Message boards :
Number crunching :
Serious BOINC Problem????
Message board moderation
Author | Message |
---|---|
Cruncher-American Send message Joined: 25 Mar 02 Posts: 1513 Credit: 370,893,186 RAC: 340 |
In both 7.2.28 and 7.0.64, I noticed the following in the Projects tab: I run SETI, share 100 with Einstein, share 0, as a backup. Recently I changed Einstein in both to "No New Tasks". When I now click on SETI, the left-hand column changes to the one for SETI. HOWEVER, it says NNT, which I did not set for SETI. But SETI is asking for new tasks. If I reset SETI to "Allow New Tasks", Einstein is also changed to Allow in the display when I click on Einstein. That is not correct. Furthermore, Einstein IS asking for tasks now when it shouldn't, since I never changed it from NNT. What the heck is going on? Did I miss something? |
David S Send message Joined: 4 Oct 99 Posts: 18352 Credit: 27,761,924 RAC: 12 |
In both 7.2.28 and 7.0.64, I noticed the following in the Projects tab: Incredibly obvious first: are you sure they aren't both highlighted in the main pane of that tab? Sometimes things in Boinc Manager are highlighted without looking like they are. David Sitting on my butt while others boldly go, Waiting for a message from a small furry creature from Alpha Centauri. |
Cruncher-American Send message Joined: 25 Mar 02 Posts: 1513 Credit: 370,893,186 RAC: 340 |
Incredibly obvious first: are you sure they aren't both highlighted in the main pane of that tab? Sometimes things in Boinc Manager are highlighted without looking like they are. Alas, that is NOT the case. With either one highlighted alone (I clicked in the project area, but not on a project, to turn off the project-specific fields and gray out the ones at the top left) than picked one. They both say NNT yet Seti is merrily requesting new tasks and filling his queues back up. |
bluestar Send message Joined: 5 Sep 12 Posts: 7031 Credit: 2,084,789 RAC: 3 |
With two of my four BOINC projects set to NNT, I may of course select more than one project my means of CTRL- or SHIFT-CTRL, or just clicking one or more projects while holding down the CTRL- key at the same time. If a project is set to NNT, selecting one project which still allows tasks together with another which is set to NNT gives different options than when both projects are set to NNT or are allowed to receive new tasks. I have Seti@home set to allow new tasks, but I am also busy doing a couple of other things here as well right now. So perhaps it may need checking whether a project is not set to NNT before attempting to download more tasks. |
Cruncher-American Send message Joined: 25 Mar 02 Posts: 1513 Credit: 370,893,186 RAC: 340 |
What does that have to do with what I asked? I have TWO projects, one of which is set to NNT (Einstein) and one which IS NOT (SETI). No matter which I select on the Project tab, they both show NNT. That is the apparent bug I was questioning/reporting. |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
What does that have to do with what I asked? I have TWO projects, one of which is set to NNT (Einstein) and one which IS NOT (SETI). No matter which I select on the Project tab, they both show NNT. That is the apparent bug I was questioning/reporting. How did you set Einstein to NNT, if not by using the controls on the Projects tab? I've reproduced what you claim to have done. If I highlight the Einstein line in the list (where the status says "Won't get new tasks"), the command button says 'Allow new tasks'. If I highlight SETI (where the status is blank), the command button says 'No new tasks'. If I highlight both projects at once, the command button is disabled (greyed out), because it can't perform two contradictory actions at once. OK, so I'm using BOINC v7.4.36 to test that, but the v6.12.34 machine beside it behaves the same way - and I've not noticed any changes as I've run through the intermediate builds between those two. |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
When I now click on SETI, the left-hand column changes to the one for SETI. HOWEVER, it says NNT, which I did not set for SETI. But SETI is asking for new tasks. The button shows the option you have, not the state that it is in. When you see "No new tasks" it's in the state of "Allow new tasks". Just as that you will see "Suspend", because it is in the state of "Resume". So it is normal that you will have Seti ask for work when it shows the "NNT" button. If I reset SETI to "Allow New Tasks", Einstein is also changed to Allow in the display when I click on Einstein. That is not correct. I suspect what is happening is what David is saying, despite you saying it isn't. In 7.0 BOINC still coloured the whole field a uniform value, I think all blue. Possibly that 7.2 did that as well. At least 7.4 has the alternating white and brown (or light grey) bars back. When your Windows colour scheme matches the background colour of these 7.0 and 7.2 fields, it's difficult to see if you selected just one or more than one project. The easiest way to test is to exit BOINC & restart it. Then choose one project only and do the changes. |
Jim_S Send message Joined: 23 Feb 00 Posts: 4705 Credit: 64,560,357 RAC: 31 |
In both 7.2.28 and 7.0.64, I noticed the following in the Projects tab: I had the same problem on my Intel box after the outage...Didn't affect My AMD box though. I Desire Peace and Justice, Jim Scott (Mod-Ret.) |
Cruncher-American Send message Joined: 25 Mar 02 Posts: 1513 Credit: 370,893,186 RAC: 340 |
I had the same problem on my Intel box after the outage...Didn't affect My AMD box though. Now that is interesting - my boxes are NVIDIA based. |
Jim_S Send message Joined: 23 Feb 00 Posts: 4705 Credit: 64,560,357 RAC: 31 |
I had the same problem on my Intel box after the outage...Didn't affect My AMD box though. So is the Intel box that did it (GTX750). I Desire Peace and Justice, Jim Scott (Mod-Ret.) |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
I can't get my head round any logical link between the manufacturer of the CPU or GPU, and the selection/command operation of the BOINC Manager GUI. But I'll throw in another random observation. Over the last few weeks (possibly a couple of months), I've been noticing that machines with a USB optical mouse have started registering a doubleclick when I'm quite sure I only clicked the mouse button once. That can wreak havoc with toggle command buttons, like BOINC uses. Either I'm developing Parkinson's without noticing any other symptoms, or somebody has slipped a HID or USB driver update onto the systems without me noticing. |
Jim_S Send message Joined: 23 Feb 00 Posts: 4705 Credit: 64,560,357 RAC: 31 |
I thought it was only My Mouse getting Flaky. |
rob smith Send message Joined: 7 Mar 03 Posts: 22199 Credit: 416,307,556 RAC: 380 |
... and I thought it was only me :-( Must have been one of MS's "updates" - designed to p folks off so they waste money on a newer version of Windoze Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Jeff Buck Send message Joined: 11 Feb 00 Posts: 1441 Credit: 148,764,870 RAC: 0 |
But I'll throw in another random observation. Over the last few weeks (possibly a couple of months), I've been noticing that machines with a USB optical mouse have started registering a doubleclick when I'm quite sure I only clicked the mouse button once. That can wreak havoc with toggle command buttons, like BOINC uses. Verrrrry interesting! Perhaps the mice are starting to rise up against us. I started having the exact same problem on my daily driver (Windows Vista) about 4-5 months ago, but with a totally opposite type of mouse, an old Compaq PS2 ball mouse. Most annoying was when I'd click the close button on one window and the phantom 2nd click would also close the window that was directly beneath it. Initially I thought the problem lay with the KVM switch I was using (which was always a little flaky), but the problem didn't go away when the switch was retired and the mouse was plugged in directly. Thorough cleaning and scraping of contacts helped briefly, but not completely and not for long. I finally discovered last month that I had a spare USB optical mouse laying in a drawer, plugged it in, and the problem seems to have gone away. |
OzzFan Send message Joined: 9 Apr 02 Posts: 15691 Credit: 84,761,841 RAC: 28 |
I don't even think it is something recently pushed out by Microsoft. I was having mouse issues while playing with an old Pentium III 1GHz running Windows 2000 SP4, certainly too old to be receiving any updates to it at all. Ended up having to press the hardware reset button because I couldn't even get the mouse to cooperate and go to the Start button to initiate a reboot. |
Wedge009 Send message Joined: 3 Apr 99 Posts: 451 Credit: 431,396,357 RAC: 553 |
For all of Microsoft's faults, I should point out to the conspiracy theorists that these observations are mostly likely due to the fault of the mice. Being mechanical devices, they do have a tendency to fail after some time. Mechanical switches inherently have some 'bounce' do them - the electrical contacts don't cleanly switch from off to on, or vice-versa, but 'bounce' back and forth between off and on for a split second. There are methods to mechanically and electronically 'debounce' the switch but it's not hard to imagine that these methods can break down after time, causing the double-click-on-single-click phenomenon and other mouse-click faults. I've experienced this myself with an older mouse recently, as have many of you. This might be common knowledge to some, but for once I don't think we should be pinning this one on Microsoft. Soli Deo Gloria |
Aurora Borealis Send message Joined: 14 Jan 01 Posts: 3075 Credit: 5,631,463 RAC: 0 |
For all of Microsoft's faults, I should point out to the conspiracy theorists that these observations are mostly likely due to the fault of the mice. Being mechanical devices, they do have a tendency to fail after some time. Mechanical switches inherently have some 'bounce' do them - the electrical contacts don't cleanly switch from off to on, or vice-versa, but 'bounce' back and forth between off and on for a split second. There are methods to mechanically and electronically 'debounce' the switch but it's not hard to imagine that these methods can break down after time, causing the double-click-on-single-click phenomenon and other mouse-click faults. I've experienced this myself with an older mouse recently, as have many of you. As anyone gone into the mouse properties and tried to adjust the double click sensitivity? |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
Most of the time it works normally (as expected), with single and double clicks performing their respective roles. I first experienced the problem with my youngest mouse, a Dell OEM branding about a year old, which seemed an unusual design: the "buttons" are flexible extensions of the mouse body moulding, rather than being discrete hinged parts. I thought that was a likely point of failure, tried the mouse on a different machine, observed that it misbehaved there too, and retired it - haven't used it since. But other mice in my nest have started misbehaving too. |
Jeff Buck Send message Joined: 11 Feb 00 Posts: 1441 Credit: 148,764,870 RAC: 0 |
I think this Wiki might explain what's going on. |
Mark Wyzenbeek Send message Joined: 28 Jun 99 Posts: 134 Credit: 6,203,079 RAC: 0 |
But I'll throw in another random observation. Over the last few weeks (possibly a couple of months), I've been noticing that machines with a USB optical mouse have started registering a doubleclick when I'm quite sure I only clicked the mouse button once. That can wreak havoc with toggle command buttons, like BOINC uses. My mouse developed the same problem. I replaced it. The Universe is not only stranger than you imagine, it's stranger than you can imagine. SETI@home classic workunits 1,405 CPU time 57,318 hours |
©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.