A THIRD science application for malariacontrol |
Message boards : Malaria Control : A THIRD science application for malariacontrol
Author | Message |
---|---|
This post was last updated on 9. of May 08. |
|
ID: 3773 | Rating: 0 | rate: / | |
Thanks for the update and of course the work \"behind the scenes\"! This really sounds interesting and I\'m lloking forward to run the new app! The insights from those calculations will help us to improve the main malariacontrol application in the future. Will this only be an intermediate application to further improve the main app or will it stay as a stand alone application? Thanks ____________ |
|
ID: 3774 | Rating: 0 | rate: / | |
thanx 4 the thanx:) the app will probably stay for quiet some time, a year or so, depends on what comes out of it.. there might be no work for some time and then it starts again.. the actual task it performs might also change at some point later, but we would announce that.. ____________ Michael |
|
ID: 3775 | Rating: 0 | rate: / | |
Neat! |
|
ID: 3779 | Rating: 0 | rate: / | |
Looking forward to getting a few of the Java wu\'s, just made sure I\'ve got the latest Java installed! |
|
ID: 3780 | Rating: 0 | rate: / | |
Neat! Java has been on the BOINC agenda for some time, but as far as I know there\'s currently no public project that\'s running a Java app (please correct me if I\'m wrong). A few BOINC-Java related activities:
|
|
ID: 3782 | Rating: 0 | rate: / | |
This sounds like a very interesting approach, but I am curious if you would be willing to provide more details regarding the unsolvable equations. Given the MLE framework in the modeling of the article cited above, the individual simulations make sense. I am curious, however, whether you considered Bayesian approaches to the problem since a set-up with strong priors would seem to make sense (at least given what I was able to get from a cursory skim of the article and Metroplis-Hastings approaches to Bayesian MCMC models). |
|
ID: 3787 | Rating: 0 | rate: / | |
Calculation is done by a java program, contained within the standard boinc-\"wrapper\" application. You need to have java installed.. if not, you will be prompted to do so. Will this stop BOINC dead in its tracks though, or are we to test that as well? I\'ve got the latest Java installed, but could try to uninstall it, if needed. ____________ Jord. BOINC FAQ Service |
|
ID: 3791 | Rating: 0 | rate: / | |
Calculation is done by a java program, contained within the standard boinc-\"wrapper\" application. You need to have java installed.. if not, you will be prompted to do so. I\'ve had a hard time in the past installing Java on Ubuntu....doesn\'t want to get past the \"license\". If I can\'t solve that, I\'ll have to remove my Linux machines from Malaria to run Java apps on my Windows machines. I do not know of a way to select running your test apps on a per OS basis. |
|
ID: 3797 | Rating: 0 | rate: / | |
I do not know of a way to select running your test apps on a per OS basis. Ever thought about using the different venues? Set your Linux machines up with a different venue than your Windows machines. (default home location) ____________ Jord. BOINC FAQ Service |
|
ID: 3799 | Rating: 0 | rate: / | |
I do not know of a way to select running your test apps on a per OS basis. Does that allow some computers to process \"test\" WUs and others not to? I thought the option to run \"test\" WUs was global to the project. |
|
ID: 3800 | Rating: 0 | rate: / | |
No, the test option is set by the project preferences, which works on a venue basis. So you can set a separate venue without the test application for those machines that can\'t run them and one (the default venue for instance) for those machines that can run the test applications. |
|
ID: 3801 | Rating: 0 | rate: / | |
No, the test option is set by the project preferences, which works on a venue basis. So you can set a separate venue without the test application for those machines that can\'t run them and one (the default venue for instance) for those machines that can run the test applications. I do not see that the project preferences work on a venue basis. Where am I missing that setup. I go to my account and can either set general prefs or project prefs. I do not see any setting under general prefs that will let me use work, home or school and set project prefs differently for one of those venues. Thanks. |
|
ID: 3802 | Rating: 0 | rate: / | |
No, the test option is set by the project preferences, which works on a venue basis. So you can set a separate venue without the test application for those machines that can\'t run them and one (the default venue for instance) for those machines that can run the test applications. Edit........found what you were saying. I did not have the venue settings set up under project prefs..........not sure I have ever been aware that was avail.......I\'ll rethink the way I\'m using venues. Thanks for your help. |
|
ID: 3803 | Rating: 0 | rate: / | |
Calculation is done by a java program, contained within the standard boinc-\"wrapper\" application. You need to have java installed.. if not, you will be prompted to do so. I did get the JRE 6 set up on my Linux boxes.......I couldn\'t see how to \"test\" the Java, but I\'ll watch for some of the Malaria \"test\" WUs to see if they work. |
|
ID: 3804 | Rating: 0 | rate: / | |
Thanks for your help. You\'re welcome. Always glad to be of some help. :-) ____________ Jord. BOINC FAQ Service |
|
ID: 3805 | Rating: 0 | rate: / | |
I do not know of a way to select running your test apps on a per OS basis. Jord, thanks for the tip! Will re-venue some of our boxes as well. Cheers, Rog. ____________ |
|
ID: 3806 | Rating: 0 | rate: / | |
I haven\'t seen a java wu as yet. Has anyone? |
|
ID: 3808 | Rating: 0 | rate: / | |
I haven\'t seen a java wu as yet. Has anyone? Not yet.....we were getting worried that we had a problem so thanks for the post. We haven\'t seen any mappredictors either.....Cheers, Rog. ____________ |
|
ID: 3811 | Rating: 0 | rate: / | |
I haven\'t seen a java wu as yet. Has anyone? Paul, they will be release on the 17th or there after. See top post. A new science application called \"optimizer\" will be launched from Monday 17.September onwards, or shortly there after. Hmmm, I could have re-read this last night before telling j2satx about venues. In addition, only windows hosts will get work. Oh well. :-) ____________ Jord. BOINC FAQ Service |
|
ID: 3812 | Rating: 0 | rate: / | |
I haven\'t seen a java wu as yet. Has anyone? Missed that myself...getting old eyes...... The venue info was/is still valuable.......I\'m considering using it to differentiate between my 32-bit and 64-bit puters. Thanks again. |
|
ID: 3814 | Rating: 0 | rate: / | |
Could have sworn it originally said the wu\'s would be released from Friday onwards......maybe I\'m just too eager! Thanks for pointing it out though Jord. |
|
ID: 3816 | Rating: 0 | rate: / | |
hi all, |
|
ID: 3869 | Rating: 0 | rate: / | |
Just noticed I have some in my cache....looking forward to crunching them! How long will they take? |
|
ID: 3870 | Rating: 0 | rate: / | |
hi, we were looking into bayesian alternatives first, but i don\'t see a way how to do it.. first,introducing prior knowledge about the shape of the survival distribution of infections would be dangerous, because we know that the duration of infections depends on other factors (e.g. age -> more acquired immunity), but do not know exactly how (since we don\'t have data that could give prior information about immune people). You might get a result, but you\'ll have no idea, whether it is true.. so you can as well let it be. Second,we have not found a way to introduce a shape parameter (which you could use to incorporate such prior knowledge), which still allows us to calculate the frequency of observed sequence patterns.. suggestions always welcome.. More detail about the Mathematics of what we\'re doing is here. have a nice weekend.. ____________ Michael |
|
ID: 3872 | Rating: 0 | rate: / | |
Nice, checking in Task Manager reveals that not optimizer_1.18 is using the CPU, but something called transmission_11 .. How\'s that for confusion. :-) |
|
ID: 3878 | Rating: 0 | rate: / | |
So it ran for all of 10 minutes. Finished all right. |
|
ID: 3879 | Rating: 0 | rate: / | |
Just noticed I have some in my cache....looking forward to crunching them! How long will they take? My box has completed 2 wu so far: the first took 36 seconds, the other 5774 seconds. ____________ |
|
ID: 3880 | Rating: 0 | rate: / | |
Processed 10 \'opty\' WUs: no problems encountered. Run time varied from 325 to 7183 secs. Ten corresponding *.jar found as described....Cheers, Rog. |
|
ID: 3881 | Rating: 0 | rate: / | |
more than 30 PC run since more than 6 month. I hope you don\'t make more of this \"delete after next reboot\" things with bigger temps. |
|
ID: 3882 | Rating: 0 | rate: / | |
I\'m getting a quite few going into computational error due to max cpu time being exceeded...... |
|
ID: 3884 | Rating: 0 | rate: / | |
more than 30 PC run since more than 6 month. I hope you don\'t make more of this \"delete after next reboot\" things with bigger temps. ksba, we\'re really sorry about this.. in your case I understand that its kind a tiresome to reboot 30 pc\'s if you don\'t have to.. We will certainly fix this, before leaving the testing state, and might have found a way to do it. We\'re working on it.. In the meantime, I would recommend to opt out of running the optimizer app, by checking the \"no\" box (for \"run optimizer app\") in your account-> project settings. So you can still get other testing work, just not from this application... and long live kantonsschule baden.. although, I was in Zofingen, that\'s also not bad.. maybe one day ksba will enjoy a similarly high international reputation as zofingen, and this will then be highly merited:) ____________ Michael |
|
ID: 3903 | Rating: 0 | rate: / | |
I\'m getting a quite few going into computational error due to max cpu time being exceeded...... gas giant, thats strange, but I saw them. We raised the maximum cpu time a bit.. please check if it helps.. the application terminates normally after 2 hours maximum(physical time), and you get credit for it, even if the calculation did not come to an end by then (explanation below). You must have extremely fast computers to \"use up\" so much cpu time whithin so little physical time.. :) To all others, thank you for the feedback! note: Some of you noticed the huge variation in calculation time. This is normal, and is because certain parameter combinations result in a lot of work to do, and others not. E.g., if you have a very high rate of new infections, and at the same time your infections last for a long time, individuals will accumulate a high number of infections which have to be kept track of. In the opposite case, if the infection rate is very low, and clearing rate of infections is high, then there will be almost no infections in a given individual.. thus it is very quick to calculate. The workunits which take very very long are likely to be outside realistic parameter combinations, therefore we set a cutoff, currently at two hours - which we might reduce if possible. Those should appear mainly in the beginning of a new run, and when the search algorithm comes closer to the solution, they get less and less.. ____________ Michael |
|
ID: 3904 | Rating: 0 | rate: / | |
I will post it here because no answer yet |
|
ID: 3905 | Rating: 0 | rate: / | |
Hi, |
|
ID: 3906 | Rating: 0 | rate: / | |
Have had a couple opt-wu\'s on my Athlon (windows) host now and all seemed to run fine. |
|
ID: 3909 | Rating: 0 | rate: / | |
Hi, hi, abort it, if it took longer than 2 hours. All the things you mention do not indicate a failure though.. the \"optimizer\" process is the boinc wrapper, transmission_<version_num>.exe is the java process.. so its normal that optimizer doesn\'t use cpu. There is no progress bar, because the application doesn\'t do checkpointing. ____________ Michael |
|
ID: 3910 | Rating: 0 | rate: / | |
I will post it here because no answer yet try to reset the project. if it doesnt work, reinstall java. That the output file is being absent means that the java app was not running for some reason - and didn\'t produce an output file.. ____________ Michael |
|
ID: 3911 | Rating: 0 | rate: / | |
ksba, we\'re really sorry about this.. in your case I understand that its kind a tiresome to reboot 30 pc\'s if you don\'t have to.. Well, you don\'t have to. I just tested deleting the temp*.jar files from my WINNT\\Temp folder by hand and it didn\'t hurt my system one bit. I had accumulated around 10 of them. Deleted them, cleaned out my recycle bin. I don\'t see the need to reboot, if there\'s no need to reboot. Just doing a little housecleaning by hand once in a while.:-) ____________ Jord. BOINC FAQ Service |
|
ID: 3912 | Rating: 0 | rate: / | |
I will post it here because no answer yet You say I must install the java platform first? I havent any java installed, thought that BOINC is doing all alone and the app is integrated within?! |
|
ID: 3915 | Rating: 0 | rate: / | |
I will post it here because no answer yet This is a new set of data that Malaria is sending out. It auto senses if you have Java installed and is run only if you have the \'run test apps\' turned on. It also only runs on Windows machines. In another section they just said that this run is over and the results have been sent to the scientists. ____________ |
|
ID: 3917 | Rating: 0 | rate: / | |
I will post it here because no answer yet That is not correct. Map Predictor run 5.20 just closed out. The Optimize app is described in the first post of this thread and it requires the user to have a valid Java environment available to run. |
|
ID: 3919 | Rating: 0 | rate: / | |
I will post it here because no answer yet Ok, thx, I have overread this below ^^ |
|
ID: 3920 | Rating: 0 | rate: / | |
It is working now but found something odd, suspend a wu and after some time the WU finished successful. |
|
ID: 3921 | Rating: 0 | rate: / | |
First opt WU run here. It ran 0.05 secs and quit with 0 credit claimed. |
|
ID: 3922 | Rating: 0 | rate: / | |
Yup, it looks like it was. Notes: 1. Javaw.exe runs at normal priority instead of low...according to task manager. 2. No progress indicator (has been noted before) and no Time to Completion Still running...will continue to monitor. |
|
ID: 3923 | Rating: 0 | rate: / | |
It finished after about 45 mins. The stderr.out shows the same contents as the WU that ran for 5/100s of a second...claims 0 credit. This is from the log: Project Date Message malariacontrol.net beta 9/25/2007 9:22:22 PM Starting opt_1_-44_5_390943439_3 malariacontrol.net beta 9/25/2007 9:22:22 PM Starting task opt_1_-44_5_390943439_3 using optimizer version 118 malariacontrol.net beta 9/25/2007 10:07:39 PM Computation for task opt_1_-44_5_390943439_3 finished SETI@home 9/25/2007 10:07:39 PM Resuming task 07mr07aa.4995.15614.15.6.190_1 using setiathome_enhanced version 528 --- 9/25/2007 10:10:45 PM Resuming network activity malariacontrol.net beta 9/25/2007 10:10:46 PM [file_xfer] Started upload of file opt_1_-44_5_390943439_3_0 malariacontrol.net beta 9/25/2007 10:10:51 PM [file_xfer] Finished upload of file opt_1_-44_5_390943439_3_0 malariacontrol.net beta 9/25/2007 10:10:51 PM [file_xfer] Throughput 696 bytes/sec malariacontrol.net beta 9/25/2007 10:11:08 PM Sending scheduler request: Requested by user malariacontrol.net beta 9/25/2007 10:11:08 PM Reporting 1 tasks malariacontrol.net beta 9/25/2007 10:11:12 PM Scheduler RPC succeeded [server version 601] malariacontrol.net beta 9/25/2007 10:11:12 PM Deferring communication for 11 sec malariacontrol.net beta 9/25/2007 10:11:12 PM Reason: requested by project |
|
ID: 3924 | Rating: 0 | rate: / | |
9/26/2007 11:15:45 CEST |malariacontrol.net beta|Starting task opt_3_-23_5_110673634_2 using optimizer version 118 |
|
ID: 3932 | Rating: 0 | rate: / | |
The application uses the BOINC wrapper which means that there is no progress shown. If you want to report the results then just press the update button. |
|
ID: 3933 | Rating: 1 | rate: / | |
My Sempron 3000 stop by 0,03 claimed 0,00 but the Time is allright, not the 0,03 and Client state done. |
|
ID: 3941 | Rating: 0 | rate: / | |
hi all, Hi Michael, Perhaps I\'ve come to a wrong conclusion. My first 3 opt_ WUs each claimed close to 0.05 secs (although two of them ran 45 mins or longer) run time and 0.00 credit. When their wingmen came in I was awarded credit anyway. I didn\'t see anything valid in the stderr_out section of each WU, so I assumed they returned garbage and I turned off the option to receive these WUs. Do you want me to continue crunching opt_ for any debugging purposes? or did these actually return valid data I\'m unaware of? WU1 WU2 WU3 Since WU1 only ran a few seconds and I had a Firewall message pop-up asking for permission for Java, it is probably no good. The others I just don\'t know about. Randy |
|
ID: 3946 | Rating: 0 | rate: / | |
hi all, |
|
ID: 4023 | Rating: 0 | rate: / | |
hi all, Since you have our email addresses wouldn\'t it be helpful to email those owners and give them the chance to fix that or perhaps stop sending units to those machines. If one of my machines has a problem and no one tells me, how would I know? What if I didn\'t check them every day? They could just eb set to do their own thing and I could go months without doing anything to them. I have 19 mahcines on line and then take about 30 minutes or so everyday to check and send the units on all of them. I have the time and ability, some don\'t, or chose not to take the time. I had one machine a while back returning nothing but junk results, a fellow user pm\'d me and I took it offline for a couple of days and then brought it back online. It now seems to be fine. If that other user hadn\'t contacted me I might not have noticed for quite a while. It was returning units and getting new ones, it was just returning unusable results. Just a thought. ____________ |
|
ID: 4024 | Rating: 0 | rate: / | |
From what I have seen up to now, this is far more than what I expected. Glad to be helping. I set my account for my home machines to run the test applications only to help put more power into testing and leave the other work that can be run by every one for them. I\'ve had a few machines that seem to be more of a problem so I just suspended MC.N on them as they have other projects they can contribute too. It seems like we are doing well enough so I will just leave them alone. I\'m happy either way. |
|
ID: 4039 | Rating: 0 | rate: / | |
2 WU (4492943 and 4492975) timed out within less than 2 hours, and another one eded in 36 seconds (4492945) |
|
ID: 4094 | Rating: 0 | rate: / | |
23-10-2007 18:54:39|malariacontrol.net beta|[error] Can\'t rename output file opt_19_-26_5_9235911_2_0 |
|
ID: 4110 | Rating: 0 | rate: / | |
Keep ketting these error\'s after reinstalling the new java 6.3 I just installed the new Java 6.3 and it works fine. Maybe there is another issue at work here.... ____________ Regards, Bob P. |
|
ID: 4111 | Rating: 0 | rate: / | |
Keep ketting these error\'s after reinstalling the new java 6.3 Something I forgot about.... You also need a BOINC client of version 5.50 or newer, because of the wrapper mechanism that is being used by MCDN. This is just one issue, there could be others too. |
|
ID: 4116 | Rating: 0 | rate: / | |
Fixed the problem by a reboot of the machine.(didn\'t know java needed this!) |
|
ID: 4117 | Rating: 0 | rate: / | |
These optimizers shouldn\'t get any faster: https://malariacontrol.net/workunit.php?wuid=4533448 :-) |
|
ID: 4132 | Rating: 0 | rate: / | |
hi all, |
|
ID: 4326 | Rating: 0 | rate: / | |
hi all, So I am guessing it checks to see if java is installed? If java is not installed, how does the bundled jre run? ____________ |
|
ID: 4331 | Rating: 0 | rate: / | |
So I am guessing it checks to see if java is installed? No, the Java Runtime Environment file will run Java for you without you needing to install Java separately. Sort of a virtual Java for that application only. ____________ Jord. BOINC FAQ Service |
|
ID: 4334 | Rating: 0 | rate: / | |
Dear all, |
|
ID: 4374 | Rating: 0 | rate: / | |
Dear all, What platforms are supported, is it still only Windows XP ? I tried to run on some Windows 98 hosts with not so good luck. They either showed client compute error or as Done with 0.00 completeion time and 0.00 credit but credit is still pending. About 3-4 out of 12 may have completed this way. |
|
ID: 4403 | Rating: 0 | rate: / | |
Hi Keith, I vaguely remember there was an issue with the wrapper approach, because win98 is not fully compatible with the windows api.. create_process() seemed to be a problem.. since we have taken the newest wrapper version of boinc, and we do get valid results from hosts with win98, I assume this issue has been resolved in the new wrapper versions - so no, it should be compatible also with win98. The fact that some workunits terminate very quickly is because they realize early that the result is not going to be meaninful and terminate.. Unfortunately we can only do this at runtime, once the workunits are sent out.. and a wide range of durations is normal with this application, since it\'s a lot about exploring new types of models instead of calculating established ones. Anyway, that\'s a personal opinion, but if I had old computers I think I wouldn\'t let it run on win98, I would take windows xp and tune it to the max (there\'s a lot of tutorials on the web), switch off all the things you don\'t need to run them as boinc client. i once did this with a 400Mhz laptop, and it actually worked ok, not slower than with something like win98, but much more reliable.. but as I said, that\'s my personal taste :) cheers Michael ____________ Michael |
|
ID: 4409 | Rating: 0 | rate: / | |
my first WU resulted in an error: |
|
ID: 4410 | Rating: 0 | rate: / | |
I vaguely remember there was an issue with the wrapper approach, because win98 is not fully compatible with the windows api.. create_process() seemed to be a problem.. since we have taken the newest wrapper version of boinc, and we do get valid results from hosts with win98, I assume this issue has been resolved in the new wrapper versions - so no, it should be compatible also with win98. Is there a minimum memory requirement ? I\'ll check my machines, when I can, and see if I can determine if is just the tasks or if it is machine specific. Most of those I have no choice in the o/s they run. I do not have reason to upgrade them,un-necessary expense, and some of the applications we run are Windows98 only, we do not have the WindowsXP version or there is not one available. Running in an emulation mode causes problems with some of the applications, so we just keep some Windows98 hosts around. We still have some software from 1989 that the boss likes and uses daily. |
|
ID: 4414 | Rating: 0 | rate: / | |
my first WU resulted in an error: I have seen this as well. Mainly when, for some reason, the hard disk is tied up by another application and BOINC can\'t do what it needs to do when a wu finishes. |
|
ID: 4417 | Rating: 0 | rate: / | |
Is there a minimum memory requirement ? re: Windows98 has anyone reached a conclusion to the Win98 question? I can see it\'s worth getting the latest java version first. I\'m just setting up BOINC on an old 600mhz athlon with 384MB and it only runs win98. the small WUs here at malariacontrol are attractive for such a limited system. It\'s maddeningly frustrating to discover the minimum system requirements on many of the BOINC projects. I wish every project had a sticky forum thread in \'number crunching\' with that title - although Rosetta and WCG are explicit enough if you dig around. Sadly Rosetta went down just as I started to install so I haven\'t been able to try and test that one yet! ed: ...ah, I\'ve just seen you can switch on/off each sub-application separately in one\'s own preferences page; if some are known to be ok then hopefully someone can advise which are good and which are doubtful. /pg |
|
ID: 4503 | Rating: 0 | rate: / | |
Can you expalin why some hosts are showing ZERO time and claimed credit. Its not just Windows98. There is another thread started in which this seems to be an issue. It apperas to us that the hosts claiming zero time are all marked valid, but the effect it has is it lowers the claimed credit and possibly the cpu claiming zero is getting too much or cheated out, but there is no way to tell without knowing the actual time. I think the users would like this issue addressed. |
|
ID: 4598 | Rating: 0 | rate: / | |
hi keith and others, first, sorry for the long delay.. I think something with the \"subscribe to this thread\" function does not work as it should, please report somebody else doesn\'t get notified about new posts..will look into this as well. We investigated the issue, and it is solved now. From now on, people who use 0 cpu time do not get any credit, and don\'t pull down the granted credit for others. Since this is a sensitive issue, I would prefer not to explain exactly how it came to that and how we solved it. I can only say that it was most likely not intended by the involved users, but a bug, or say an \"unfortunate\" co-occurence in our software and the boinc software, which together granted too much credit in some cases.. We will now try to see how much credit we have to subtract from - only a handfull of probably unknowing but lucky - users, in order to restore justice .. regards Michael ____________ Michael |
|
ID: 4674 | Rating: 0 | rate: / | |
For me I would like to say THANKS for finding and solving the problem! We users do not always need to know the whys, sometimes it is of a sensitive nature and we just do not always need to know. The important point is that you figured it out and stopped it in the future, again THANK YOU! ____________ |
|
ID: 4683 | Rating: 0 | rate: / | |
hi all, |
|
ID: 5194 | Rating: 0 | rate: / | |
looking forward to new results (am very excited :) and thanks again for crunching So much for that then. 7 hours of No work from project, there was work but not for the applications you have allowed. I\'ve turned the Malariacontrol simulation back on and got fed from that immediately. Will try again later this weekend. ____________ Jord. BOINC FAQ Service |
|
ID: 5196 | Rating: 0 | rate: / | |
looking forward to new results (am very excited :) and thanks again for crunching LOL...yeah I even allowed test applications, but nada. It\'ll be interesting to see when the first wu\'s come through. Live long and BOINC! ____________ Paul (S@H1 8888) |
|
ID: 5197 | Rating: 0 | rate: / | |
I\'m sorry guys, I closed the tap yesterday night, panicking a bit before going for a beer, because I had the impression something was wrong: One of my workunits finished three times but \"without a finished file\", as it said.. Did this happen to anybody else? Please check in your messages whether you keep on crunching the same workunit or not.. Migth be just a problem with my computer, though.. ok, and maybe I have to send out some more workunits in order to be able to ask someone :) thx ____________ Michael |
|
ID: 5205 | Rating: 0 | rate: / | |
I\'m sorry guys, I closed the tap yesterday night, panicking a bit before going for a beer, because I had the impression something was wrong: One of my workunits finished three times but \"without a finished file\", as it said.. Did this happen to anybody else? Please check in your messages whether you keep on crunching the same workunit or not.. Migth be just a problem with my computer, though.. ok, and maybe I have to send out some more workunits in order to be able to ask someone :)thx This is for Optimizer 1.28, right ? I see: host #1 ran 1 task for 39 minutes, finished with success host #2 ran 1 task for 16-3/4 hours and exited with \'resource limit exceeded\' Log shows messages: maximum CPU time exceeded then [error] Can\'t rename output file X task X finished Output file for task X absent host #2 is running a second task which has been running for 1-3/4 hours so far. |
|
ID: 5207 | Rating: 0 | rate: / | |
Please check in your messages whether you keep on crunching the same workunit or not.. That might be a problem with your BOINC version as well. I\'m alpha testing 6.1.8 just to see if I can get it to do those things, but that\'s difficult if all I get is the No work from project message. :-( But I\'ll go back to waiting for an Optimizer. ____________ Jord. BOINC FAQ Service |
|
ID: 5208 | Rating: 0 | rate: / | |
Please check in your messages whether you keep on crunching the same workunit or not.. thanks for the information, I have a wu running for more than 5 hours (2h should be absolute maximum - abort them if they exceed 2h).. so something is wrong. There are at least 2 issues: 1. it doesnt stop after 2h, means I have to go through the code again and make a new application version 2. made a mistake when configuring the \"estimated fpops\" parameter for the workunit..one zero to much.. how stupid..that means your clients get somewhat confused and think they are supercomputers, they get a very high opinion of themselves and think they are 10 times faster than they actually are, coz they finished so much work in no time - which causes them to pick up more work than they actually can do before the deadline (they\'ll realize their mistake in time though and will get back to the ground).. it\'s not too bad for you, I think, just for us a bit, but not much, since we are still testing.. to get out of this, just do reset project..easy to fix on our side I generally think it\'s ok to abort the workunits, since there is a chance you won\'t get credit for it.. sorry about that. Hv to start from scratch again. so there will be no work for some time now until we\'re ready.. I\'ll be looking into this over the weekend sorry for inconveniences Michael ____________ Michael |
|
ID: 5209 | Rating: 0 | rate: / | |
I generally think it\'s ok to abort the workunits, since there is a chance you won\'t get credit for it.. sorry about that. Hv to start from scratch again. Like we\'re here for the credits... I\'m not. (You\'re not either... :-)) But nice to see a developer say \"Hey, look, I messed up. I\'ll be available for a public flogging later this weekend!\".. refreshing. :-) Just holler when the new new new Optimizers are ready to be tested. Do you also tell your boss that he added a zero too many to your check? ;-) ____________ Jord. BOINC FAQ Service |
|
ID: 5210 | Rating: 0 | rate: / | |
I\'m sorry guys, I closed the tap yesterday night, panicking a bit before going for a beer, because I had the impression something was wrong: One of my workunits finished three times but \"without a finished file\", as it said.. Did this happen to anybody else? Please check in your messages whether you keep on crunching the same workunit or not.. Migth be just a problem with my computer, though.. ok, and maybe I have to send out some more workunits in order to be able to ask someone :) I had this task run for 13:46:56 and then received the following messages: 8:55:16 AM Aborting task opt_27_-18_5_662409262_0: exceeded CPU time limit 49609.375000 8:55:24 AM [error] Can\'t rename output file opt_27_-18_5_662409262_0_0 8:55:25 AM Computation for task opt_27_-18_5_662409262_0 finished 8:55:25 AM Output file opt_27_-18_5_662409262_0_0 for task opt_27_-18_5_662409262_0 absent |
|
ID: 5211 | Rating: 0 | rate: / | |
I guess I should go abort those two opt WU\'s that have been running for seven hours then... |
|
ID: 5218 | Rating: 0 | rate: / | |
I guess I should go abort those two opt WU\'s that have been running for seven hours then... abort them.. actually you should abort all of them now.. a new series will come in 1-2 hours.. you will get credit for those where all results of a workunit hv been processed, but there are some where only 1 or 2 per wu were done.. am sorry about that, but I hv to cancel them, since it is not guaranteed that another result of those would succeed. the bug is fixed, and running well in our small-scale testing environment.. please wait for the new series.. regards Michael ____________ Michael |
|
ID: 5223 | Rating: 0 | rate: / | |
update: |
|
ID: 5225 | Rating: 0 | rate: / | |
you will get credit for those where all results of a workunit hv been processed, but there are some where only 1 or 2 per wu were done.. am sorry about that, but I hv to cancel them, since it is not guaranteed that another result of those would succeed. Michael, I think this has had some unintentional consequences. Please see this thread: WUs with errors \"cancelled\" TIA |
|
ID: 5243 | Rating: 0 | rate: / | |
- maximum duration is 2h, anything that runs (much) longer is likely to be flawed, report please.. 10-Feb-08 17:48:51|malariacontrol.net beta|Starting opt_24_-103_5_41959952_2 10-Feb-08 17:48:51|malariacontrol.net beta|Starting task opt_24_-103_5_41959952_2 using optimizer version 129 10-Feb-08 19:49:20|malariacontrol.net beta|Computation for task opt_24_-103_5_41959952_2 finished It runs for 2 hours 29 seconds. Well done. :-) Task ID. Finished well on BOINC 6.1.8 ____________ Jord. BOINC FAQ Service |
|
ID: 5245 | Rating: 0 | rate: / | |
Working well here..... |
|
ID: 5250 | Rating: 0 | rate: / | |
And they also run for shorter times. |
|
ID: 5253 | Rating: 0 | rate: / | |
Run seems to progressing (mostly) OK. Just reported a string of 25 WUs with only 1 bad unit: |
|
ID: 5270 | Rating: 0 | rate: / | |
Developers of the optimizer application might be interested in a problem where the optimizer fails to respond to a \"suspend\" request from BOINC. |
|
ID: 5572 | Rating: 0 | rate: / | |
Had to abort this Opt WU. I just noticed it...it started at 10:41pm last night (after I went to bed) and ran until I aborted it at about 7:15pm tonight. MCN LTD is up to 63,700 now and will just have to work itself off against my other project.
Project Date Message malariacontrol.net beta 3/5/2008 10:41:51 PM Starting opt_44_-519_5_144224089_2 malariacontrol.net beta 3/5/2008 10:41:51 PM Starting task opt_44_-519_5_144224089_2 using optimizer version 130 Einstein@Home 3/6/2008 10:33:56 AM Task h1_0804.45_S5R3__152_S5R3b_1 exited with a DLL initialization error. Einstein@Home 3/6/2008 10:33:56 AM If this happens repeatedly you may need to reboot your computer. malariacontrol.net beta 3/6/2008 10:33:56 AM Task opt_44_-519_5_144224089_2 exited with a DLL initialization error. malariacontrol.net beta 3/6/2008 10:33:56 AM If this happens repeatedly you may need to reboot your computer. malariacontrol.net beta 3/6/2008 10:33:56 AM Restarting task opt_44_-519_5_144224089_2 using optimizer version 130 malariacontrol.net beta 3/6/2008 7:16:55 PM [error] Can\'t rename output file opt_44_-519_5_144224089_2_0 malariacontrol.net beta 3/6/2008 7:17:03 PM Computation for task opt_44_-519_5_144224089_2 finished malariacontrol.net beta 3/6/2008 7:17:03 PM Starting wu_139_415_104113_0_1204726367_2 malariacontrol.net beta 3/6/2008 7:17:03 PM Starting task wu_139_415_104113_0_1204726367_2 using malariacontrol version 556
[/edit] [edit2]corrected starting time[/edit]
|
|
ID: 5627 | Rating: 0 | rate: / | |
Another reported problem with Optimizer application (optimizer_1.32_windows_intelx86 - Entry point not found) here |
|
ID: 5630 | Rating: 0 | rate: / | |
Krunchin-Kieth, myself and others were chatting about Win98 higher up this thread in late Nov/Dec - {here} - so I thought I\'d try switching on the new optimiser 1.33 application. Well four or five tasks went through and failed so its back off again, herewith an example: |
|
ID: 5778 | Rating: 0 | rate: / | |
Hi, createProcess() is a function called in the wrapper application, so this problem must exist for projects using the wrapper approach - hmm, no, wrong, we used the wrapper earlier with the mapping application and it worked on win98 ( except for creating a nasty console window for every task). The things different in this application are only 2, as far as I can think: unzipping of the java runtime environment, and running a java virtual machine..but then why does createProcess() fail? I personally have no idea so far, any suggestions welcome. If you watch the slot directory, do you see a directory called \"jre\" created there? if its not there, that would be an explanation. Any suggestions? ____________ Michael |
|
ID: 5924 | Rating: 0 | rate: / | |
FYI |
|
ID: 5925 | Rating: 0 | rate: / | |
Best post that on the main page, News section as well. If only to reach those people that read by RSS only. |
|
ID: 5926 | Rating: 0 | rate: / | |
Test of new optimizer app version |
|
ID: 6286 | Rating: 0 | rate: / | |
On one computer I got this message: |
|
ID: 6288 | Rating: 0 | rate: / | |
Message boards : Malaria Control : A THIRD science application for malariacontrol