Posts by Moderator9

21) Message boards : Number crunching : There is nothing in my Team page. Why? (Message 18810) Posted 16 Jun 2006 by Moderator9 Post: I use IE, when I click on EspiTeam (Spain - Espa񡩼/a>, I get a blank page. tony Hm...I get a '?' instead of the 'n' in Espana. Perhaps you should try ñ instead whatever you typed for the 'n' ? Hmmm, What I pasted to the link I provided, is exactly what appears under his page. Must be a browser thing. tony Well it must be more than that. Safari, FireFox, and IE all seem to display it fine on a Mac. I do not have a Windoze machine handy to try it on but I will do so later. In any case I can't duplicate the error.
22) Message boards : Number crunching : Problems with download of WUs: Either now work or overcommitted. (Message 18806) Posted 16 Jun 2006 by Moderator9 Post: ...Hi Moderator, yes, I did read the FAQ. I did not touch the system for 5 days, it's set to connect every 3 days, the time-to-switch is set to 36 minutes and "Keep in memory" is enabled. Shorter connect times are not acceptable because SETI@Home and sometimes even E@H tend to break down for several days on a regular basis. What setting are you using for the "Time" Parameter? The longer time settings are designed for longer connection intervals, and the shorter times are for shorter intervals. But you have to be careful not to play with the setting too much if you are running a very long work queue. But here is the problem. With the long connect time the system will always want to load a lot of work units. If I read your reply correctly, you are processing three projects. The fact that these other projects might go down should not affect the way you are setting your system. With (at least) three projects you will always have work on your system. For example I run the same projects as you (in fact others as well). My connect time is set to .2 days. There are never more than 3 or 4 work units in my queue for any single project, yet there is always work. If a project is offline for a time, the system simply process for one of the other projects, until the problem gets fixed. My systems are never forced into EDF mode, and they are very stable. In your case you are forcing the system to run in "bursts". This will prevent the system from stabilizing for a very long time. Under the conditions you describe it could take 30 to 40 days for your setup to become truly stable, depending on your "time" setting.
23) Message boards : Number crunching : There is nothing in my Team page. Why? (Message 18803) Posted 16 Jun 2006 by Moderator9 Post: There is nothing in my Team page. Why? here is my team's URL: http://boinc.bakerlab.org/rosetta/team_display.php?teamid=2475 How can i solve it? I am not seeing a problem with the display. Perhaps the character set you are using on your system is not displaying the page correctly.
24) Message boards : Number crunching : Problems with download of WUs: Either now work or overcommitted. (Message 18793) Posted 16 Jun 2006 by Moderator9 Post: If it were me... I'd let it run the way it wants to for 4 days. So after 4 days I'd update to Rosetta, if it still sends no work then I'd do the "reset" project with Rosetta. In general, BOINC will handle it, and assure Rosetta gets it's proper resource share. I see both your machine's seem to be getting many client errors. Have you been able to track that down? I followed your advice and let it run for several days without any interfearance. I did not get any new work for 5 days but tonight it downloaded 30 work-units and is overcommitted again! Obviously the scheduling of Rosetta does not work at all. The scheduling of Rosetta is the scheduling of BOINC and yes the BOINC scheduler does not work. You may try to reduce your reconnect time to something belwo 2 days and you will see less overcommitted messages. Besides being overcommitted is in fact no problem other than it sounds stupid. You (probably) won't miss deadlines and that's the important thing tralala, can't be. Works perfectly with all other projects. Never had a problem with E@H, SETI@Home and Pirates on my Macs and additional Climateprediction on my P4. Must be a Rosetta problem then. Have you read the FAQs that discuss the time parameter? If you are constantly adjusting the system it will never settle down. You have to balance the connect time and the time setting with reasonable values and then let the system settle itself out.
25) Questions and Answers : Macintosh : rosetta running even when closed (Message 18756) Posted 16 Jun 2006 by Moderator9 Post: I assume you have tried using the "Activity Monitor" to abort the process. If not, run the activity monitor, select the process, and force quit the application. I have seen this happen once on one of my Macs as well. It is usually caused by BOINC aborting itself and then being restarted manually. It then looses control of the application for some reason. Failing the force quit, you can of course reboot the system by holding the power button until it shuts down and then restarting. In any case you will have to restart the system to get everything organized again. After the restart, the system should run normally.
26) Message boards : Number crunching : Is this for real??? (Message 18755) Posted 16 Jun 2006 by Moderator9 Post: I find it hard to believe that Rosetta or Boinc can make the point system in place now a fair and even field, way to Manny clients, OS's, and type CPUs , conch's, And Hyper Overclocking to deal with. Back to my original Question about RAC How can a member get a Rac that is 8X his total points I might be slow but I am having trouble understanding how the RAC is formulated I know it is not of any real value here. But I do like to understand how things work Well, this link provides the formula by which it is calculated. I am certain the WIKI also has more detail about this. But the short of it is the number does not mean anything. It only gives a basic measure of a contribution over time. The reason that very low scores produce very high RACs is because the time element of the formula does not balance out in the short term properly. For the formula to work properly you need a number of samples taken over a longer period of time. Remember this is a "average". the more samples used in the calculation, the more accurate the average will be. Think of it this way. If the RAC is calculated for a single day, and a person submits one result for 25 credits in the first hour, you would logically project that that person should produce 600 credits over the 24 hour period based on the submission in the first hour of the day. Over time the falsehood of the assumption dissolves when it turns out the system actually only produces that single result in a 24 hour period. Eventually that RAC would become 25 over time.
27) Message boards : Number crunching : Report Problems With BOINC SERVER UPGRADE (Message 18745) Posted 15 Jun 2006 by Moderator9 Post: I just tried attaching again and got the same error message. Also it says Incomplete request received. I used my account key in place of my password because I've forgotten it. Could this be the problem? jas Yes that could be the issue. The advice provided by David Kim is the way to go. Just change the password to something you know and use it to attach.
28) Message boards : Number crunching : Is this for real??? (Message 18742) Posted 15 Jun 2006 by Moderator9 Post: Dear Moderator9, First of all, I'd like to apologize for warming up a topic that has been discussed to death. I wanted to know, if I'm doing something wrong. I was looking for a fault in my installations. After reading Jose post and my posts again, I agreed with Jose and that is why I asked to remove my posts. They could cause disturbance. That is something I don't want. This is the most important reason, why I don't want to cause any further trouble. Thank you for your detailed explanation. I will try to be more careful in the future. Kind regards, Jochen @Laurenu2: As recommended, we should continue this discussion in private. I'd be happy to hear from you. My e-mail adress is: jr <at> jr101.de Jochen, Please do not misunderstand my post. I was not upset with you about this thread. But I will do as you asked.
29) Message boards : Number crunching : Is this for real??? (Message 18727) Posted 15 Jun 2006 by Moderator9 Post: I really did not want to offend or attack someone. If someone felt attacked or offended by me, please excuse me. cu, Jochen [Edit:] I just tried to edit my posts, but I can't anymore... Is there a moderator that could delete them? Jochen, Yes, I can do that. But which ones do you want deleted? As a personal aside (Jose need not read further), The Rosetta forums are open for discussion of any topic, so long as the discourse is polite and professional. While I see no need for my intervention as a Moderator in this discussion beyond explanations as necessary, this topic has really been discussed to death. There are 70,000 people participating in Rosetta. Very few have raised this as an important issue. But from my point of view it is counter productive to keep raising the issue in the context of the thinly veiled threat that masses of people will leave if it isn't fixed right now. The project team is aware that a very few systems may be claiming higher credits than might be justified by the system type. They are working on a credit system that does NOT rely on information that might be altered by a user (or the users BOINC client). When this is in place there will be an announcement. In the meantime it is unreasonable to expect the project to drop everything and spend a lot of the limited programming resources available to specifically address this problem ahead of other issues. Let me be clear about this. They are already working on the issue, but it is NOT going to be handled ahead of the CASP work which is being conducted against a very unforgiving set of deadlines. The new credit system will be integrated into RALPH as new versions are prepared for testing. I along with a few others were asked by the project to take a look at some of the credit claims made by some systems, and recommend a solution. The fast solution would be to simply delete the users who "seem" to have questionable credit claims and purge the credits from the system. I think you can all agree that would not be the best solution. It is arbitrary, and open to a lot of subjectivity. So that means something must be done that would require a more surgical approach. But that also means it will take more time. For every credit granting solution I have seen proposed from the user community, I have seen an objection raised, from another part of the user community. In other words, one persons solution is another persons poison. At present, the Flop counting method is the most fair proposal on the table, and some people do not like that either. No matter what credit system is used, there will be someone who thinks it is unfair. The Rosetta project team, working in concert with their counterparts among the BOINC developers, have a working plan they are implementing. But it will require coding and testing on RALPH before it will be deployed in Rosetta. That will take time. How much time is anyones guess, it depends on a lot of factors. But no matter what that schedule works out to be, arguing over it on the forums will not influence or speed up the process. Nor will threatening to leave the project if it is not fixed fast enough. The fact is that of the 149,000 systems currently shown as attached to the project less than 70 have been singled out by the user community as problematic. That represents a very, very small percentage. While it could be argued that there are more than 70 and they just have not been reported, the current evidence does not support that position. There is a larger set of systems that fall into a grey area where they may be claiming higher than normal credits, but not so high as to be completely impossible to justify. With these systems there may be any number of reasons why the claims are in fact legitimate. I for one am not prepared to enter into a witch hunt and demand that these system owners submit justifications for their credit claims. What is forgotten in all this is that there are in fact a lot more systems that actually claim LESS credit than they should. Either the benchmarks are off, or there was some other process that reduced the benchmark when it was run. Also many Linux systems claim LESS credit than they should just because they are running Linux. The project should also provide universal EXTRA credit to those systems as a matter of fair-play, but I don't see many of you arguing for that. So my points are these. The project is working on the problem. It is not as widespread as these discussions would have people believe. The immediate solutions available to the project will not make everyone happy. Even the long term solutions will not make everyone happy. While I frequently see people say that "The science comes first...", that is almost universally followed by a comment like "if you don't fix the credits, I and a lot of people will leave". This project is doing work that WILL save lives, maybe even yours. How long should the project team shut down the real work to fix the credit system? Either the science comes first or it does not. You can't have it both ways.
30) Message boards : Number crunching : Rosetta work file downloads (Message 18712) Posted 15 Jun 2006 by Moderator9 Post: i just did a test on the file: hom001_aat325_09_05.200_v1_3 (from within hom001_aat325_09_05.200_v1_3.gz). the gzipped file that rosetta downloads is 6486 kB. i tried recompressing using zip and rar compression and got sizes of about 6420 and 3300 kB, so rar gives about half the size of zip and gz. so maybe you could try using another compression format (instead of gz, maybe bz2 or 7z --- i'm too lazy to download and test these) and you might be able reduce the download size. this would be very helpful for the dial up users. This has been discussed before (more than once). The issue is not as simple as just changing the compression methods used. This is all intertwined with the BOINC software as well. The project is currently using the data transmission and compression system that is integrated into BOINC. In the last few weeks, the size of the files has been reduced by 33%. The user adjustable "time" setting was developed to reduce the need to download work units and still keep a system busy, specifically for modem users. If you are on a modem, you should have the "time" setting adjusted for as long a period of time as it will allow. Right now that is 24 hours. So if your setting is 24 hours, you would have to download a single work unit every 24 hours of CPU time. The maximum time adjustment will eventually be on the order of 4 days. So under those circumstances, the amount of data downloaded over time will be significantly reduced. In any case there will not be any change to the compression used during CASP. This is just a very bad time to introduce new issues that could cause new problems.
31) Message boards : Number crunching : Is this for real??? (Message 18711) Posted 15 Jun 2006 by Moderator9 Post: @Laurenu2: You are one of those, having a BOINC version 5.5.0 with far too high benchmark results. What kind of BOINC client are you using? Why are the benchmarks that high? Can you tell me what computer node are you referring to that is to high. I even get lost in my network 70+ nodes Please give computer name not computer ID # I have been restocking with new AMD X2's, are these the ones your talking about If you look at the Top participants My first listing is about 20+ pages down and the people in front of my first listing are using lesser computers then I. So I would guess I am not getting that much more points for the work done But as I said to the system Lords it would be best to make it all even for the same work done Rac is not calculated by the Project. It is calculated by all the stats sites. However you can artificially raise the RAC by extending the time between results reporting to the system. In other words, assuming two identical systems, one reporting in every few hours, and the other reporting in one a week. The system that reports results once a week will show a higher RAC. The Rosetta project is doing what it can in dealing with any credit issues. Since the calculation of RAC is not part of the project, that will have to be done somewhere else.
32) Message boards : Number crunching : Report Problems With BOINC SERVER UPGRADE (Message 18706) Posted 15 Jun 2006 by Moderator9 Post: I've been tryingt to get my computer to share it's resources wiht the Rosetta@home project, but when I try to register I keep getting an error message that states: Failure to attach. I've installed the lastest BOINC 5.4.9 and originally attempted to work with Rosetta on 13 Feb 2006. What do I need to do to get this program to work? jas Since you already have an ID here on the Rosetta site, when you use the BOINC project attach function you must check the "use existing account" button when it askes for the user ID and password. It should then attach and begin loading software.
33) Message boards : Number crunching : Report Problems With BOINC SERVER UPGRADE (Message 18702) Posted 15 Jun 2006 by Moderator9 Post: I'm in the same boat, just the one line saying not enough memory then the box inexplicably sits idle for 24 hours unless I manually intervene. Is anyone working on this? I found 4 of my boxes sitting idle this morning again due to the same problem. What a waste of time and resources. When this happens I'm getting about half the work I should be. Multiply that by the thousands of computers ruuning Rosetta and this must be a huge problem. I a Post lower down in this thread, Rhiju and Bin are in fact aware of and working on this issue. However the problem is not as wide spread as you might believe. Most systems that are seeing this error message are also still getting work.
34) Questions and Answers : Macintosh : Rosetta/BOINC hangs at x% (Message 18677) Posted 15 Jun 2006 by Moderator9 Post: You may want to take a look at this FAQ. I think it may explain your issue. The short of it is that the Work unit will stay at 1% for quite a long time, before progressing depending on the particular type of work unit and protein. This is also somewhat dependent on the time parameter you have set in your preferences. In any case all work units will complete at least one model and that may take as long as 5-8 hours for a large protein. During that time the percentage will show some value less than 2%.
35) Message boards : Cafe Rosetta : instability with new version? (Message 18639) Posted 14 Jun 2006 by Moderator9 Post: I am not following rosetta closely but it seems that the layout of the screen has changed, and following that there has been crashes and freezes among the workgroup computers. Is there any active attempt to make it more stable? We can't let our computer freeze everytime it goes into screensaver for extended periods. It didn't happen before. Please advise. Thank you. Yes, of course people are working on the issue. It is not really that the screen saver is new or even really a constant issue. However it had to be modified to accommodate the CASP work wherein there is no known structure at the beginning of the process and the RMSD value is also unknown. As a result, to properly display the information for a CASP target, the screen display had to be modified. Moreover, while the issue may surface while using the screen saver, it is not likely to be isolated to that function. There have been massive changes to the nature of the work in preparation for CASP, that are more likely to be the cause of any issues.
36) Message boards : Number crunching : Rosetta work file downloads (Message 18628) Posted 14 Jun 2006 by Moderator9 Post: Hi there, i am new to this project and just have a question about the download/upload of work files. I am on a restrictive internet connection, and would like to know how much downloads/uploads are used on average when working on a project. Are we talking megabytes? >10mb per work unit? More? Is there a way i can contribute without getting assigned huge projects that make my internet download bill skyrocket? any help appreciated -mitch The initial download might be as large as 15Mb, but this includes the Rosetta application. The average Work Unit should be about 3Mb. The average upload is less than 500k. the project has a user selectable "Time" parameter which is designed specifically to help your situation. See this FAQ.
37) Questions and Answers : Windows : Suis en france & comprend pas anglais, besoin adresse e-mail pour messages (Message 18611) Posted 14 Jun 2006 by Moderator9 Post: Bonjour, je ne parle pas le français. Mais j'essaierai d'aider. Pouvez-vous faire vos questions simples ? Rosetta traite-t-il votre ordinateur ? Avez-vous besoin de l'assistance ? Pour envoyer un e-mail, s'il vous plaît utiliser le lien dans ma signature. "Moderator Contact".
38) Message boards : Number crunching : The SETI thing was over credits? (Message 18609) Posted 14 Jun 2006 by Moderator9 Post: I must say that I'm not here to bash SETI. I give credit where credit is due. The SETI staff not only invented the Idea of volunteer DC, but they proved it could work. That said; I stopped crunching for SETI (Not for credits or cheating issues) because of the shape the message boards got into. One never really felt confident in asking questions or felt very welcome by the community of volunteers. Only now is there some air of civility emerging, but the atmosphere is still very charged over there. Here, even the single PC user gets a warm welcome by all of the computer farm owners who can get RAC's over 2000. And I get a clear picture as to what my contribution is. Here you never feel small with a single machine. Thank you for the kind words. SETI has fostered a rather unique and historic way to bring basic science research to the masses. But each project is unique and each has it's own issues to address. The Rosetta project team tries very hard to support all contributers, large or small. The user community here is full of helpful people who make the experience better for everyone as well. Many of the more prominent among them have already posted in this thread. Their contributions to the project have not gone unnoticed by David Baker and David Kim either. Also don't forget the system admin staff working totally out of view of the wider user community without whom the servers would not run. Currently the project is improving daily thank in no small part to the users who are willing to suffer the issues first in Ralph so the Rosetta users will have a better experience. But most importantly it is thanks to all of you, the users, that this project is moving in a positive direction.
39) Message boards : Number crunching : Report Problems With BOINC SERVER UPGRADE (Message 18607) Posted 14 Jun 2006 by Moderator9 Post: When I try to update my "Message board preferences" I get the following error: Couldn't update forum preferences. Unknown column 'minimum_wrap_postcount' in 'field list' Is there a fix listed for this? David Kim reports that this issue has been fixed.
40) Message boards : Cafe Rosetta : Moderator contact thread archive (Message 18597) Posted 13 Jun 2006 by Moderator9 Post: When I try to update my "Message board preferences" I get the following error: Couldn't update forum preferences. Unknown column 'minimum_wrap_postcount' in 'field list' Is there a fix listed for this? I have only seen this one other time, and it was related to the size of an avatar. I will see if the system admin can find anything about this. EDIT: I send you a message off line regarding moving your original message to the server upgrade problem reporting thread (here). It should get faster attention there, and I have sent a message to the system admin.

Previous 20 · Next 20