Archives

August 2000

XXVI
Nope...this isn't about an old Super Bowl...but this day back in MCMLXXIV partner in crime Rat Bastard was born in a litter of one.  Happy Birthday Dude!  Team Lamb Chop reporters found the Bastard himself at the bar stumbling around, and snapped a picture...unfortunately, it looks like he has already hit the sauce pretty hard.   Here's to at least another XXVI more! 

Beta II.LXXIV is Out
Yup it is out.  There is only a couple of fixes in this one.  The first was to fix a memory leak in the previous beta, and also fixed the client from finishing at about 84% on some work units.  They also released a couple of other clients, addition to the WinXCVIII, Mac, and Solaris betas.  There are two SCO versions out and also a linux version (setiathome-2.74.i386-pc-linux-gnu-gnulibc2.1.tar).  I will keep ya notified if I hear anything interesting (I probably wont install it on my machine though).  If everything goes well with this beta....expect to see the final version 3.0 client in about 2 weeks. 

Stats and more Stats
Now I have ye old cable modem up and running pretty well, I am not limited to the 24.4K connection that I used to have for uploading the stats stuff.  Last night, I added a bunch 'o stuff on the stats pages.  I will give you a rundown on the additions I made last night.

• On the stats main page, I have included some team/member stuff in numeric form.  This includes, the time for today's and yesterday's updates.  The team work unit totals for yesterday and today.  Total members for yesterday and today.  # of new members joined today, and # of members lost today.  Also included the total work units added to the team today (net), and also the work unit production today.  When I say "production", I mean actual WUs posted from within the team, and this doesn't include WUs lost from members leaving or WUs brought in from new members.  Finally I added a breakdown of how many members fall into a bunch 'o WU categories...unfortunately Team Lamb Chop has over 190 members without a work unit processed :(
• On the member charts pages, I have added several things, I now have included *ALL* of the team members in the stats, including the members who have posted 0 work units.  This now means that there are now charts for members 2701-2800, and also 2801-2900.  Also I have included the projected rankings for team member positions 201-400!  Go check it out man!
• On the member graphs pages, I have added two graphs including graphs for members 301-400.
• On the team daily stats pages...there are a couple of changes there also.  The top member WU/day now covers the top 50 instead of the top 30....I have also included a chart showing the sorted work unit average ranking for the top 50 there also.  On the bottom of that chart is a bar graph showing how the top 50 rank relative to each other in their averages.
• In the next couple of days I will try to get some of the overall team stats into some type of chart form or in some better graphical form.....who knows when that will happen for sure though!

New Version of SetiSpy
Released in anticipation of the S@H version 3.0 client release Roleof has released a new version of SetiSpy.  This version will include more information about the work units themselves, and also track, chart and log the pulses and triplets in addition to the gaussians in each work unit.  This version also fixes a couple of bugs, and provides more user friendliness.  Go download it NOW!     -zAmboni

Fixed.
Welp i guess the problem that they were having is fixed now.  Granted I have no idea what was wrong and how they fixed it, but it appears that the work unit credit for those who had the error message is gone forever.  The servers were down for maybe 15-30 min this afternoon, and when they were back up the problem experienced earlier was fixed.  Several team members who posted some results at this time, had the error, and as of this time the credit for the work units were not posted.  The only thing I have seen about this was posted on sci.astro.seti earlier this afternoon:

For a few hours this morning our server stopped accepting results since it didn't recognize legitimate SETI@home clients as, well, legitimate clients. This has been fixed. Sorry about all the confusion and credits M.I.A..
- Matt - SETI@home

Warning!
There are reports coming in all around that there may be a problem with uploading results to the Berkeley servers.  These problems are manifesting themselves in the form of "WARNING: Platform XXXXXXX not found; continuing in test mode".  Where XXXXXXX is the client being used.  There are *alot* of people who are experiencing this problem right now, and there is no word from the Berkeley guys what is up.  When this error message is shown the work units do not appear to be credited to that person's account...and it is happening across different platforms and clients.  I can't say what the best thing to do is....I am not sure if those work units results are actually being kept track of over there, or if they eventually will be credited to accounts.  I just turned off the auto connect for my SETI Queue and will hold out until it looks like things are straightened out.

This is just an idea....but I think they are working on getting things prepared for the version 3.0 client release and are getting things checked out to prevent older clients from sending in results.  The problem is...the version 3.0 client isn't released yet and this shouldn't be happening....but that is only my "theory".   Will keep you informed on what is going on over there.   -zAmboni

Planned Server Outage
Sometime in the middle of the result upload problem this afternoon there was a planned outage posted on the technical news page:

There will be a planned 2 hour outage on 8/30/00, starting at 13:00 UT. While the main reason for this outage is unrelated to SETI@home (Space Sciences Lab is upgrading the UPS configuration on all network gear), we will be using the time for hardware and software upgrades of our own.

Prepare.  And if you aren't using a queuing program already...you should start!

Dressed to a T
For all of you Ars Technica aficionados out there now you can purchase some Ars lovin' to wear around.  There are now available an Ars Technica t-shirt that you can guy from Copyleft.  I have already ordered mine, go get yours now!   -zAmboni

Respect My Authoritah!
The SouthPark Crew is back and with avengence...they cranked out 300+ work units for the day and looks to give Admins spend CPUs and Knight a run for their money in that category.  It has been a while, but they are welcome to continue crunching work units for TLC!   :) .  For those of you who may be new to the team, yes it is that South Park.  SPAC uses the machines that they render the show with for crunching SETI when they are not using them for the show.  They have been on the team since I have been around (over a year) and they used to be in the team #1 spot for a while.  Sometime in December, they had a crunch to do some shows, and their work unit production dropped off to nil.  They haven't been heard of SETI wise until yesterday when they posted their first work units in about half a year!  Lets hope that they will continue crunching and help us kick some Compaq butt.  If you want to learn more about what the SPAC actually do in the whole South Park experience, check out an interview with the SPAC crew that Ars Technica did sometime last year.  Hrmmmmmmm....dunno if we actually should be using SETI@Home to crunch work units to find signs of alien life.  Maybe we should just go ask Cartman.  Many of you have already known that Cartman has had some previous contact with the aliens...anal probe and all :).  Heck....I may just have to email JJ, and see if I can get him to post something on the SETI/RC5 Forum!  -zAmboni

Welcome Back Old Friend!
You may have noticed something in the top WUs for the day that hasn't been seen for several months.  The SouthPark Admin Crew posted a 30 work units today.  Looks like Kyle, Cartman and Kenny have given their computers a break for a while and they are now getting some SETI lovin'.  Welcome back guys, hope ya stay a while :).  BTW....how about some team p1mp1ng in the next episode?????  :^).

*Still* Not Ready for Prime Time?
While South Park is entrenched in the prime time viewership, looks like the version 3.0 of the SETI client still needs some working.  I gave it a shot last night, and things seemed well...albeit a bit slower than what I wanted.  There was one problem I sent off in a bug report....the two work units I did last night finished at ~84%.  not a big bug...but still a bug.  And then I ran across this on alt.sci.seti today:

Thanks for the initial beta bug reports.  Just this morning we found a memory leak in 2.73 (whoops), so we should be releasing 2.74 today or tomorrow to fix this.  If you haven't already downloaded 2.73, you may want to wait.  Thanks for your effort in beta testing!
-Eric Heien

So look for another new beta out sometime tomorrow/tonight.  I think they want to really get this thing out the door.  They are releasing a new beta even before getting all the bug reports back.  Most likely they want to make sure the client is running fine and doing all the important stuff right and get it out....then maybe fix the "cosmetic" problems (like the % done thingy") with other released after the "official" v 3.0 rollout.   -zAmboni

Ya Ya Ya Its Here
The new version 3.0 beta dubbed v2.73 has hit the SETI site.  The email that they sent didn't really say anything too much about the revision or what has changed since v2.72.  Here is what they sent (well most of it):

The SETI@home beta client version 2.73 fixes several of the remaining major bugs that the beta testers have reported.

NOTE ON TESTING:
Please do your best to thoroughly beta test this version.  Try everything you can think of to make it crash, and install it on as many computers as you can.  Before we release version 3.0, we want to make sure there are no major problems.  Smaller issues will be dealt with in later versions (3.0x).  Any bugs, small or large, should be reported to beta@setiathome.berkeley.edu, even if you have already reported them for previous versions.

In particular, we have been receiving some reports of excessive CPU heat, resource usage, and sluggishness.  If you notice any of these, please let us know your CPU temperature, swap file size, available resources/memory, and other system information, both for version 2.04 and 2.73.

I cant really say I know anything about the CPU heat or resource usages on my computer with previous betas.  I personally think that the CPU heat "problem" is sort of natural but may have gone unnoticed by many because they weren't really "testing" the client before.  If you peg your CPU at 100% with version 2.04 and if you peg it at 100% with v2.73 there really shouldn't have been any differences with CPU temps (right?).  There did seem to be some sluggishness with the previous betas, but that was mainly in movement of the client window, and opening and closing the client from the system tray.  I'm trying out v2.73 right now, and cant really say if it is faster/slower/same as the previous betas.

Lookie Here Lookie Here!
There was some movement in the overall teams yesterday (If you didn't notice).  Team MacAddict passed up the flailing Microsoft for the overall #7 spot.  Prepare to hear MacZealots everywhere gloating how MacOS is superior to Windoze just because their SETI team is ranked higher.  You can always fight back saying that at least Win Me is going to hit the shelves way before MacOS X hits the shelves...  -zAmboni

Status Updates
Still no news on the 2.72 beta, perhaps it will be out tomorrow. The rest of this paragraph is a filler, designed to prevent the front page from looking empty and forlorn.

On the bright side, things are looking better for the benchmark page. Knight has been working with a friend on getting a database setup, which is now done. So administration of the results is going to be much easier than it was in the past, which will make life easier for our resident celebrity Max. The only thing left to be done is to pipe the results to a page. I won't lie and claim to know anything about the process in detail, I'll leave that in the hands of the experts. Oh, and Knight too. -Rat

New Beta Tomorrow?
Just picked this up from the SETI newsgroups:

I should be sending beta 2.72 out today or tomorrow, and if all goes well with that version, 3.0 should be out a week or two after that.
-Eric Heien

Its about 10pm ET right now, and I don't have the email yet...maybe tomorrow then.

Weekly Update is Up!
Actually it was up sometime last night, but I didn't let anyone know it was up :)   But if you want to check it out you can check it out here.  The big mover of the week was octane.  He was tearing up the 250-400 member range moving up a whopping 132 spots in the past week.  He will find the path a bit more treacherous trying to move up some more...it gets real tough to move up within the top 250!  You can tell that from the rest of the top movers for the week.  One reason the movers numbers are depressed for the past week, was because of at least 12 members who joined Team Lamb Chop in the top 300.

The top producers of the week look all too familiar.  Admins spend CPUs was tops for the week, but slowly creeping up on him is Knight.  The second place (for the week) Knight has been slowly increasing his output per day and will soon be challenging AsC for the top producer! Memnoch and svdsinner are battling it out for 3rd and 4th....with Memnoch taking 3rd for the week. 

In the top 50 there has been some significant team placement moves going on in the past week or so.  Topping off the list is svdsinner.  He has moved passed Tobler Design into the #3 spot.  Memnoch is about to put some moves on Xanrel...he is closing in on the #5 spot and wants it soon...probably tomorrow!   Moving down a bit, Knight is kicking butt and taking names.  He has moved up over 4 spots in the past week and has planted himself firmly into the #8 position.  The problem is there is quite a gap between the #8 and #7 spots.  It may take him a bit of time to reach #7!   Both Beryllium and VulTure's Den moved up a spot each to #9 and #11....while aenima is hoping to move past OoklaTheMok for the #13 spot tomorrow.  romkey and NSteussy are battling it out for the #16 spot...but romkey may win the war...looking to move up one to #16 in the next couple of days.  rosborn put the moves on wesley barnes and moved up to #19 in the past week.

Moving down a bit more....there are some pretty heated races going on.  Greens in Oz with a flurry of work units today, and on the 17th has passed up Action for #22.  Joker keeps cranking 'em out and moving up, up one spot to #25.  Dan Dan the Linux Man passed R. Marvin for #29...and now R. Marvin and Daryle A. Tilroe are neck and neck for the #30 spot...right now R. Marvin holds the edge.  JerkyChew clawed up one spot to #33.  Mr. Catbert, moved swiftly past both XYZZY and ojohnson up to #36....and Geordie_Cos380 moved past ojohnson also to #38.  Whiner has been duking it out with doma's SETICluster, and came out on top at #43.  wiz_of ez was up two spots and into #47....and finally MadMac passed both Panders and The Phrog or the #48 spot.   Way to go all!

Server Downtime Explained
The servers were down for about a half an hour this afternoon.  There was no previous warning for the downtime, but this popped up in the technical news reports *after* the outage was over:

August 21, 2000
We are having a half hour outage today to measure the database speed and backup unprocessed results, as we fell significantly behind over the weekend in regards to inserting results in the database.

well it would be nice if they posted this *before* the outage so ppl could be prepared and not wonder what was going on....

SETI Driver WARNING!!!!!
This is taken directly from a post on the SETI newsgroups from Michael Ober (creator of SETI Driver):  He warns you NOT to use the latest version of SD:

To all using SETI Driver,

The New Client button has a bug in it.  This bug can possibly cause SETI Driver to delete the S@H client.  I will have it fixed before the next
command line client version is released.  In the meantime, DON'T use it. The purpose of this button is to allow a clean upgrade of your S@H client,
so it isn't really needed until CLI 3.0 is released.  The New Client button allows you to

Actions when clicking the New Client button.
1.  A file open dialog box is created asking you to select a new SETI*.exe client.  This client will become your new S@H default client.
2.  Copy the current S@H client from the SETI Driver directory to each of the WU directories, renaming it "SetiAtHomeOldClient.exe" in the process. Windows will fail this copy if there is already a file named SetiAtHomeOldClient.exe in the WU directory, so if multiple new clients are released while your cache is clearing, you are still protected from restarting a WU when changing clients.
3.  Stop all WU processing - Windows won't allow the next step otherwise.
4.  Delete the S@H client from the SETI Driver directory.
5.  Copy the selected client (step 1) into the SETI Driver directory.
6.  Restart WU processing.

WU processing checks for the file "SetiAtHomeOldClient.exe" in the WU directory and uses it if it there.  If it isn't, the S@H client in the SETI Driver directory is used.  All WUs are processed with the -stop_after_process switch.

Transmit processing works as follows:
On each WU that is ready to transmit do the following
1.  Check for SetiAtHomeOldClient.exe and use it if it is found, otherwise, use the S@H client in the SETI Driver directory.
2.  Create a SETI nn - Transmitting processing with the -stop_after_xmit and any proxy/socks switches on the command line.
3.  When the transmit finishes, delete SetiAtHomeOldClient.exe.  This is the only time it is safe to change S@H clients without loosing WU processing time.

=========  BUG  BUG  BUG  =============
This is where the bug lies - I didn't properly detect this file and will actually attempt to delete the default client instead if there is no SetiAtHomeOldClient.exe file in the WU directory.  This bug can only manifest itself if there are no WUs actually being processed at this point in time.  Windows won't allow you to delete a file that is open, and Windows keeps all executable files open as long as the program they contain is running, which is why I didn't catch this in testing - I always have a WU processing, even when I'm developing.
============ END BUG ================

Version 1.5.0.0 fixes this bug, but isn't ready yet.  I am having to completely rewrite the transmit control code so I can add full SETILog support to it.

Where Are They Now?
Nostalgia time....lets turn back the clock to the day I started keeping stats for Team Lamb Chop.  Originally I started to only keep stats for the club teams.  Then about a month later I decided what the heck...lets keep stats of the top 100 members of the team.  At that time I was only limited to the top 100 because that is all the stats the Berkeley guys gave us.  These are the first numbers I had taken for the team and will show you where they are now.

Heck....While I'm at it lets take a look at the same date in the Club Team scene. At the time I took the data here, I only kept track of the club teams.  Heck the other teams were out of our league (so I thought!)   * note:  the data for the club teams were taken from the club team page on the SETI site and do not necessarily reflect the current work unit numbers for each team!

When compiling the numbers for the above tables....I ran across something quite interesting.

Almost one year to the day that Art Bell passed Team Lamb Chop in the standings we passed Art Bell back again (to stay this time!).  Hey we did it two days short of one year....but who is counting the days? :).  Hopefully one year from now we will be past the top three...something I never imagined.  Lets make EVERY team look up at the Lamb!  -zAmboni

All You Need...
If you don't have the proper tools to do a job, they can easily be replaced by 1) a hammer, 2) chisel, 3) vice grips.  Earlier tonight i needed to take the strut off my VW GTI...but of course I needed a "special tool".  Of course you will never be able to find any "special tool" at any auto shop or even a local garage.  My normal repairman just uses a impact wrench to get a strut off (I even watched him).  Problem was my GTI was 10 miles away in my garage at home...with no impact wrench.  A normal socket wrench didn't work....socket wrench + vice grips was a no go also.  Not wanting to get a tow to the shop to get it repaired "properly", I decided to manufacture my own "impact" wrench with a hammer and chisel.  Vice grips were used to hold the perfectly round and shiny strut shaft (ok they semi gripped them).....then use the hammer and chisel on the top strut nut.  Many "impacts" later, give it a shot with the socket wrench and "voila!", the nut is loose!  Strut is replaced, game over man.  This now ends the improper lesson in car repair, back to your normal programming.

Duplicate Result Problem Solved?
Eric Heien reported on the SETI newsgroups that they (hopefully) have fixed the problem with users getting the dreaded duplicate result message:

Hi everybody,
We think we solved the problem with the server reporting duplicate result errors.  We believe that the table we use to look these up had just got too big.  Coupled with a backup we're currently doing, this caused some lookups to take too long, resulting in an error being generated which the server interpreted as "Duplicate Result".

In other words, you shouldn't see this problem anymore.  If it's still occurring, please let us know. 

This is actually another good reason to get a work unit caching program such as SetiQueue.  Reason why is that when you use SetiQueue, and a result is dropped for being a "duplicate" work unit, SetiQueue places the result in a "Dropped" folder. Sometime later, when problems are over, you can just copy the dropped results back into the SetiQueue folder and reconnect, and hope the work units will be uploaded this time.  This feature, and keeping things going during planned and unplanned outages are reasons why you *should* be using some type of work unit caching program.

In addition...
to the other Team Lamb Chop discussion outlets, there was one that I totally forgot to mention.  There is a Team Lamb Chop email list.  If you have trouble connecting to IRC, and/or lack web access at work, this may be a way for you to join in on the fun.  You can sign up for the email list by checking out the account maintenance site.  Props go out to Kevin Carpenter for maintaining the list!

...and Finally
Another thing I forgot to mention, was that the weekly stats were updated a couple of days ago.  Sorry for not posting some more team stats updates on the front page here, but it has been quite a busy week on my end.    -zAmboni

Funny Strange...
C. Eric Smith took a break from his ultra top secret project to drop me a line.  CES and Jason Johnston has been a great help in keeping the TLC server up and providing free web hosting for the site.  He sent along a link for fromago, a site that they are hosting and wanted to pass it along to the masses.  Its a mix of fish, humor and internet dating.  Definitely an interesting read!  

While you are at it go over and check out the Nimbus Networks site.  If you are looking for a web hosting site drop on by.  They have been quite reliable, and the servers are fast.  If there is any problems Jason is always on call to replace that toasted motherboard at 4 o'clock in the morning!

Funny Ha Ha!
Team Lamb Chop unveils the "Official Humor Site" of Team Lamb Chop.  Where else could you find Retarded Monkeys, Exploding Breasteseses, Heatsinks for the Homeless, and the Unfrozen Caveman CEO?  None other than the (in)famous BBspot.com.  The site is a mix of wit, irreverence, clever writing and imaginative ideas.  BBspot actually beat all the Apple rumor sites to the punch by bringing news of the new line of Apple Computers that include the dMac, vMac and biMac.  They also don't pull any punches by asking the tough questions such as What (OS) would Jesus Run?  Stop by, check them out, you will not be disappointed!  -zAmboni

"I Extend to You, A Laurel and Hardy Handshake"
Sorry about the misplaced Blazing Saddles reference, but I want to welcome all of the new members to Team Lamb Chop over the past couple of days. Lets see how quickly we can kick some serious Compaq, Sun and SGI butt!  Also, I would like to thank the entire Team Lamb Chop membership for helping us whoop Art Bell.  Lets keep it going and the top 3 better watch out...we just may have to open a can of Whoopass on 'em!

For all new members, and any members of Team Lamb Chop who haven't dropped in...there are two outlets to help discuss SETI and meet other team members.  The first is to drop in on the SETI/RC5/Distributed Forum at the Ars Technica Open Forum.  The second (and third) outlet is to hop on IRC and join #Seti on either kulish.com (port 6667) or irc5.shacknet.nu (6667).  There are usually several team members who hang out to talk SETI, or anything else you have on your mind.  Right now the two servers aren't connected to each other, but hopefully they will be soon.  See ya there!

Server Problems At the Home Offices
Some of you may have had problems connecting to the SETI servers today...There were a couple of "unplanned" (read: someone f*cked up) power outages over there on the Berkeley Campus.  There was a post on the SETI site stating the following:

There were two unexpected back-to-back power outages this morning at around 10:00 PST. We have temporarily switched to backup power, but there may be another outage later today or tomorrow when we switch back to the regular power grid. UPDATE: This mornings power outage happened due to a switching problem at Lawrence Berkeley National Labs, on the same hill as the Space Sciences Lab. The situation has been corrected and we should not experience additional outages.

Things didn't end there though.  The servers went down again for a bit to shut down the servers from the UPS and bring them back to the regular power grid....Eric Heien posted on the SETI newsgroup:

Just thought I'd let you know that we are having an unexpected outage right now.  I'm not able to post on the web server, because it has gone down and we're waiting for the "OK" to bring it back up.

Basically there were two back-to-back power failures around 10:10 this morning, we're not sure what caused them.  The power is still off in parts of the building, so we're waiting until it's all functional before bringing the servers back up.  As we have UPS backup, no data was lost and the machines were able to gracefully shutdown on backup power.

More Server Problems????
I am not sure what is causing this, but there has been several people report this both on the team forum, and on the SETI newsgroups.  Some people are getting their results rejected by the servers because of "duplicate results".  The problem is that these work units were not done by the users before, but the servers said they were, rejected the results, and didn't credit the work units done for them.  Personally I had one out of 11 work units uploaded today that said it was a duplicate result.  And it wasn't a duplicate.  One poster on the newsgroup said he used SetiQueue (which saves a copy of uploaded results) and he resent the offending results later, and they were accepted by the servers.  I haven't tried this myself, but may later on.  I don't know what is up, but I will relay any info I hear along to y'all  -zAmboni

-Front Page