Message boards :
Number crunching :
Pressing F5 causes error in sixtrack screensaver gui.
Message board moderation
Author | Message |
---|---|
![]() Send message Joined: 2 Sep 04 Posts: 378 Credit: 10,765 RAC: 0 |
Sixtrack 4.66 Run the boinc client, select 'view screensaver. when screensaver is running, press f5. Get an error window about unable to unregister class. http://members.shaw.ca/joevoxxer/sixtrack466f5.jpg pressing OK results in not being able to restart screensaver again. Have to exit boinc, restart boinc, and can then rerun screensaver. update: this does not affect the screensaver running in full screen screensaver mode (from using the screensaver preview in the display properties for an xp machine viewing full screen screensaver. I'm not the LHC Alex. Just a number cruncher like everyone else here. |
![]() ![]() Send message Joined: 1 Sep 04 Posts: 157 Credit: 82,604 RAC: 0 ![]() |
Still using Sixtrack 4.64 I did the test with the graphics, not the screensaver. Same problem: show graphics is not working after that. Win XP Pro SP2. |
![]() ![]() Send message Joined: 23 Oct 04 Posts: 358 Credit: 1,439,205 RAC: 0 |
> Still using Sixtrack 4.64 > > I did the test with the graphics, not the screensaver. > So did I; the screensaver itself gave the error-message 'that it couldn't start the appl. graphic' with the bouncing BOINC Icon, and restart this behaviour every 5 sec.. After a restart of the Client the WU was shown as 0% complete and 0 CPU time (but it had , for sure, 60 minutes crunched before.), when I then activated the 'show graphic' the state switched to 100% and calculation error. The result I'm talking about: WU ID: 84862, result ID with Exit-status: 1073807364 (0x40010004) ? I don't test scrensaver on this host again; 'show graphics' works! (WinXP Pro SP2, ATI X800 XT) greetz littleBouncer ![]() ![]() ![]() |
![]() ![]() Send message Joined: 30 Sep 04 Posts: 112 Credit: 104,059 RAC: 0 |
Sounds like an important bug has been found... Now let's see how important it is to the project developers and how high of a priority they place on fixing it. Good job to those who discovered this....Guess that's why LHC is still considered a BETA project eh....? :) ![]() ![]() |
![]() Send message Joined: 2 Sep 04 Posts: 378 Credit: 10,765 RAC: 0 |
> Sounds like an important bug has been found... Now let's see how important it > is to the project developers and how high of a priority they place on fixing > it. > > Good job to those who discovered this....Guess that's why LHC is still > considered a BETA project eh....? > > :) > I've seen worse bugs. Not like I've found a nuke button which makes an entire system self destruct. ;) Previous job I had, I made it a challenge to kill an embedded operating system in as few keystrokes as possible. I got it down to typing two words in a diagnostic terminal and pressing enter and the system would seize up completely, and crash the kernel. I'm not the LHC Alex. Just a number cruncher like everyone else here. |
![]() Send message Joined: 3 Sep 04 Posts: 212 Credit: 4,545 RAC: 0 |
> Sixtrack 4.66 > Run the boinc client, select 'view screensaver. > when screensaver is running, press f5. > > Get an error window about unable to unregister class. > http://members.shaw.ca/joevoxxer/sixtrack466f5.jpg > > pressing OK results in not being able to restart screensaver again. > > Have to exit boinc, restart boinc, and can then rerun screensaver. > > update: this does not affect the screensaver running in full screen > screensaver mode (from using the screensaver preview in the display properties > for an xp machine viewing full screen screensaver. > > Thanks for information. We'll try to fix this, but our graphics developer is not currently working for us so it might take some time... Markku Degerholm LHC@home admin |
![]() Send message Joined: 27 Jul 04 Posts: 182 Credit: 1,880 RAC: 0 |
I have a test version on the alpha server were F5 at least doesn't crash the client. It does't do anything at all. But that is better than crashing. Chrulle Research Assistant & Ex-LHC@home developer Niels Bohr Institute |
©2025 CERN