Posts by old_user12561

1) Message boards : Number crunching : She don\'t like my AMDs (Message 34829) Posted 28 Aug 2008 by old_user12561 Post: I think whoever said that the HDAM3 doesn\'t run well on AMD64 was correct... the last two crashed at exactly 36 mins.... Anything else seems to run fine on her. I was just alarmed at the last message thinking I was in for another crash. Thankfully it is running quite smoothly as of now. HAve no idea of what happened with the old computer except that every WU crashed on it. Hadn\'t had a similar problem on any other project. The slab seems to be ticking along. Thanks a lot!!!
2) Message boards : Number crunching : She don\'t like my AMDs (Message 34823) Posted 28 Aug 2008 by old_user12561 Post: Thanks for your response. No, I did not do a forum search and have gone over the readme so many times to try and figure out exactly why my AMD machines keeps crashing WUs, I can\'t begin to count. That must have been something I missed. Since I had already started this thread I figured this would be a good place to ask.
3) Message boards : Number crunching : She don\'t like my AMDs (Message 34820) Posted 28 Aug 2008 by old_user12561 Post: Ok here is the latest message for the Slab model I am running on the AMD 8/28/2008 12:38:08 PM\|climateprediction.net\|Task hadsm3fub_jxym_005962145_0 exited with zero status but no \'finished\' file What the heck? Should I just give up? I use this comp for work and actually take it offline about 4 hours a day to LINUX live. I always take it off after a save point and shut down properly.
4) Message boards : Number crunching : She don\'t like my AMDs (Message 34804) Posted 27 Aug 2008 by old_user12561 Post: I am running 6.2 Boinc for Windows for the 32 but system but the HDAM still crashes.... have two slabs running on it now but not happy.....
5) Message boards : Number crunching : HELP!!! - Numerous trickles at end of phase (Message 34802) Posted 27 Aug 2008 by old_user12561 Post: Thanks for the info. It just kept sending trickles every 5 secs or so with a sucessfull message. I guess I will have a better understanding of it next time. Thought it was stuck in a loop at 66 percent. Once restarted it started progressing again.
6) Message boards : Number crunching : CPU clock and percentage not counting, but trickle been sent (Message 34799) Posted 27 Aug 2008 by old_user12561 Post: Funny that this is the same point at which mine was stuck (66.666%) and sending trickles every five seconds according to BOINC. Restarted the BOINC app and she seems happy.
7) Message boards : Number crunching : HELP!!! - Numerous trickles at end of phase (Message 34797) Posted 27 Aug 2008 by old_user12561 Post: It has completed phase one... no sign of the phase two completion on my account page, though the percent complete has remained the same at 67%. Hopefully all is well. I would kill it if that happened. I did snooze the project before shutting down but by the looks of the messages it was sending a trickle request ever 5 seconds or so for 4 hours with no computation going on.
8) Message boards : Number crunching : HELP!!! - Numerous trickles at end of phase (Message 34793) Posted 27 Aug 2008 by old_user12561 Post: NVM have no idea what that was about except maybe getting to the end of the phase. Opened my comp and nothing but \'trickle up\' for miles and miles every 5 secs... Restarted BOINC and it uploaded a zip which seemed to have taken care of the problem.. Wasn\'t even crunching just trickling.
9) Message boards : Number crunching : HELP!!! - Numerous trickles at end of phase (Message 34792) Posted 27 Aug 2008 by old_user12561 Post: This task http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=7582132 has suddenly started sending trickle up messages every 30 secs..... WTH? Make that every 5 secs
10) Message boards : Number crunching : Running other projects with CPN on dual Core (Message 34666) Posted 17 Aug 2008 by old_user12561 Post: It\'s not \'Rosie\', it\'s your BOINC. It\'s got more work than it can handle, and won\'t allow anything else to download. It will stay like that for days, or even weeks, and then, when it\'s worked out a \'work schedule\' to include both climate models and other projects, it\'ll suddenly realise that things aren\'t as bad as it first thought, and allow work from other projects. This is normal for computers running other projects that suddenly find a climate model added to the mix. It also seems that you haven\'t altered your preference settings in your climate project account to select the type of model that\'s sent to you. If you do this, you can get shorter models, so that other projects don\'t suffer. And if you start turning the climate models on and off so as to allow work from other projects, you\'ll just prolong the time that it takes BOINC to create its \'work schedule\'. PS Don\'t get lulled into thinking that the climate models aren\'t needed for 3 years. That\'s just to fool BOINC into not panicing. The models are required ASAP. Leave some of them too long, and the research for that model type may well be over and done with before you finish the model. Actually I HAVE modified my preferences getting after the downlods from 4 to a 10 day buffer. Actually never expected to get two climate models with a project already running and a four day buffer. Had no intention of monkeying with the comp to \'prolong or shorten any runtime. It was just a question. You answered it in your first few sentences. The preference you see on my comp page is an experiment as I was having trouble running the HADAM3 on the AMD so I deselected all. My orignal preference was for those type models ONLY.
11) Message boards : Number crunching : Running other projects with CPN on dual Core (Message 34662) Posted 16 Aug 2008 by old_user12561 Post: It depends upon the settings in Your Account. Well each product is evenly meeted out CPU time. I have requested up to 10 days of work from Rosetta. Down to the last 5 2.5 hr tasks and still getting no new work. I am runnin a Couple model with a ridiculous completion time and CPN downloaded a DAM3 before I could hit \'no new tasks\'....9-). Question is... with the high completion time is Rosie thinking I can\'t handle anymore work even though the completion date is some 3 years from now? Ooops forgot, I don\'t think its an issue of space since projects are limited to 57GB of space on this harddrive unless of course it has anything to do with the memory required by the HDAM3 though there is 3GB of memory installed.
12) Message boards : Number crunching : Running other projects with CPN on dual Core (Message 34657) Posted 16 Aug 2008 by old_user12561 Post: While trying to switch out comps to assign certain projects to certain comps, I realized that no project get new tasks. Does the completion time on the CPN preclude other tasks from downloading until its crunched down a bit?
13) Message boards : Number crunching : She don\'t like my AMDs (Message 34649) Posted 15 Aug 2008 by old_user12561 Post: Thanks for all the replies. To the person who I bet it was Vista. I dont think Vista was out in 04 when I started this project. I am running them happily on Vista on two intels, the AMDs just don\'t like it. The one I attempted to run on this year has 4GB RAM, 1024 dual L2, and two 3000+ processors. Nothing that CPN could throw at me should be too big for the machine.In 04 it was XP SP1. Again thanks all. I decided to just crunch CPN on the faster dual intel (though not as fast as the AMD) Rosetta on the other intel and keep SETI on the AMD. Though I would really like to dedicate the power of the AMD machine to CPN since the WUs ARE so big. Forgot I AM running 32 bit Vista on a 64bit capable processor... thinking about upgrading to 64 bit Vista, one reason I just stuck 4GB of memory in her. Forogt I am running boinc 6.2 for 32 bit Vista as well.
14) Message boards : Number crunching : She don\'t like my AMDs (Message 34648) Posted 15 Aug 2008 by old_user12561 Post: Thanks for all the replies. To the person who I bet it was Vista. I dont think Vista was out in 04 when I started this project. I am running them happily on Vista on two intels, the AMDs just don\'t like it. The one I attempted to run on this year has 4GB RAM, 1024 dual L2, and two 3000+ processors. Nothing that CPN could throw at me should be too big for the machine.In 04 it was XP SP1. Again thanks all. I decided to just crunch CPN on the faster dual intel (though not as fast as the AMD) Rosetta on the other intel and keep SETI on the AMD. Though I would really like to dedicate the power of the AMD machine to CPN since the WUs ARE so big. Forgot I AM running 32 bit Vista on a 64bit capable processor... thinking about upgrading to 64 bit Vista, one reason I just stuck 4GB of memory in her.
15) Message boards : Number crunching : She don\'t like my AMDs (Message 34620) Posted 13 Aug 2008 by old_user12561 Post: Ok I started this project 4 years ago and switched to an AMD... immediately got nothing but comp errors. Asked about this here and was told everything from the machine was overclocked (which I wouldn\'t even begin to know how to do) even though it was straight out of the box. Fast forward 4 yrs. Dual core AMD 3000+... again...nothing but comp errors and this is the only project I am having this issue on. Runs fine on my dual intels.
16) Message boards : Number crunching : Boinc version 6 released - comments thread (Message 34619) Posted 13 Aug 2008 by old_user12561 Post: How do I get the thing to quit booting with the BIOS without completely uninstalling it? IF I have to uinstall is there a way to save my work? I really don\'t like the idea of not being able to disable ANYTHING on my comp.
17) Questions and Answers : Windows : upgrade to boinc 6.2.16 (Message 34618) Posted 13 Aug 2008 by old_user12561 Post: I upgraded on my comp... the problem is that I use the comp for work as well and I boot to LINUX from a disk bypassing the windows operating system when I work. Can you say SNAFU? YIKES. Could not log onto work because the default automatically runs it from the BOOT and once I switched the comp back to windows every WU in process immediately went to comp error. I dont like the idea of amything defaulting on my comp where you can\'t turn it off and on at will and unless you catch it before you install it, you can\'t. You wind up with every message ever sent on the thing taking up disk space. Close it out and reopen and three days worth of messages are still there.... that is short of shutting your whole system down. There are times when I DON\'t want BOINC to run as in when I am working in windows and need my full 3GB of memory. Anyone know of a way to get the thing out of BIOS with running out your tasks and uninstalling it?
18) Questions and Answers : Windows : Client error (Message 5729) Posted 28 Oct 2004 by old_user12561 Post: > Margo B: > > Generally what you wrote in your other message about Daily quota exceeded is a > message that you receive as you computer has tried to download over 2 models > in one day. Generally this indicates that the previous 2 models that it > downloaded have crashed and you would have seen in the Boinc "Messages" tab > some information about an error code. From the above it appears that the > computer keeps crashing the models daily. > > I am not familiar with the text that you have pasted in so I cannot comment > further though it does seem that your computer was running models > successfully. Have you made any changes to your system on or just before Oct > 6? > > Another thought, do you have the option to "Leave applications in memory while > preempted" set to Yes? I think I read somewhere that have a "No" to this > could have negative results. > > You might want to try looking at some of the stickies and other information on > the Climateprediction.net discussion forum at: > http://www.climateprediction.net/board/index.php > > Which is accessable from the "Message Boards" menu item on the left hand side. > There could be some systems stability or other issues that could be looked > into. > > DaveN > Thanks Dave but Im inclined to believe that I have a bad WU. I had not had this problem before and this is about the time a new WU downloaded. Further the text that you see pposted is the Reults tab from this website. I read it to understand that this is the reply when you results are sent to be computed. I don't know what the problem is but I still have only one work unit that is running no work and showing 66% complete. I was wondering if this could have soemthing to do with reattaching to SETI, another BIONIC program I am running. I have been receiving credits from the server until sometime in the middle of last night. The application is defaulted to remove from memory when complete. Guess I'll have to report a bug to get this straightened out.
19) Questions and Answers : Windows : Client error (Message 5725) Posted 28 Oct 2004 by old_user12561 Post: My last post was about receiving daily quota exceeded messages from server. After exploring a little further I realize that all my results since 10/8 have returned the following client errors for computing. Is this my problem or CP\'s: Results for user Result ID click for details Work unit ID click for details Sent Received Server state explain Outcome explain Client state explain CPU time (sec) granted credit 363391 311313 28 Oct 2004 0:06:23 UTC 28 Oct 2004 0:08:18 UTC Over Client error Computing 17.47 0.00 363386 311308 28 Oct 2004 0:03:52 UTC 28 Oct 2004 0:06:23 UTC Over Client error Computing 1.36 0.00 363383 311305 28 Oct 2004 0:02:23 UTC 28 Oct 2004 0:03:45 UTC Over Client error Computing 2.30 0.00 363379 311301 28 Oct 2004 0:01:03 UTC 28 Oct 2004 0:02:23 UTC Over Client error Computing 2.16 0.00 363378 311300 27 Oct 2004 23:59:44 UTC 28 Oct 2004 0:01:03 UTC Over Client error Computing 1.47 0.00 363376 311298 27 Oct 2004 23:58:16 UTC 27 Oct 2004 23:59:44 UTC Over Client error Computing 0.75 0.00 363374 311296 27 Oct 2004 23:53:54 UTC 27 Oct 2004 23:58:16 UTC Over Client error Computing 1.23 0.00 281577 256897 6 Oct 2004 9:40:04 UTC 27 Oct 2004 23:53:54 UTC Over Client error Computing 1161434.45 2948.95 141785 142058 4 Sep 2004 22:03:06 UTC 8 Oct 2004 9:30:35 UTC Over Success Done 2147619.55 5444.21 -------------------------------------------------------------------------------- Copyright Â© 2002-2004 climateprediction.net
20) Questions and Answers : Windows : Daily quota exceeded (Message 5721) Posted 28 Oct 2004 by old_user12561 Post: I\'m running version 4.13 on Windows XP and keep getting a message that says \'No work available, daily quota exceeded\'. I thought this dowloaded the entire workunit to you comp at one time. I\'m showing 66% complete under projects but nothing under work for CP.