LHC@home

June 13, 2005 - 19:00 UTC
There have been many failures over the past week. Another bug was found (and quickly patched) in our upload/download file server causing it to hang until reboot. Once that was remedied, both the scheduling server and the main web server had separate issues due to extremely high load.

In case you haven't noticed, we recently changed the URL setiathome.ssl.berkeley.edu - instead of pointing to the old SETI@home "classic" project, it now leads users to the new BOINC-based version. As expected, this vastly increased the number of new users joining the BOINC project, and therefore increased the strain on our back-end servers. Soon we will stop new classic account sign-ups altogether, and eventually stop accepting classic results outright (with advance warning) - each step potentially increasing the demands on our hardware.

At this exact point there is no new hardware that BOINC could use as its various servers fail for one reason or another. This is because the classic project is still active and using up half of our server farm. This was soon change.

The classic "master science database server" (a 6 CPU Sun E3500) will be the first machine to be repurposed. We're busy migrating most of its data onto a new database server (an 8 CPU E3500). This migration had been slowed by recent (recoverable) disk failures, but should finish in a month or so. Before then, however, we are going to move the BOINC scheduler onto it. The actual file upload/download handler will remain on its current server, thereby spreading the whole scheduler system over two machines.

As soon as possible, we will add a second webserver (and maybe a third). The BOINC web site contains far more dynamically-generated content than the classic site, and therefore needs more power behind it. We don't really have any spares, so some machines will have to double as web servers and whatever else they are currently doing.

And as if that wasn't enough to worry about, the BOINC replica database has continually fallen further and further behind the master database (because the load on the master increases and the replica hardware is relatively inferior). Then yesterday it was rendered useless as a binary log on the master got corrupted. This didn't damage the master database - only the replica. So we're going to have to build the replica from scratch (or hold off until we somehow obtain between hardware for that).

More to come as things progress...

---

IMHO they should already have stopped classic sign ups. BOINC SETI needs more computer power and they have already decided to go with BOINC. The sooner they terminate classic, though it did have some good features and did some pioneering work, the sooner they will have the use of all their equipment to devote to BOINC.

---

In my Opinon they shouldn't have moved to BOINC. BOINC for me causes more problems then it fixes. The one redemming Quality about BOINC is that it allows me to run other projects. Nothing else. I'll miss SetiSpy, SetiDrive and all of my old teams that are going to quit has soon has Classic ends.

---

Why not try to fix those problems? BOINC is much improved over SETI classic.

What kind of problems are you having with BOINC?

Jim

> In my Opinon they shouldn't have moved to BOINC. BOINC for me causes more
> problems then it fixes. The one redemming Quality about BOINC is that it
> allows me to run other projects. Nothing else. I'll miss SetiSpy, SetiDrive
> and all of my old teams that are going to quit has soon has Classic ends.
>

---

> Why not try to fix those problems? BOINC is much improved over SETI classic.
>
> What kind of problems are you having with BOINC?
>
> Jim
>
> > In my Opinon they shouldn't have moved to BOINC. BOINC for me causes
> more
> > problems then it fixes. The one redemming Quality about BOINC is that
> it
> > allows me to run other projects. Nothing else. I'll miss SetiSpy,
> SetiDrive
> > and all of my old teams that are going to quit has soon has Classic
> ends.
> >
>


The biggest one is the screensaver problem. It pisses me off all the time. So now i just turn it off.

---

Ok, so what problems are you having with the screensaver?

Jim

> The biggest one is the screensaver problem. It pisses me off all the time. So
> now i just turn it off.
>

---

> Ok, so what problems are you having with the screensaver?
>
> Jim
>
> > The biggest one is the screensaver problem. It pisses me off all the
> time. So
> > now i just turn it off.
> >
>

Problem getting it to run and stopping it when i want to use the computer.

---

A "future directions" on classic "predicts" the following schedule:
"Souther Hemisphere data recording 12-03
Southern Hemisphere Client 01-04
BOINC Astropulse 12-03"
My concern about forever putting off the migration in order to please all is that we are stuck scientifically from moving on. I would rather take the lumps of many complaining, and just close down to get it done rather than being stuck while pursuing the illusive dream of a smooth transition and making everyone happy. After that "bump" we could devote all seti resources to BOINC which will sooner solve problems. Avoiding that bump will just keep us stuck. After classic closes either we will have enough resources or we won't. SETI can deal with it at that time.
Further, I believe the bump will not be that bad. BOINC seti would use less computer time because, though it may be more complex, in BOINC seti you would only crunch a piece of data a minimum number of times instead of crunching repeatedly and unnecessairily just to keep the classic seti crunchers happy.
Classic seti has had plenty of time to get their final crunch on. I would like to see some new direction to seti rather than the same old issues revolving around how we can please both classic and BOINC users. Classic seti was a historic ground breaking endeavor that will go down in history, but it is time to move on. The forums on classic seti should be closed immediately (notice was already given) and e-mail should go out announcing the specific date that classic seti will stop taking downloads and the specific date they will stop taking uploads. Then the project can close and we can move on.

---

ED said:
My concern about forever putting off the migration in order to please all is that we are stuck scientifically from moving on. I would rather take the lumps of many complaining, and just close down to get it done rather than being stuck while pursuing the illusive dream of a smooth transition and making everyone happy.


But the thing is nobody cares what you want to do. You are not one of the scientists running this project. There is no us. WE do not have servers, etc. We are a bunch of people who were asked to donate some processing power on *our* machines. That is all you have control of, how much to donate your computers and who to donate it too. It is *their* project and they get to make the decisions. And I am sure they have many more reasons for what they are doing then you ever thought of.

---

--
David Stites
Mount Vernon, WA USA

David brings up a valid point on what the relationship between project managers and crunchers should be. I believe it is a bi-lateral relationship.
On the one hand, the projects need crunchers to accomplish their missions. They need feedback on what works and what does not work to be more efficient. I have seen suggestions in the forums incorporated to various projects to make them better. Crunchers care about these projects and support them with our computers and electricity.
Of course, the scientists have the expertise and should lead the project, but they owe the crunchers an occasional update as to the big picture and how things are going so we can all feel a part of the project. That is why all the projects I know of have some type of forum and some type of news on the front page.
On the other hand, users need to realize that projects aren't huge groups of people and resources who exist soley to make them happy as crunchers. Often times there are just a few people (SETI has two paid positions!)who are doing their best with limited resources. I too get tired of some people complaining as if they are losing their house when a project goes off line for a few hours.
Because I believe in a two way communication, I posted my concerns. SETI does not have to respond, of course, but I hope they do communicate a little more in their news or technical news as LHC has done.
Congratulations on 6,255 classic seti work units David! That is truly an awesome amount! I was crunching seti pretty steadily since it opened and I only had 874 wu's. You have really made a major contribution to science!

---