion of our web pages (read: everything except lam-mpi.org) down on milliways before the IPCRES rollout ceremony next week. Will likely have IMAP, SMTP, CVS, and other things setup by then as well.

I saw a roadmap presentation from Compaq about the alphas while I was there, mostly dealing with the future of alpha as it is intertwined with the itanium. It was NDA, so I can't say much, but it was interesting. We'll see what happens.

I bought an "A" parking permit for IU. Not as expensive as I thought, but it still wasn't cheap. And I'm not guaranteed a parking spot, either. Not like ND (plenty of parking, just perhaps farther away). City campuses are different, I guess.

Brian and Jeremiah labored to bring down, convert, and generally update the web pages from nd.edu to iu.edu. This was especially painful because some of the web pages are stored in the CVSROOTs of actual projects, not directly in the CVSROOT of the web pages. So we ended up bringing the entire CVSROOT down from nd.edu as well (which is probably a good thing).

I'm going to spend more time today fighting mailman and mhonarc setups on milliways. We're going to CVS our local setup so that things should be a bit more centralized this time.


Here's some quickies:

  • TNT is running a witchblade marathon tomorrow -- all 12 episodes, back to back. I missed the first 15 minutes of the first episode or so; I'll probably catch them then.

  • I committed a patch to xiph.org today. It's my first in a long time. It allows us Sun users to compile from CVS without being forced to use gcc/gmake. It was an automake thing, actually.

  • I'm working on the OSCAR 1.1 press release. It's like herding cats.

  • The power management settings still aren't right on my laptop; it doesn't quite suspend / restore properly yet. I finally found the script where all the magic happens; I may dwiddle around with it and figure it out over time.

  • I finally got and submitted paperwork to get paid for my Army time in December of 1999.

  • My dad noted that on my birthday a few days ago, this is a brief span of time where he is exactly (well, close enough) twice as old as I am: 30 and 60.

  • Ookiness with "#! /usr/bin/env python" and "#! /usr/bin/python" on RH 7.1 (i.e., milliways). It turns out that the former seg faults sometimes (took quite a while to figure that out). WTF?!?!?

  • JeffJournal has moved to lists.squyres.com in preparation for moving everything out of nd.edu.
  • Perk mailed asking about setting up his own DSL CAN. I think I sent him more information than he expected. :-)

  • Saw the DVD version of Apollo 13 last night w/ Tracy (she gave it to me for my birthday). It's a great "engineers rock!" film.

  • squyres.com continues to get hit by Code Red attempts. <sigh>

  • Don found a neat feature of OpenSSH that's quite helpful for squyres.com's setup. In the "Host" section of $HOME/.ssh/config you can not only list a different username, but a non-standard port as well. This even applies to scp; how cool is that? So I can add the following to my config file:

     Host MY.SERVER.NAME
    Port 2222

  • gcc 3.0.1 came out this week. I can get it to build and install, but anything I compile with g++ seems to complain about missing symbols, and I can't quite figure out the problem. Oh well.

August 29, 2001

It's like a pastel black

Janna came over for dinner last weekend; that was fun.


Annoying Mozilla bug: accidentally added a server to the "block images from this server" list (that option is right below "view this image" on the right mouse click popup). Oops... didn't meant to do that. Let's go remove it. Took a little looking around to find the menu to remove servers from the "block images from this server" list, but I finally found it. I found the server in the list, selected it, and clicked on "Remove server". The server disappeared from the list, so I "Ok"ed out of the window and reloaded the web page. The image was still gone. Hmm. Go back to the "block images from this server" window -- yep, the server was still listed there. Repeat the whole procedure. Nope, the server is still on the list. Quit Mozilla and repeat the whole procedure. Nope, the server is still on the list. This is clearly a bug.

Luckily, this list is in a text file that I could go edit to remove the server from the list. But that was annoying. It's also fairly inconsistent with its behavior of various MIME types. You enter in an action and it doesn't seem to stick. I'm not gonna go into detail, but suffice it to say that it's fairly annoying. I find myself gravitating back to Netscape 4.77.


Found a bug in LAM's configure script that showed up when using autoconf 2.52 and the --without-romio switch. Suffice it to say that it was a combination of autoconf being "too smart" and us having an early test in the script inside of an "if" block. autoconf 2.52 moved up a bunch of setup kinds of tests to be inside the if block. So if you did --without-romio, all these setup tests would get skipped, and Much Badness would occur from there.

Took a while to figure that out, though...


The Pervasive Technology Labs rollout ceremony this past Tuesday seemed to go well. Lots of important people were there, including press, etc. I met and chatted with the folks from the other two labs to find out what they were about. Sounds like they are doing some interesting things; we'll probably be collaborating with them in the future. Should be interesting and fun.

We spent a good deal of time the week before moving a lot of stuff down from nd.edu to iu.edu. Our new home is http://www.osl.iu.edu -- the Open Systems Lab. It was actually fairly complicated and took a fair amount of coordination. Almost everything has been moved, with two notable exceptions:


  1. It seems that python is unhappy on milliways (our main server). Rob thinks it may have something to do with the fact that milliways is a linux SMP box. The following program sometimes core dumps:

     #! /usr/bin/env python
    print 'Hello, world'

    It even [sometimes] core dumps if you use /usr/bin/python instead of the env stuff. Weird.


  2. lam-mpi.org hasn't been moved down yet. Since that's not obviously in nd.edu, we let that one stay on ND servers for the time being, and concentrated on moving everything else before the rollout ceremony this past Tuesday.


During the last two weeks, I've gotten really stalled in my drive across Louisville to get to Bloomies. Perhaps it's because school has started again...? For example, this past week, it took me about 45 minutes to cross Louisville, where it usually only takes about 10. It's all highway, but the traffic has been slowed down to stop-and-go. Might have to leave a little earlier next time to see if that helps any.

George from UITS suggested an alternate route from Bloomies to I64. It takes about the same time as taking IN 46 to I64 (about an hour), but exits on I64 about 20 miles south of where IN 46 meets I64. So it's a net savings in distance. Yay George!


Back to work...

About August 2001

This page contains all entries posted to JeffJournal in August 2001. They are listed from oldest to newest.

July 2001 is the previous archive.

September 2001 is the next archive.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.34