As a charter member of the "relative handfull", (since December 29th), I want to thank the Einstein staff for their work with the MDQ. The practice I got in the running dry, detaching and reattaching to avoid the dead box club will not be missed. Yes, tweakster is warm and happy. My house is warm and happy. Our mild winter (so far) here in the midwest has meant that my air conditioning has come on briefly at times to maintain my 78 degree environment. My Einstein farm is literally heating my home. For those of you who want to tinker with the screen saver to deliver messages of doom, just remember, for the pro's, it's the first thing we disable.
Regards-tweakster
If you read this a little closer, it's usually the ones that use the screensaver that are causing problems. It also was suggested to do a POP-UP message first, then an afterthought to add it to the screensaver.
I personally use the "BLANK" screensaver, so it would never notify me in that. A POP-UP from the software itself would only be the way.
Since each project is different, and has different thresholds in work allowed, the programming should come from the project software, and relay it through the BOINC software.
tweakster:
It is probably a moot point now but I found a different way around the MDQ problem. I bumped up my 'connect to network about every' time. This will not help if you are already in the middle of a siege of 1 hour nasties. Over time I collected enough extra work to ride out one such a siege (turned in 76 results with 4 cpus while the quota was still 16). Three day connect time ultimately collected enough WU to extra WUs almost ride out my worst case scenario of 144 ultra-short WUs in a day.
As a charter member of the "relative handfull", (since December 29th), I want to thank the Einstein staff for their work with the MDQ. The practice I got in the running dry, detaching and reattaching to avoid the dead box club will not be missed. Yes, tweakster is warm and happy. My house is warm and happy. Our mild winter (so far) here in the midwest has meant that my air conditioning has come on briefly at times to maintain my 78 degree environment. My Einstein farm is literally heating my home. For those of you who want to tinker with the screen saver to deliver messages of doom, just remember, for the pro's, it's the first thing we disable.
Regards-tweakster
Happy to see your RAC numbers climbing.
Also happy not to be getting your electric bill.
MarkF and Pooh; Seems to me we need a separate forum for the "relative handfull". I disable any screen saver and set the rig to turn off the monitor after 2 minutes. My computers are not considered part of the decor, nor do I pump amps into them to serve as a fascinating night light as might the CompUSSR customers. How about a forum addition that is restricted to the top 1000, 500,... whatever crunchers to give some motivation to our fellow travelers to upgrade and do some ka crunching to get there? Back in December when I got hit with the 1 hour nasties, it was "who ya gonna call". I figure if they can share our dialogs they just might learn something.
Bruce; Thank you for the regards and concern about my commitment to my utility provider. I suffered with S__i for some three years. I have a great collection of modems. My motivation with Einstein is to find out how good you guys can get. The 24 divided by 1 equals 16 turns out to be just a bump in the road thanks to your attention. I am willing to turn up the heat if your record of "full service server support" keeps my rigs hot (even in July).
If there were a way for the software to automatically send a POP-UP message to a user who has returned a set number of BAD results, and keep popping up as each bad one goes in, and if it does get down to minimial, after another set number it should totally stop, and keep popping messages up, telling them of the situation.
Alas, in a perfect world.
Pop ups can be a disaster in DC programming, if they destroy the ability to leave the progam running unattended. Any pop-up, even for an error condition, should be one that does not require a user response to clear it.
This is to make it possible to sort out the problem via a remotely connected manager (which could be the official boinc_manager running remotely, or could be an independent one like BoincView)
Secondly, some clients run on boxes with no GUI support at all - a pop up box based on the assumption that the GUI is there can then cause more problems and mislead the user into thinking that the software needs a GUI for normal operation.
The Predictor app has a pop-up in it that causes both these issues to arise - we do not need any more of these.
My preference would be to allow the quota to drop to zero, email the user whenever it drops from one to zero, and provide a link in that email and on their account page that they can use to bump the quota to one when they think they have fixed the fault. This would allow many attempts to fix the fault to be made on the same day, but with a user reset between each failed attempt an unattended rogue box would not eat too many WU
This change would have to come from the BOINC team rather than the Einstein folks.
hehe, I have screen saver set to none. I don't need a screen saver :D I don't even use the power saving, send monitor to standby... When I'll be away from the comp for awhile, I just turn the monitor off. As to having it go in standby, 2 mins or so, nah... That can be a tad annoying if one's reading something, and well... Or watching a TV show for that matter, and em TV shows can last for awhile. Watching a DVD could present the same issue if it's wanting input, especially if it's something like Lord of the Rings: extended version :D
Quote:
The Predictor app has a pop-up in it that causes both these issues to arise - we do not need any more of these.
My preference would be to allow the quota to drop to zero, email the user whenever it drops from one to zero, and provide a link in that email and on their account page that they can use to bump the quota to one when they think they have fixed the fault. This would allow many attempts to fix the fault to be made on the same day, but with a user reset between each failed attempt an unattended rogue box would not eat too many WU
This change would have to come from the BOINC team rather than the Einstein folks.
River~~
I've seen that popup in Predictor before... To make a long story short, it was when the science app saw an update, that dealt with where certain parameters were contained. This was, as I remember because some dialup users didn't like downloading the same file with each WU, so it was hard coded into the app, and then referenced by the WU sometime back.
However, when the first WUs came out using that (which promptly needed to be upgraded again), people started getting this particular Fortran error, and practically everyone was getting the same one. Was a coding bug, related to the upgraded app, when first implementing this change...
Perfectly understandable from a programming standpoint (as moving the data from the WUs, to being hard coded into the app itself isn't a small change); though also a tad bad, when I came home from a full day at school or something, saw the error on my monitor, and had lost possibly up to 10 hours of crunch time or so, as BOINC wouldn't schedule anything else to run until a user responce was given to that dialog... Course, beyond lost crunch time, when nothing could schedule (prior to the sorting of the fortran errors out), was wonder about how this could effect our daily quota as the real problem was bad WUs that were effecting everyone, and not a hardware issue with these people's comps...
I've seen that popup in Predictor before... To make a long story short, it was when the science app saw an update, that dealt with where certain parameters were contained. This was, as I remember because some dialup users didn't like downloading the same file with each WU, so it was hard coded into the app, and then referenced by the WU sometime back.
Yes, well, based on my data which I have just looked at again, the version of the application that had those problems was 4.28 ... guess which version is still in use?
If you guessed 4.28 you get the lead balloon.
But, focus stealing is bad, pop-ups should almost never be used. And I am not convinced that there is a need for them with BOINC. If the point is to have it run as an unobtrusive application then this is not a good idea.
If the operator configures themselves into a corner I have little sympathy. But, we should not incraese the risk that we lock up someone's machine. I have even seen cases where the use of a pop-up causes a problem where the person cannot acknowledge the pop-up because something else stole the focus but you can't get the focus back because the pop-up needs to be acknowledged first ...
RE: As a charter member of
)
If you read this a little closer, it's usually the ones that use the screensaver that are causing problems. It also was suggested to do a POP-UP message first, then an afterthought to add it to the screensaver.
I personally use the "BLANK" screensaver, so it would never notify me in that. A POP-UP from the software itself would only be the way.
Since each project is different, and has different thresholds in work allowed, the programming should come from the project software, and relay it through the BOINC software.
tweakster: It is probably a
)
tweakster:
It is probably a moot point now but I found a different way around the MDQ problem. I bumped up my 'connect to network about every' time. This will not help if you are already in the middle of a siege of 1 hour nasties. Over time I collected enough extra work to ride out one such a siege (turned in 76 results with 4 cpus while the quota was still 16). Three day connect time ultimately collected enough WU to extra WUs almost ride out my worst case scenario of 144 ultra-short WUs in a day.
RE: As a charter member of
)
Happy to see your RAC numbers climbing.
Also happy not to be getting your electric bill.
Bruce
Director, Einstein@Home
MarkF and Pooh; Seems to me
)
MarkF and Pooh; Seems to me we need a separate forum for the "relative handfull". I disable any screen saver and set the rig to turn off the monitor after 2 minutes. My computers are not considered part of the decor, nor do I pump amps into them to serve as a fascinating night light as might the CompUSSR customers. How about a forum addition that is restricted to the top 1000, 500,... whatever crunchers to give some motivation to our fellow travelers to upgrade and do some ka crunching to get there? Back in December when I got hit with the 1 hour nasties, it was "who ya gonna call". I figure if they can share our dialogs they just might learn something.
Regards-tweakster
Bruce; Thank you for the
)
Bruce; Thank you for the regards and concern about my commitment to my utility provider. I suffered with S__i for some three years. I have a great collection of modems. My motivation with Einstein is to find out how good you guys can get. The 24 divided by 1 equals 16 turns out to be just a bump in the road thanks to your attention. I am willing to turn up the heat if your record of "full service server support" keeps my rigs hot (even in July).
Regards-tweakster
RE: If there were a way for
)
Pop ups can be a disaster in DC programming, if they destroy the ability to leave the progam running unattended. Any pop-up, even for an error condition, should be one that does not require a user response to clear it.
This is to make it possible to sort out the problem via a remotely connected manager (which could be the official boinc_manager running remotely, or could be an independent one like BoincView)
Secondly, some clients run on boxes with no GUI support at all - a pop up box based on the assumption that the GUI is there can then cause more problems and mislead the user into thinking that the software needs a GUI for normal operation.
The Predictor app has a pop-up in it that causes both these issues to arise - we do not need any more of these.
My preference would be to allow the quota to drop to zero, email the user whenever it drops from one to zero, and provide a link in that email and on their account page that they can use to bump the quota to one when they think they have fixed the fault. This would allow many attempts to fix the fault to be made on the same day, but with a user reset between each failed attempt an unattended rogue box would not eat too many WU
This change would have to come from the BOINC team rather than the Einstein folks.
River~~
~~gravywavy
hehe, I have screen saver set
)
hehe, I have screen saver set to none. I don't need a screen saver :D I don't even use the power saving, send monitor to standby... When I'll be away from the comp for awhile, I just turn the monitor off. As to having it go in standby, 2 mins or so, nah... That can be a tad annoying if one's reading something, and well... Or watching a TV show for that matter, and em TV shows can last for awhile. Watching a DVD could present the same issue if it's wanting input, especially if it's something like Lord of the Rings: extended version :D
I've seen that popup in Predictor before... To make a long story short, it was when the science app saw an update, that dealt with where certain parameters were contained. This was, as I remember because some dialup users didn't like downloading the same file with each WU, so it was hard coded into the app, and then referenced by the WU sometime back.
However, when the first WUs came out using that (which promptly needed to be upgraded again), people started getting this particular Fortran error, and practically everyone was getting the same one. Was a coding bug, related to the upgraded app, when first implementing this change...
Perfectly understandable from a programming standpoint (as moving the data from the WUs, to being hard coded into the app itself isn't a small change); though also a tad bad, when I came home from a full day at school or something, saw the error on my monitor, and had lost possibly up to 10 hours of crunch time or so, as BOINC wouldn't schedule anything else to run until a user responce was given to that dialog... Course, beyond lost crunch time, when nothing could schedule (prior to the sorting of the fortran errors out), was wonder about how this could effect our daily quota as the real problem was bad WUs that were effecting everyone, and not a hardware issue with these people's comps...
RE: I've seen that popup in
)
Yes, well, based on my data which I have just looked at again, the version of the application that had those problems was 4.28 ... guess which version is still in use?
If you guessed 4.28 you get the lead balloon.
But, focus stealing is bad, pop-ups should almost never be used. And I am not convinced that there is a need for them with BOINC. If the point is to have it run as an unobtrusive application then this is not a good idea.
If the operator configures themselves into a corner I have little sympathy. But, we should not incraese the risk that we lock up someone's machine. I have even seen cases where the use of a pop-up causes a problem where the person cannot acknowledge the pop-up because something else stole the focus but you can't get the focus back because the pop-up needs to be acknowledged first ...