Archives

January 2001

Now And Then: A TLC Graphical Retrospective

NT/2K SETI Service/Scheduling

Seti Driver Guide Updated 11/19/00

S@H Response to Hacked Client
There was a post on alt.sci.seti tonight from Eric Korpela on the aftermath of the hacked client being revealed (this is a semi-official response):

Sorry about letting the lack of response from most of the SETI@home team posters.  I'm sure you can understand that we've been a bit busy.  Thank you, Lawrence Kirby, for helping everyone keep a little perspective.

Yes, the patch does exist, and yes, it does what is claimed.  It requests a workunit and uploads a blank result a few seconds later. Unfortunately we had become complacent against this type of attack following the release of the version 2 clients.  We are no longer complacent, and are watching carefully for users using this patch.  Most make themselves pretty obvious.

The damage to the completeness of the science database appears to be minimal.  Lawrence is correct that it is unlikely that the same workunit would be sent to two different cheaters.  There is also an additional effect that Lawrence didn't mention.  Workunits are not deleted immediately after a second result is returned.  The deletion occurs only when the splitters have filled up the available disk space, and then the deletions are in fairly small chunks.  Since we're constrained by the data rate at the telescope (and we don't run the splitters any faster than that), the net effect of a million bogus results returned quickly is that the average number of times a workunit is sent goes up to compensate.   It's not enough to fully prevent any loss, but it helps. 

We're investigating ways to prevent type of hack from working in the future on both the server and client side.  Some of the options people here have mentioned are being invertigated.  We'll try solving the problem in the server first.  Because of the architecture it's a whole lot easier to prevent the phony results from ending up in the database than it is to prevent cheaters from getting credit for the phonies. 

Another option would be to no longer give credit for null results.  We expect that many users would rebel against that, and stop processing any workunits that don't find a signal in the first 5 minutes.

Some of the options presented here would greatly exceed our database processing capacity to implement.  They would certainly work, they're just not feasable in our current setup.  Others would require that we increase our workunit storage by a factor of 8 or so.  If we had that much disk space around it would get used in the science database....

If we're not able to solve the problem in the server, expect a Windows commandline 3.04.  I've come up with a method of prevent the null result problem that would make the patch job more difficult.  Of course there's no way of making the patch job impossible.

At any rate, the damage isn't too bad, and we're working on a solution...

Eric

Its A v3.03 World
Welp...it looks like the axe has fallen.  The Berkeley guys shut down the server that was feeding work units for the pre v3.03 clients.  Here is what Matt had to say about the switch:

There has been a steady increase in the number of 3.03 results we've been getting back, though we're still considerably far from 100%. As far as I'm concerned, there's been enough time and enough warning so I'll suggest at today's meeting that we make the switch like now. We've been slow because we can't turn on warnings to the CLI clients without gettings DOSed (though that's greatly improved over the past week as people finally started making the switch).
- Matt - SETI@home

So for those of you out there who have been waiting for the last minute to switch over to v3.03, that minute has come and passed.  There is one note to this though.  Not all of the clients have been ported over to v3.03.  On the downloads page there are still at least 7 clients that have not been ported over.  I guess if you were using the pre v3.03 clients then at the moment you are SOL  :(.

-Update: hrmmm I couldn't connect with the old SetiQ earlier today, but that must have been the problem listed below.  I just tried (9:00pm EST) and everything went through...it isn't down YET.

Keeping Up With the Joneses
With the switch over to version 3.03, there have been some updates to several of the favorite add ons for the SETI clients. 

First off, there is a "beta" SetiQ program that has released.  This version is now a Windows GUI based application, and will support the version 3.03 client.  There are other changes to the client that you can read about here. 

Secondly, there is the release of version 3.0.4 of SETI Spy.  This version update contains mostly different bug fixes and also some visual inconsistencies.  You can check out what is different here. 

S@H  Was Down...Now Up
There was a problem with the S#H site, and attempts to download work units even with version 3.03 were fruitless.  There seems to have been some configuration problems with their routers/servers that Matt explained here:

Due to a syntax error in an apache .conf file, one disk partition filled up, causing a general slowdown of the web server and the data server last night. This was easily fixed, and plans are being made to reduce server dependencies as much as possible (by moving them on bigger, separate disk partitions).

Addendum: Yes, the error was my fault and I am currently still kicking myself.
- Matt - SETI@home

Little Of This...Little Of That...Mix...Zap...ET?
I ran across this news item a little bit earlier tonight.  There has been earlier attempts to "create life" in the laboratory using elements found in the early universe.  These experiments were able to create some of the complex molecules that many considered to be the precursors of life here on earth.  Scientists now seem to have done the same using a deep-space environment:

 In a simulated deep-space environment, researchers have created small structures that look like the cell walls in living organisms and show early signs of the ability to convert sunlight into chemical energy. The work, announced Monday, adds to growing evidence suggesting that interstellar clouds of gas and dust provide the ultimate in Betty Crocker mixes: Just dump the enclosed ingredients into a hospitable planet, add water and stir.

Yum...ET in a test tube!

What To Say?
The internet is a strange place.  People can act like your best friends, and then turn around and stab you in the back laughing while you are laying there bleeding.  Luckily, TLC and the S@H program has not seen many of these people.  Unfortunately one of those times is here and now.  If you don't understand what I am talking about, you need to reference this thread on the distributed forum.

To summarize:  2-3 people, fake friends, multiple accounts, over 1,000,000 fake results, stab wounds in the backs of TLC, KWSN, and the entire SETI@Home Project.

(Hopefully) the last thing I will say about this situation is the following (taken from a post I made in the above thread):

All I can ask any/all of you are:
1) DO NOT download the hacked client
2) DO NOT use the hacked client even if it is out of curiosity.
3) DO NOT post links to where you can download the hacked client
4) DO NOT join TLC if you are using a hacked client
4) If you are on TLC using a hacked client, please leave the team.

 

Version 3.03 SUCKS!
Ok those aren't my words, but one poster on alt.sci.seti thought so!  But his post did get the attention of Matt Lebofsky and he replied in kind...

> I used to be able to finish a WU in 3.5 hours on my p3700@1008 Mhz.  Now it
> takes about 5 hours.  On my AMD 700@850 it used to take 5 hours, now it
> takes 7+ hours.  This sucks.  Am I able to still use v 3.0 instead of 3.03?
> These are the text clients of course.

3.03 sucks so much because we cleaned up a bunch of sucking code in this version. We found several cases existed in 3.0 where the clients could process entire workunits without ever touching the suck functions. This was an undetected conflict between the science and suck code, and has been fixed in 3.03. Now we are efficiently maximizing both science and sucking.

You are able to use 3.0 during random phases when I don't have mandatory upgrade messages turned on. I turn these on whenever I feel like it, and turn them off when 3.0 clients in tight loops load down our server.

As of time of writing, you cannot download workunits with 3.0, but you may be able to later. Within a week, no matter what, 3.0 will be dead in the water. We can only accept results from suckless clients for so long.

But...for those of you who thinks that Matt's words were a bit harsh he also added in a following post:

> Or is Matt just finally fed off with the continous complaining
> and whining going on from the speed freaks about having to change to
> the slower v3.03.

I'm hardly fed up. I feel everyone has the right to complain. People get many different things out of this project, and have every reason to be upset that our goals may hinder their personal goals. It's a big family all living in one big ol' house.

But I mean, c'mon. I think the original author couldn't have been that freaked out about an extra 1.5 hours per work unit of computing time, and so I feel a little levity in response may put that into perspective for everyone.

At least I hope. I'm one of those people who refuses to use emoticons, and therefore people automatically assume I'm being a raging lunatic or a total jerk. Oh well. When I start to see those things in books it'll be clearly apparent to me that we're all goners.
- Matt - SETI@home

New SETI Driver Out!
Mike Ober has let the cat out of the bag and into cyberspace...the new version of SETI Driver is out.  Here is the specifics on the changes:

It's on the North American download site now. There is no longer a Win95 specific version - I worked around the problem in code.

Changes from the previous version are:
- Auto Transmit loop has been changed to reflect the text client page looping instructions
- SETILog.CSV is now created only when requested
- Fixed a command line reading bug that could cause socks and proxy server settings to be mis-read
- Fixed a resource leak when clearing the "Use SETI Driver" progress monitoring checkbox
- During WU transmit, SETI Driver now boosts the priority of the transmit process to High after 30 seconds. This also means that the shortest time period between the start of consecutive WU transmits is now 30 seconds.

The European download site should be ready in a day or so.

On Again, Off Again
It has been noted in various places that the view the last 10 work units page was on earlier tonight.  But again for some yet unknown reason, they are back off line...  And I thought my Win98 box was unstable...

Error Codes and You
I saw this post on alt.sci.seti from Eric Korpela.  There has been some question on what the different error codes you get from the SETI client were, so Eric helps demystify what the codes are.  If you have been wondering, here they are:

I'm not sure it really matters what the error is, a script could easily treat all of them the same.

Here they are anyway:  Some of them are only used internally.

#define SUCCESS                 0
#define NAME_EXISTS             (-10)
#define FILE_DELETE_FAILED      (-11)
[snip]

woah....that was alittle bit longer than I expected it to be :)  Instead of putting them all in here I made a post to the Distributed Forum with the entire list.  Check them out here!

Pokemon
Ok now I wonder if Nintendo will come after us!

And I Almost got DSL...
If you have DSL and aren't using one from a baby bell, you may soon be SOL.  I *nearly* went with COVAD DSL when it was taking Mediaone ages to upgrade to two way cablemodem here.  Im glad I waited.  And the DOJ thinks that Microsoft is a monopoly?

Back to Business
I didn't want this to turn into a E/N site...but just had an overwhelming urge to post those...

First off, I posted the weekly stats for the past week.  I noticed when I was getting the weekly stuff ready...I am running out of room on my Excel spreadsheet.  I have less than a month's room on the darn thing!  I need to get cracking on some new code and a new worksheet for bigger and better things for the site :)

There has been some mad bovine action going on in the past week.  First off we had the arrival of Evil Cow whipping up the charts, and also Team Beef Roast got some front page Ars pimpage....way to go!

How is TLC doing in the standings?  Yesterday TLC passed through the 3 Million WU mark again, and is again steadily climbing.  Right now we are about 120,000 WUs behind Sun, at the current pace we should be passing Sun within two weeks.....BUT (more about that later).  How is KWSN doing?  After the past week's fallout they ended up sitting behind Compaq down in 5th place.  Right now they are in Compaq's shadow and should pass them with tomorrow's update putting them at the #4 spot. KWSN  is about 600,000 WU behind TLC right now....and even though it is too early to get a good idea of average production, KWSN is cranking out about the same number of work units as TLC, and they probably are producing a slight bit more.  With the current lineups it would take KWSN quite a bit of time to catch up to TLC since the difference in production between the two of us is pretty slight.  But as many of you know from last week, team lineups can change quite fast and furious.  Nothing is set in stone.

More About V3.03
Yes we are set to pass Sun again in about two weeks, and this is where the BUT comes in.  I have a feeling that the servers for version 3.00 and previous clients will be shut down REAL SOON!  The time for passing Sun may be stretched out a bit because of this when version 3.03 becomes *absolutely* mandatory.  Matt Lebofsky said that v3.03 was mandatory a couple of days ago, but I still have been able to upload results/download WUs.  Why is this?  For the past couple of days there seems to be a period of time where people cannot download work units from the servers (but can upload results fine).  A couple of days ago I didn't know the reason why this was happening, but there may be an explanation for the weirdness.  Mike Ober has been in touch with Matt with the changeover and the interactivity of Seti Driver, and he has this to say about the download problem.

Actually this is the whole issue. I have been in contact with Matt L. at S@H about the problem and he has confirmed that the Client Expired messages are being turned on and off. When they are on, the problem occurs. When they are off, this problem goes away.
Mike.

But why is Matt turning things on and off?  I think that there is a simple explanation.  He seems to be turning on and off the client expired messages during the afternoons over here in the US.  I believe he is guessing that these are the times that he thinks that people will be at their machines to be able to see these client expiration messages, in the hopes that they would see this and then switch over the clients.  He may be doing this at different times in the day to make sure he will get most people's attention.  Why doesn't he keep the message up to make sure that people will realize to change things over?  I think that it has to do with many people with "SETI Farms" and others who use caching programs or scripts to keep things going.  If they leave the message on and prevent the downloading of work units, SetiQ, and many of the scripts will try over and over to connect and download work units, basically DOSing the Berkeley servers. 

The way Matt is doing things now, he turns on the messages to try to get peoples attention, and then he lets things to back to normal.  I would expect this to happen probably for the next couple of days.  I think this is the last warning.  The next thing that will happen is the server shut down and you wont be able to connect at all!  My guess on the server shutdown?  I think either Thursday or Friday.  You have been warned....Upgrade!

Version 3.03 Mandatory?  Or Not?  (yet?)
Many people (including me) had some problems with submitting results and downloading new work units a couple of days ago with non 3.03 clients.  It was a bit strange since the servers accepted the results, but didn't download a new work unit.  Sometime later that night, I was able to download new work units.  Strange eh?  I sort of took this as a "shot across the bow".  Most likely they were playing around with their servers to get ready for the client switch over.  I had figured that we would hear something from the S@H guys just before they switched over, so I didn't think that the previous episode would have been the real "switch off".  I just checked the alt.sci.seti newsgroup and I believe we have the "two minute warning".

Hi, SETI@homers,

Short story:

Version 3.03 is now officially "mandatory."

Yes,  all earlier versions are obsolete, including 3.0, even though you may have only installed it as little as a month ago. Sorry for the inconvenience. We have been warning this time would come for months, and have been issuing warning messages to most clients over the past couple of weeks, and within a week or so we will shut off all old clients at the router (via a DNS switch).

Long story:
We've been screwing around with the server for the past month trying to get it to respond politely to all versions on varying platforms. Due to outdated client/server communication protocols, unpredictable behavior of third-party caching software, people running their clients in while(1) loops, and plain old bugs, this has been proving difficult but at least almost completely solvable. However there are some cases, like the windows 3.00 CLI whirring away on their compute farms, where we can't do anything about it except the hard shutoff via the DNS switch.

We never had these problems before, or at least at this scale, because we made very few versions above 1.x mandatory. However, lessons learned and new logic in place should make any future transitions *much* easier for us (and therefore you, the users).

In other words, we do try to make transitions as polite and seamless as possible. Please bear with us. Once we get everyone on an even playing field with version 3.03, there's going to be a huge sigh of relief in SETI@home-land, and we can continue working on more interesting things (for both you and us).

Thanks for your time, interest, and effort!

- Matt - SETI@home

Pre v3.03 clients are dead now...right?  Hrmmmm I'm not so sure.  I guess mainly because I am not sure what their meaning of "mandatory" is.  Right after reading this I went in and uploaded 4 results being run with version 3.00 CLI and SetiQ.  The results were sent ok, and it downloaded 4 new work units just fine....then I looked and it posted the 4 results to my total fine.  That really isn't my definition of "mandatory" :).  To me the only way it will become "mandatory" is when they shut off things via the DNS switch.  OK either that or they stop giving credit for pre version 3.03 clients.  But beware, if you want to stick it out to the bitter end, that bitter end will come "within a week or so".

Results?  When?
For the good part of the past two years S@H users have been sending the Berkeley guys results...but when will we see some results from them?  It looks like the Berkeley guys will be starting to do some serious results sifting.

>But they hope to get up to real time processing speed
>when they add another server (in seti@home II ??).
>It wasn't clear when.

The machine that will be doing this processing was assembled in the last week, and will be fully operating in the next couple weeks.

I should add something to the tech news about it...

Eric

Sad Day for TLC
This has nothing to do with the latest in who is has joined the team or who has left the team in the past couple of days.  This goes a bit deeper for those who have been a part of TLC for a while.  Yesterday Rat Bastard offered his resignation for the post of benchmarking guru.

Hello all,
As you're all aware, I've been rather absent as of late. Life in general has kept me rather busy, and I've found less and less time to devote to web ponderings. Furthermore, with the status of the new Seti clients, I've become disenchanted with the project as a whole, and simply have lost interest. As a result, I have not upgraded to the 3.03 client, and will not. I do plan on running some RC5 on my servers (just so the processors don't rust ), but even that will be no matter of importance to me.
I just wanted to post this here so that everyone knew where I was. I've had fun with the Seti project, and I've met alot of great people here. However, it's time for me to do other things. So I will post my official resignation from all Seti Community happenings. If anyone wants me, I'll be playing games with my son, who currently seems to enjoy the games 'run', 'jump', and 'ball'. 'Drool' is still a favorite too- although it's not as good as 'hug dada'.

Goodbye everyone,

Rat
Eric Nadeau

Many of the newer members may not know that much about Rat.  He has been an integral part of this site here and in Team Lamb Chop in general.  RB was on TLC already when I first joined the team.  Not too long after I started keeping team stats on my personal web site, RB came up with the idea of benchmarking the different clients using a common work unit.  At first he kept the benchmarking stuff on his own site (even though we tried to integrate things as much as we could :), and then things were combined when we moved to this domain. 

The benchmarking pages have been (and still are) a valuable tool for comparing different clients and different systems with regard to SETI@Home.  For many, tweaking for the best S@H production, means tweaking a system for best overall performance!

I have always thought that in the S@H community as a whole, know this site more for the benchmarking pages, than for the team stats pages (even though its the stats pages that keeps em coming back ;) hehehe.  It has been a pleasure working with RB over the past year or so, he has been extremely valuable to TLC, the Ars Forums, and the S@H project.

Good luck with the family, job and your gaming habits!  You will be missed by Ars, TLC, and yes even me.  Hey, he is the father of the {H}ardest Son in the SETI world!  (see Nov 25, 1999 news).  -zAmboni

I Must Also Add...
That Max is going to keep going with the benchmarking duties.  A big tip of the hat goes out to Max!  If anyone would like to help out on the benchmarking pages let either Max or I know and we will see what we can do.  I also need to check with Max on this, but we may want to get a setup for the benchmarking to do some sorting of the different columns, and maybe a better submission form.  Let us know if you can help!

Calling All Lurkers!
I know that there may be quite many a lurker out there who may stumble upon this site, or maybe even come here day after day.  I want to invite you all to drop in on the Distributed Projects Forum.  Check it out and take a look around.  We would like to hear from you even it is just to say hi!  This is not limited to Team Lamb Chop members.  If you are from *any* team, drop in and say hi.  The only thing is that registration is required to post.....but it wont cost ya anything!  Hope to see ya there.

SETI Blackout?
Many of you know that there is a pretty serious power crisis in California right now.  It has gotten to the point where there are rolling blackouts in some California regions.  Will this effect the Berkeley Campus and the S@H servers?  I haven't heard anything about this, but I really doubt if they would cut power on a state university campus, so the S@H project should be safe from this.  But again I could be wrong.

More About v3.03 Transition
Hiram (leader of the client porting efforts) posted on alt.sci.seti a couple of days ago this tidbit:

Good Morning SETI Fans:

This is the best statement I can make at this time. The team is trying to get the pre-v3.03 clients turned off as soon as possible.  As they work through various problems with the transition, the exact moment is unknown.

For your purposes, you should assume your pre-v3.03 client is obsolete and you should upgrade now.  Otherwise, when the moment does happen, your processing will be dead in the water.

--Hiram - WB6RSS

If you notice from the news a couple of days ago they state that any pre-v3.03 clients should be shut off completely "about" Jan 31, 2001.  This doesn't mean that they will shut them down on Jan 31.  The shutdown could come anytime from now until then (or after).  Right now if you are planning to stick with any earlier versions you may be risking it if they shut down the old servers say....tomorrow.  Remember.  If you have any cached work units and they shut the client down....those results are lost.  Forever lost in the dead client result vacuum.

v3.03 Bug Info Updated
A week or so ago there was news about a possible bug in the v3.03 client in which the client will stick state.sah and key.sah files in random folders, and multiple copies at that.  Matt Lebofsky updated us on things tonight, and it is not apparently a bug in the client, but those systems in which this has been seen are infected with a virus.  I'll let him explain.

More and more people are complaining about multiple copies of state.sah and key.sah files appearing in random folders of the windows systems. We've been trying to reproduce and debug this for a while now, and it's looking like it's not a SETI@home bug at all. It's a completely unrelated VIRUS!!!

Not just a virus, but I believe it's the hybris-b virus, which got spread around recently via an e-mail regarding "Snow White" or something or other. In any case, if you are experiencing this problem, get your latest anti-virus updates and clean your system! Some of the users who I've been working with regarding this problem found this virus and haven't had a problem since - I haven't heard back from the other users yet.

What's shocking to me, if this indeed is the cause of the problem, is how many systems have been contaminated!! Wow.

- Matt - SETI@home

Finally...
The team shakeup is not over yet.  One of the team members who left a couple of days ago (Support AIDS Research) was apparently inadvertent,  and they are now back.  SAR now claims the #1 spot, and knocks everyone else in the team down a notch!  It still be a couple more days until things finally settle down, and maybe then I will have an idea on how far off #1 will be off and how TLC is comparing to KWSN.  TLC moved back into the #2 spot yesterday, but #1 may be a bit off again.  I guess we will have to wait and see. 

Hard work, enthusiasm and courtesy in one package...a minion speaks.
There has been a lot of forum heat generated lately with various people coming and going for various reasons - some better put and more believable than others. But I still stand by the principle that you can drift in as you please and likewise drift out - just don't expect everyone else to like it or be as laid back as me. Many people have pointed out that it's not so much your WU tally that counts but how you conduct yourself as a contributor to helping others and making this an interesting and 'fun' place to be. Yes I'm a little naive and innocent but it keeps me from being too cynical about the 'competition'. 

So please spare a thought for zAmboni (who has invested far more than any of us) and the pressure that this place puts him under. Keeping this page fresh, correlating mounds of stats, moderating the boards, unofficially Captaining the team and trying not to get too disheartened as situations unfold is a lot to put on one person. Yes, we volunteer freely and could stop anytime but you know it is just not that simple when this place starts to feel like home and you get to know people. I feel it and I'm tucked safely away in a quiet benchmarking backwater. zAmboni dedicates a staggering amount of time to this venture and I hope you all say a quiet mental 'thankyou' as you read this. I think that'll do from me for a while. I return you back to normal service.
Max out.

Changes At The Top
Has it been a week already?  I guess it has.

One look today at the member charts will show that there has been some big changes in the team makeup in the past two days.  A heaping handfull of top members on the team have left, and as a result, TLC is no longer the #1 team in the S@H project.  Here is a list of members who have left in the past two days:  Antiguru, Support AIDS Research, www.care.org, Admins spend CPUs, Krell, Wireless Digital Intelligence, nop, Runar, Angus, and mr_mann.  The majority of the list who left have joined up with the Knights Who Say Ni!.  Add them up.  That is a loss of over a million work units from TLC.  I'm not sure what will be come of X-mas Gift From Antimoses, but I wouldn't expect for that account to stick around either.

For the most part (with the exception of Admins spend CPUs and maybe Angus), they were a pretty tight knit group of members, so it isn't surprising that if one of them left, the others would follow.  Why did they leave?  I really have no clue.  There are hints that they weren't having "fun" anymore on TLC, and decided that they wanted to move on.  Recently in the Distributed Forums an internal contest of sub teams was started to foster some type of team competitions.  I honestly haven't been following the threads that much, and there may have been a small amount of dissention in there.  But just from a cursory glance there didn't seem to be that much dissention to cause a mass exodus from the team. 

Krell, Antimoses, and mr_mann have all been valuable members, good help on the forum, and a good conversation on the IRC channel.  I am sorry to see them leave, but I wish them well wherever they end up.  Hey, at least I have moved up 8 spots in the last two days!  :)

With the fallout, I am actually happy to see some of the die hard TLC members who have been through thick and thin claiming the top spots on the team now.  The top 5 on the team now are: 1) svdsinner, 2) Memnoch, 3) Vomesh, 4) C. Eric Smith, and 5) Knight.  They have been here through thick and thin, and will be here for the long run :).

Where Are We Now?
Here is a list of the teams after the fallout:

Team	Work Units
Sun Microsystems	3105779
SGI SETI	2677223
Team Lamb Chop	2667897
KWSN	2516668
Compaq	2400662

TLC has dropped down to the #3 spot in the overall standings.  It shouldn't take too long to retake the #2 spot, I would say a couple of days.  Right now with the switching of members over to KWSN the #2 position will not last for too long.  With their new members KWSN will be out producing TLC by quite a bit, and they will probably pass us in a couple of weeks. 

Version 3.03 Transition
The S@H guys have updated the text client download page, with some useful information to most.  First off there looks to be a table showing which client you should use for the operating system that the user will be running.  It looks like they have pared down the number of different clients, and some now have multiple functionalities.  The second thing of note on the page is for the client transition.  Here is what they have to say about it:

The transition to v3.03 is now in progress. All previous versioned clients should be receiving either upgrade warning messages, or version is obsolete messages. Porting of v3.03 clients continues and there will be additional v3.03 clients available on this page in the near future. Thank you for your patience. To verify your client version:
./setiathome -version

Clients previous to version 3.03 will probably be turned off completely by about 2001-01-31 (31 January 2001).

The warnings for the client transition has started already.  Using SetiQ right now, I don't really see the warning messages :).  Seeing that they are expecting to complete the switchover by the end of the month, I hope that there will be an update for SetiQ before the switchover happens.  If that doesn't happen, I may have to switch to a different caching program.

It is going to be interesting to see the fallout over the client transition.  There may be alot of people (on all teams) that will quit production rather than switch over to the new clients.  It is also going to slow down the different races in the stats.  It may slow down our overtaking of SGI and Sun, but I guess we will find out sooner or later.

Bunch o' News Today
I have been a bit lax on keeping up with the news, so there is going to be a big update today :).  First off is a new entry on the tech news page on the S@H site. 

January 8, 2001
Over the past several weeks, and especially last weekend, hardware problems on campus have been affecting the router which carries the SETI@home traffic. These problems cause large dips in our maximum bandwidth allowance, and therefore we've been dropping many connections to our server. Central campus is committed to fixing this problem as soon as possible.

Warning Out To Clients Already?
There has been talk of the Berkeley guys already rejecting results from older 2.04 clients.  I am not sure of the details of this since I haven't been running any of these clients.  It does make me wonder how they are going to handle the shutdown of the pre 3.03 clients.  If you take a look at their text client download page, you may notice that there are a bunch of clients that haven't even been updated to version 3.00 vintage.  Which of these clients will no longer be supported?  I have no clue, and there isn't any work on their site about how things will be handled.   The Berkeley guys wanted to have things switched over to v3.03 completely "in a few weeks" but how can they do that if many of the older clients aren't updated yet?  To fully switch over to all version 3.03 clients, they will have to shut down the old data server DNS, and this would affect ALL pre version 3.03 clients.  I cant see them doing a full switch over till all pre 3.03 clients are updated to version 3.03.  But again, who knows what they are thinking out there.

Problem with v3.03?
I saw this post on the alt.sci.seti newsgroup a couple of days ago from Matt Lebofsky, and this may or may not apply to you.  Some people (only a few though) have reported a problem, that they may need your help in solving.  I'll let Matt tell you about it:

Hello, Everybody!

So we're getting reports from windows users using the new 3.0x clients saying that multiple random directories on the machines are getting peppered with state.sah and key.sah files. Sometimes it's not so bad - a couple of extra files. Sometimes it's appearing in over a thousand directories, including on the desktop and in the start menu, etc.

I am usually in charge or tracking this stuff down, and am wholly unable to reproduce this bug on any windows machine in my possession. Nor have I seen with my own eyes a machine with this problem.

So I have a question or two:

1) Have you seen this behavior?
2) If so, do you know why it happened?
3) Have you seen this behavior in clients earlier than 3.0?
4) Have you seen this behavior in software other than SETI@home?

Something like this would normally be a show-stopping bug, but only one or two people report it a day, and as I said, I can't reproduce it. My guess it's some klunky virus-checking or disk defrag type software that's not properly handling open files (the key.sah and state.sah files are constanly opened and closed by the client to save the current state of progress).

Any info would be vastly appreciated.

Matt - SETI@home
[email protected]

I personally haven't seen this behavior or have heard of any others, bit if you have seen anything like this contact them with some info.

Site Stuff
i did alittle housekeeping with the site today.  First off I expanded the charts from members 3200 through 3800+.  WIth the new influx of members, a couple more pages needed to be added.  I also archived December's news and pared things down on this page.  Finally I have updated the weekly stats for the past week.  Check them out if you must!

TLC: Now and Then
I have just posted something new.  At the beginning of the new millennium, we take a look back at the last and see how TLC has changed since I first started keeping stats for the team.  It is a graphical interlude into the past.  Over the past year I have just been amazed at some of the progress the team has made, and looking at the graphs made my jaw drop even more.  Take a look for yourself and see how a team with such humble beginnings was able to rise to be the cream of the crop!   This first part of two takes a look at the team as a whole and its progress over the past year or so, hopefully in a couple of days I will have a member Now and Then also.  Keep an eye out for that!

Welcome!
I would like to welcome all of the new members to the team in the past day.  I guess all of you weren't driven away by my typo on the Ars front page yesterday/today.  (BTW: thanks to all of you who reminded me of that via email ;).   There needs to be some cleanup on the front page here, I will do that soon...heck I want all the new members to know what has been going on for the past month or so!  I'll get on that ASAP....ok as soon as I finish up some old business.  Been playing System Shock 2 for the past couple of days.  It may be bit of an old title, but it definitely some quality work and a pretty good game. 

v3.03 Mandatory Soon?
I missed this on my brief look through the newsgroups today, but I caught this in a thread started by Poof.  Here is what Matt Lebofsky had to say today:

We will start making the version mandatory starting as early as this week. We need to make some tweaks to our server to make the version warning messages work properly. We had a bug where people downloaded 3.03, finished their work unit and sent it back, and then their version was "obsoleted" by accident. How annoying. This'll be fixed and tested today or tomorrow. Then the mandatory upgrade messages will start, and a couple weeks down the line we're shutting down the old clients by shutting them down at the router (via a DNS switcheroo).

FYI, over 25% of our current results are coming in from 3.03 clients.

- Matt - SETI@home

Right now I am not sure how queuing programs will handle the upgrade messages, but I guess soon we all will find out.  At least we have a couple of weeks for Seti Queue to get updated for the version 3.03 client.

Yea Yea Yea...
It should be 4000 YEARS :).  I'm trying to get them to change it on the front page...

Weekly Stats r Up
Ok they are done *right* this time. Check them out here.  I noticed that the WU chart on the weeklies for last week was incorrect, I changed the graph and you can check that out here.

I am going to have to do something new with the stats pages.  I have gotten rid of some of the charts that are no longer needed now we are in the #1 spot.  But I need to come up with something else now :).  Ill be working on them soon.

-Front Page