Message boards :
Number crunching :
Host showing 200 units in progress ?
Message board moderation
Author | Message |
---|---|
Merlin.huff Send message Joined: 5 May 02 Posts: 16 Credit: 337,728 RAC: 0 |
One of my computers is showing over 200 work units in progress http://setiweb.ssl.berkeley.edu/show_host_detail.php?hostid=674664 It has never downloaded them and even if it did would never be able to crunch them in time. <img src="http://www.boincstats.com/stats/banner.php?cpid=eb8a66ae028924190ff09df10e79003e" /> |
Compare Watches Send message Joined: 24 Apr 04 Posts: 81 Credit: 2,862,596 RAC: 0 |
Blimey - looks like 9th May is D-Day for you - better leave the machine on eh? Shouldn't think it's anything to worry about - they'll drop off and get sent out again to other machines within a couple of weeks. Which version of the client software are you running out of interest? Tomcat's Taekwondo Pages |
Chilean Send message Joined: 6 Apr 03 Posts: 498 Credit: 3,200,504 RAC: 0 |
> One of my computers is showing over 200 work units in progress > > http://setiweb.ssl.berkeley.edu/show_host_detail.php?hostid=674664 > > It has never downloaded them and even if it did would never be able to crunch > them in time. > It says 441 or around that, cant remember. But that number shows how many how u downloaded and how many have uploaded, after a while those WU will be deleted of that list. |
Paul D. Buck Send message Joined: 19 Jul 00 Posts: 3898 Credit: 1,158,042 RAC: 0 |
Do you have a high number in the preference to "contact the server every 'x' days?" If this number is too high you can download too much work. For most BOINC systems the number should be, in my opinion, no higher than 5 in almost all cases. If you have an always on connection and are connected to more than one project the number should be no higher than 1 day. Most people in this position use 0.1 day to 0.5 days. I use the 0.5 days, have 5 projects running (CPDN, Einstein@Home, LHC@Home, Predictor@Home, and SETI@Home), and for my 7 computers I have 123 work units queued with 15 in-flight at all times. With this quantity of work (most of which, LHC@Home and Predictor@Home runs in about one hour) I have never run dry even during the last "madness" outage when all projects but one were off-line for one reason or another. |
Prognatus Send message Joined: 6 Jul 99 Posts: 1600 Credit: 391,546 RAC: 0 |
|
Merlin.huff Send message Joined: 5 May 02 Posts: 16 Credit: 337,728 RAC: 0 |
I have my settings at 3 days work. The thing is that host has no record of those work units been downloaded, as far as I can tell Seti think I have those units but I have never recived them. If there had been a problem also I would have though the daily quota should have stoped me from downloding that many work units in one go. As asked I am using Boinc Manager 4.25 Also that host is switched on 24/7 and only crunches seti. I monitor it every day and it has never had more than 3 days work. What I was worried about is the 240 odd work units that say "In Progress" <img src="http://www.boincstats.com/stats/banner.php?cpid=eb8a66ae028924190ff09df10e79003e" /> |
jimmyhua Send message Joined: 16 Apr 05 Posts: 97 Credit: 369,588 RAC: 0 |
I checked your computer, and it only has 27 WUs that are in progress? Nevermind I see it now on page 7! That's weird! Jimmy |
Digger Send message Joined: 4 Dec 99 Posts: 614 Credit: 21,053 RAC: 0 |
<blockquote>I have my settings at 3 days work. The thing is that host has no record of those work units been downloaded, as far as I can tell Seti think I have those units but I have never recived them. If there had been a problem also I would have though the daily quota should have stoped me from downloding that many work units in one go.</blockquote> Merlin, I have had that happen to me in the past, where the system 'thinks' it downloaded units to me, but actually did not. A careful inspection of my log files revealed (like yours did) that none of them had even started to download to my machine, so I know it wasn't an error on my side. They were 'assigned' to me I guess, but somehow missed the part where they were supposed to download to me. Mine were only a few units though, and they eventually just timed out and dropped off my results page. With so many of yours going to be listed as 'no reply' all at once though, I wonder if your quota is going to be lowered as a result? Dig |
Bruno G. Olsen & ESEA @ greenholt Send message Joined: 15 May 99 Posts: 875 Credit: 4,386,984 RAC: 0 |
That does seem odd indeed. I didn't go back all pages, but I did notice alot of wu's "sent" at different times during the 25th. I checked my rss reader and there isn't any mention of problems for that date. I thought it might have been some error that deleted them before thay were actually sent when the backup was made, but the backup was made two days later... |
Merlin.huff Send message Joined: 5 May 02 Posts: 16 Credit: 337,728 RAC: 0 |
I do hope my quota is not cut because of this. But I get the feeling it will <img src="http://www.boincstats.com/stats/banner.php?cpid=eb8a66ae028924190ff09df10e79003e" /> |
ThePhantom86 Send message Joined: 30 Jan 02 Posts: 268 Credit: 1,970,082 RAC: 0 |
I've had ghost untis before but never 200+. Every now and then it shows I should have a unit and I don't. I've only had this happen on Seti and Predictor. |
Brian Oliver Send message Joined: 25 May 99 Posts: 50 Credit: 910,871 RAC: 0 |
I repeatedly had ghost units on Predictor, I asked what was going on, bringing it to peoples attention so that it could get fixed or looked into ect. After the 4th or 5th time of this happening and my reporting it i got an e-mail from the Predictor staff telling me that i was full of it and it is NOT possible for such a thing to happen ect ect. And that i oviously did not know what I was talking about (i am a CCIE, MSPT, MCSE So i think i know something about what is going on) and i needed to learn more about computers and read the FAQ's. To say the least this bothered me and i have since chosen not to crunch any more Predictor WU's evan though at the time i was in the top 500 and had crunched over 3% of all WU's sent out by Predictor. I have NEVER had such a Problem with SETI, Ghost units, Rudeness, Poor cutomer service or condesending attitudes. Wiki de BOINC <img src="http://www.boincstats.com/stats/banner.php?cpid=51a3eaaef8df276544f56140a8a65413"> |
Divide Overflow Send message Joined: 3 Apr 99 Posts: 365 Credit: 131,684 RAC: 0 |
Brian, that's too bad about your experience with Predictor. :( I see ghost WU's come across once every other week or so for my home system on Einstein@Home. My work laptop crunches for SETI and I've only see a ghost WU show up very rarely. I'll be glad when this minor bug gets tracked down and fixed, but it's only been a very minor nuisance to me. I'd feel differently if I was getting hundreds show up like Merlin here... |
Brian Oliver Send message Joined: 25 May 99 Posts: 50 Credit: 910,871 RAC: 0 |
That was my point... getting the issue fixed or at least looked into with an acknowalgement that there is or might be an issue. I did NOT expect to be treated so rudely. I was getting these "ghost" units about 30% of the time that I connected and other people were reporting them as well. The other fun thing was that it would randomly download 40-80 WU's (that would show up and could be worked) with a bank of .1 Wiki de BOINC <img src="http://www.boincstats.com/stats/banner.php?cpid=51a3eaaef8df276544f56140a8a65413"> |
Digger Send message Joined: 4 Dec 99 Posts: 614 Credit: 21,053 RAC: 0 |
<blockquote>I do hope my quota is not cut because of this. But I get the feeling it will</blockquote> Merlin, I imagine that if your 'credit rating' here suffers as the result of this issue, one of the Devs could probably get you fixed back up again. I'd be surprised if one of them hasn't already seen this thread and is currently checking into it. Nobody should hold you accountable for 200+ units that never even downloaded to your machine. I do have to say that I like the built-in accountability that Boinc now has. If someone isn't minding the farm, and continually downloads an excessive number of units that either result in errors or don't get returned at all, then they get put on rations until they get their problems fixed. Nice feature. :) Dig |
ABT Chuck P Send message Joined: 15 May 99 Posts: 91 Credit: 316,669 RAC: 0 |
<blockquote>I do hope my quota is not cut because of this. But I get the feeling it will</blockquote> ============= Your daily DL allowable quota will take a hit (it will be limited) but will increase again as you return other completed work units. |
Ingleside Send message Joined: 4 Feb 03 Posts: 1546 Credit: 15,832,022 RAC: 13 |
<blockquote>That was my point... getting the issue fixed or at least looked into with an acknowalgement that there is or might be an issue. I did NOT expect to be treated so rudely. I was getting these "ghost" units about 30% of the time that I connected and other people were reporting them as well. The other fun thing was that it would randomly download 40-80 WU's (that would show up and could be worked) with a bank of .1 </blockquote> v4.25 fixed atleast one bug in parsing of scheduler-reply, but there can still be other bugs hidden away somewhere. |
Ingleside Send message Joined: 4 Feb 03 Posts: 1546 Credit: 15,832,022 RAC: 13 |
<blockquote>I do hope my quota is not cut because of this. But I get the feeling it will</blockquote> Since the quota is doubled each time you reports a "success"-result, even if set to 1 you will be back to normal after 7 results so shouldn't really be a problem except if on dialup. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.