I must admit that because of the instability of the project over the past week or so I reduced my resource share to Rosetta from 45% to 16%. I will be happy to reverse it when I observe stability again. My results page indicates that I have several WUs that are either in progress or past due. As far as I know I processed and returned everything that was sent to me. Lost credit & lost science? This is a psychological downer.

Some of the older versions of the BOINC Client Software could "lose" work because of errors during the transfer process. The later versions have additional checks to help prevent this problem.

Without knowing which version you are running it is hard to say why the work may have been "lost".

To Paul Buck and David Kim:
I'm posting here only because this is where it started for me. Is the system stable? I bumped up my resource share to Rosetta (Nov 9) thereby increasing the flow of WUs to my computer. Three times today I requested work and received the msg "Scheduler Request Failed...with a return value of 500", followed by the msg "No Scheduler Responded". I figure no big deal, will get some work in the next request, and I did eventually. Hours later I was looking at my Result Page and it shows WUs supposedly sent to me that I never received and all of them were sent at the exact time I received the "No Scheduler Responded" msg. I've discovered this same phenomenon at LHC so it isn't unique to Rosetta (I think). Now I have WUs all over the place that I can't possibly complete since I never got them. I'm on v5.2.5 and using cable modem. I don't feel so bad about NOT getting WUs since I don't waste CPU time on them, as opposed to reporting a completed WU and not getting credit. Still, your servers are expecting the return of WUs that I never got and can't return. Be glad to help you with this if it really matters; at least I now recognize what is happening to me.

---

It may be a problem with your client. It sounds weird to me and I haven't seen any posts with similar problems. The system is stable. If it's not too much trouble, can you try the most recent BOINC client version. They have had quite a few releases in a short period of time so I imagine they may have been ironing out some issues. The current release is 5.2.7 for windows. Don't worry about the fact that our server thinks you have those work units. They will eventually expire on our end and it will not effect anything.

Since Sunday, Vatson, a graduate student, Bin, a postdoctoral researcher in my group, and I have found that this dramatic movement towards the native structure occurs for a number of different protein systems.

Congratulations on the breakthrough, they're always a plus.

You mention that several protein systems show this "change to a slope", with your new algorithm. Are these all naturally occuring proteins, or are some of them "test" ones that are created in the laboratory. Assuming such a thing is possible, which I think it is, (recombinant DNA to make insulin?)

---

It may be a problem with your client. It sounds weird to me and I haven't seen any posts with similar problems. The system is stable. If it's not too much trouble, can you try the most recent BOINC client version. They have had quite a few releases in a short period of time so I imagine they may have been ironing out some issues. The current release is 5.2.7 for windows. Don't worry about the fact that our server thinks you have those work units. They will eventually expire on our end and it will not effect anything.

For what it's worth, just moments ago I changed the setting for LHC from "No New Work" to "Allow Work". Boinc requested work and I got the same "No Schedulers Responded" msg. Moments later another request was sent and the same thing happened again. I checked the LHC Results page and sure enough there were 2 WUs supposedly sent to me but I never got them. YIKES! This seems like a REAL problem I will upgrade to v2.5.6 or whatever is current. Let you know what happens.

---

... v2.5.6 ...

5.2.7 :)

---

A 500 error code indicates that your comptuer and the server are not communicating properly. The "good news" is that you have a workable problem. The bad news is that I am not very good at networking problems ... :(

So, upgrade is the first step.

Also, if you are on a home network with a router and so forth ... after the up grade and your system is running with the new BOINC Client Software. Shut the entire network down to cold iron ...

There are times when my cable modem, routers, computers, and the outside world "pick on" one of my computers and it will have issues when the rest of the home system works fine.

After that, well, now we need Tony and Tigher ... :)

---

Breakthroughs are always the best :-)

I've been running a few projects the past few days, crunching some workunits of every project I could find. I've also been reading the forums on all projects. I'm new to distributed computing projects. Three projects are running right now: Seti@home, predictor@home and rosetta@home. The other projects I've tried are dropped for various reasons.

Having spent some time reading on the various message boards of the projects, I'm confident that this is the project I'm going to do. I like regular updates on websites and I love project developers and scientists who spend time to explain the project they are doing to regular people like me.

It's like listening to a speech or going to a meeting where a scientist is trying to persuade a room full of people to spend money on his project. He has to persuade you with his best arguments and explanations. Asking people to donate their computer time isn't done in a conference room or in some fancy government building, but it's done in a public forum on the internet. Right here, on this message board and this website.

People need to know they are doing something useful and reading: "A BREAKTHROUGH" is of course the best argument possible. This actually says: Thank you for your (computer) time. You've helped us: a breakthrough. Something that couldn't have been done without your help.

Great project, great people!

---