Posts by old_user18325

1) Questions and Answers : Windows : No scheduler responded . . . (Message 9769) Posted 22 Feb 2005 by old_user18325 Post: > Yes, one of the servers is down, and has been since last Thursday or Friday. > There's a thread in the "Number Crunching" forum discussing it. > > So far there has been nothing forthcoming about the cause or the expected fix > date. In fact, no comment from Oxford at all! > > Ian > My BOINC client seems to connect just fine. Is there more than one scheduler for the project?
2) Questions and Answers : Windows : Model just started over for some reason (Message 6048) Posted 11 Nov 2004 by old_user18325 Post: > The CLI client service doesn't write to those files. It used to write > messages to the application event log, but that was disabled because they were > swamping everything else. The CLI client service now silently discards all > BOINC log messages. In any case, you're more likely to get useful information > from the CPDN specific logs in the climatepredition.net/2x82_000157746 > directory. stdout_um.txt was updated at about the same time as the restart: Starting hadsm3 model for ID# 2x82_000157746... Changing to slots directory C:\Program Files\BOINC\slots\1
3) Questions and Answers : Windows : Model just started over for some reason (Message 6029) Posted 11 Nov 2004 by old_user18325 Post: > Go to BOINC folder look at Stderr or Stdout, they may have some info on the > problem, > > Actually, the files have not been written to since 10/31, about when I switched from the GUI to the CLI running as a service. I don't know if the CLI client is supposed to be writing to these files or not.
4) Questions and Answers : Windows : Model just started over for some reason (Message 5976) Posted 8 Nov 2004 by old_user18325 Post: > If the same model restarts the trickles will not be recounted. Once it gets > past the place it was before the trickles will start being counted again. > <br>John Keck -- BOINCing since 2002/12/08 -- <a> href="http://www.boinc.dk/index.php?page=user_statistics&project=cpdn&userid=191"><img> border="0" height="80" src="http://191.cpdn.sig.boinc.dk?188"></a> > > Any idea why this would occur in the first place? There must be a error somewhere for this to occur. Is there someplace I can check to see what that error might be?
5) Questions and Answers : Windows : Model just started over for some reason (Message 5971) Posted 8 Nov 2004 by old_user18325 Post: I have a workunit, 2x82_000157746_0, which has recently started over for some reason. No indications as to why this occurred, but it had sent up the 11th trickle. I run the CLI version of BOINC set up as a service. This is somewhat disconcerting. It has sent up the 1st trickle since restarting, but doesn\'t report anything on the web page for that workunit. I\'m wondering if it is even still functioning properly anymore. I am also running another model on the other processor, so I don\'t just want to detach/reattach, since that would toss the other model, which is running properly. Any thoughts/suggestions would be appreciated.
6) Questions and Answers : Wish list : Pls explain more about the RSS news feed (Message 5122) Posted 7 Oct 2004 by old_user18325 Post: > Checking the home page every few days would accomplish much the same thing. > The stuff you see on the RSS feed are the news items on the page. This info is > only sent as often as it is posted. (which isn't often) If there are several websites that you check every few days for updates, such as the websites of the other BOINC projects, then it may become worth your while to download an RSS feed reader like Pluck, which also sets up RSS feeds from several other sources from news and technology sites as well. However, they can be somewhat slow if you have a large number of feeds for the reader to have to go out and search for periodically.
7) Questions and Answers : Windows : Picking up a new workunit caused the other one to dump (Message 4560) Posted 23 Sep 2004 by old_user18325 Post: Since I have been having so much trouble with the BOINC GUI since Seti is having scheduler trouble, I decided to pick up another CPDN workunit to keep the other processor busy doing something other than maxing out on the 4.09 GUI process that wouldn't even appear on the taskbar. In order to have some control over the operation, I started the CLI version (after, of course, killing the GUI process tree) and performed an update (-update_prefs http://climateprediction.net/) after changing CPDN to 200 so that it would send more work. According to the output on the CMD window, it did indeed pick up another workunit and start to process it. To validate that the BOINC GUI would then behave properly, I Ctrl-Break out of the CLI and started the GUI. The GUI did start to behave at that point, but upon checking the messages, I saw 2 errors: climateprediction.net - 2004-09-23 14:14:27 - Unrecoverable error for result 2ktx_100141526_0 ( - exit code -1 (0xffffffff)) climateprediction.net - 2004-09-23 14:14:27 - Unrecoverable error for result 2x7w_000157740_0 ( - exit code -1 (0xffffffff)) I guess that the workunits were trashed at that point, since the BOINC client downloaded 2 more workunits. So, I guess that the 5 days worth of processing that I did on the first workunit were all for nought. Has anyone else run into this scenario? Could I have done something different to preserve the first workunit? I am running the 4.03 version of CPDN. Here is the pertinent info on my system: Time zone UTC -6 hours Name HOSS Created 15 Sep 2004 21:20:49 UTC Total Credit 302.46 Recent average credit 40.04 CPU type GenuineIntel 993MHz Pentium Number of CPUs 2 Operating System Microsoft Windows 2000 Professional Edition, Service Pack 4, (05.00.2195.00) Memory 511.16 MB Cache 976.56 KB Swap space 990.54 MB Total disk space 16.94 GB Free Disk Space 2.35 GB Measured floating point speed 895.84 million ops/sec Measured integer speed 2207.4 million ops/sec Average upload rate 8.59 KB/sec Average download rate 113.42 KB/sec Number of times client has contacted server 12 Last time contacted server 23 Sep 2004 19:23:29 UTC % of time client is on 98.127 % % of time host is connected 98.127 % % of time user is active 98.126 %