Message boards : News : Acemd apps should be fixed
Author | Message |
---|---|
We are aware of the problem and working on it. Sorry for the inconvenience. | |
ID: 52451 | Rating: 0 | rate:
![]() ![]() ![]() | |
What? | |
ID: 52453 | Rating: 0 | rate:
![]() ![]() ![]() | |
Hi, | |
ID: 52454 | Rating: 0 | rate:
![]() ![]() ![]() | |
No Problem, the other Projekt are now very happy | |
ID: 52455 | Rating: 0 | rate:
![]() ![]() ![]() | |
We are aware of the problem and working on it. Sorry for the inconvenience. You're welcome. Thank you very much for the information, and for your efforts. May the Schwartz be with you (Spaceballs film quote ;-) | |
ID: 52456 | Rating: 0 | rate:
![]() ![]() ![]() | |
Hello Toi | |
ID: 52457 | Rating: 0 | rate:
![]() ![]() ![]() | |
Go to ebay to get a 'scrap' key, they are very affordable... | |
ID: 52459 | Rating: 0 | rate:
![]() ![]() ![]() | |
Go to ebay to get a 'scrap' key, they are very affordable... Ah ah ah !!! Good Idea, but such 'scrap' keys are not on ebay. It is not our key who have problem. It is server licence of project it self ! An official, worldwide, serious research PRJ who already weeks know it will happens, do as good as nothing ! And now, because th amount of WU by host, all is blocked ! The PRJ blocked it self !!! Normally, I get 4 WU at once. And are returned within 2 days. Now , host (event log) says computer has finish a dayly quota of 1 task. Of course. Each task in error decrease amount of downable task ! Next step will be "....quota of zero" ??? Or perhaps PRJ want to remove Windows hosts ??? Only Admin can anser, but nothing !!! Only " we know and we appologize". Nice. When a solution ??? Or perhaps they not need us anymore ? Answer is only known by admin... | |
ID: 52460 | Rating: 0 | rate:
![]() ![]() ![]() | |
So is this what is causing the errors on the tasks in windows? | |
ID: 52464 | Rating: 0 | rate:
![]() ![]() ![]() | |
Hello Toi Another thread indicates that WUs still run on Windows XP hosts, but not on those with Linux or more recent versions of Windows. The license for the XP version is expected to last a little while longer before it also expires. Someone on the project team wrote that a fix is expected to come only when the ACEMD 3 application is ready, which will not need these licenses. If so that version will be hard to test against the current version that need licenses. Users might make sure that they are signed up to run ACEMD 3 when it is available, and therefore help beta test this new application. | |
ID: 52465 | Rating: 0 | rate:
![]() ![]() ![]() | |
... The license for the XP version is expected to last a little while longer before it also expires. Would be great if someone from the team could tell us when the XP license will expire. Not that we could do much against it, but at least we would know. | |
ID: 52468 | Rating: 0 | rate:
![]() ![]() ![]() | |
The licenses have obviously expired for the applications sent to users (except for the XP version), but have they also expired for the server programs used to generate those applications? | |
ID: 52469 | Rating: 0 | rate:
![]() ![]() ![]() | |
The licenses have obviously expired for the applications sent to users (except for the XP version), but have they also expired for the server programs used to generate those applications? Devs had that opportunity months ago with the Linux version... | |
ID: 52471 | Rating: 0 | rate:
![]() ![]() ![]() | |
The licenses have obviously expired for the applications sent to users (except for the XP version), but have they also expired for the server programs used to generate those applications? They didn't tell us the results, though. Also, they could create a few test workunits with the current application and with more output enabled as the application starts to help pin down which item the application is built from contains the expired license, if they don't already know. Also, I haven't seen it mentioned whether a computer running Windows Vista would try to run the XP version or the newer Windows version. If no one knows, I have a computer that is probably suitable for testing this, but I cannot run that computer at the same time as the computer I'd otherwise be using. | |
ID: 52476 | Rating: 0 | rate:
![]() ![]() ![]() | |
The licenses have obviously expired for the applications sent to users (except for the XP version), but have they also expired for the server programs used to generate those applications? This isn't the 1st time this license has expired. I think last year it did as well. They know what which license expired. | |
ID: 52478 | Rating: 0 | rate:
![]() ![]() ![]() | |
Hello mmonnin, I had also the same impression. But because I manage all PRJ of all users of my team, I was not more sure. And still not any reaction of admin. The worst, is that they still send WU who will always crash... With results that all produced WU will comes in full error due to the max# of error/total/success who is set by projet on 7 , 10 , 6 | |
ID: 52479 | Rating: 0 | rate:
![]() ![]() ![]() | |
The worst, is that they still send WU who will always crash... if you happen to have a Windows XP somewhere around, use this. The XP app still works (who knows for how long). | |
ID: 52480 | Rating: 0 | rate:
![]() ![]() ![]() | |
Let me know if solved. For CUDA 80 only. | |
ID: 52481 | Rating: 0 | rate:
![]() ![]() ![]() | |
Let me know if solved. For CUDA 80 only. Solved for both windows and linux? | |
ID: 52484 | Rating: 0 | rate:
![]() ![]() ![]() | |
Let me know if solved. For CUDA 80 only. I just tried Linux and windows, Linux couldn't find any tasks and windows found a task but immediately errored. | |
ID: 52485 | Rating: 0 | rate:
![]() ![]() ![]() | |
Let me know if solved. For CUDA 80 only. Hello, no it is not solved ! It ill be solved when PRJ will make a reneal of license and fully restat whole project and all server ! But first, they need to pay licence ! | |
ID: 52487 | Rating: 0 | rate:
![]() ![]() ![]() | |
Only Windows for the Short/Long apps: | |
ID: 52488 | Rating: 0 | rate:
![]() ![]() ![]() | |
I updated the short and long apps (version 923). Should fix the problem. As far as I know Linux was not affected. | |
ID: 52489 | Rating: 0 | rate:
![]() ![]() ![]() | |
I updated the short and long apps (version 923). Should fix the problem. As far as I know Linux was not affected. If the new versions work properly, where are the workunits set up to use those new versions? | |
ID: 52490 | Rating: 0 | rate:
![]() ![]() ![]() | |
I updated the short and long apps (version 923). Should fix the problem. As far as I know Linux was not affected. Task 21304404 on normally-reliable host 45218 (Windows 7, GTX 970) has started and is running - seemingly normally. Result likely midday tomorrow. | |
ID: 52491 | Rating: 0 | rate:
![]() ![]() ![]() | |
Mine is running Long run Cuda 80's today on Win 10 pro Yeah.. Glad they got the Lic issue for Win10 fixed or whatever they did. | |
ID: 52493 | Rating: 0 | rate:
![]() ![]() ![]() | |
I now have an application running. A rather bad initial runtime estimate, but that is typical for new versions. Likely to finish in the next 5 hours. Running under Windows 10 Pro. | |
ID: 52496 | Rating: 0 | rate:
![]() ![]() ![]() | |
Anyone try a 20xx card yet? It would be nice to know if the Turing is working now too. | |
ID: 52497 | Rating: 0 | rate:
![]() ![]() ![]() | |
Anyone try a 20xx card yet? It would be nice to know if the Turing is working now too. It is still the ACEMD2 app, so Turing not working yet. | |
ID: 52498 | Rating: 0 | rate:
![]() ![]() ![]() | |
I updated the short and long apps (version 923). Should fix the problem. As far as I know Linux was not affected. What? Linux version has been dead since May! 3 full months. A post of your own. https://www.gpugrid.net/forum_thread.php?id=4925&nowrap=true#51818 | |
ID: 52499 | Rating: 0 | rate:
![]() ![]() ![]() | |
I updated the short and long apps (version 923). Should fix the problem. As far as I know Linux was not affected. mmonnin, The update is only concerning the problem about windows/nvidia licence. It not speak about Linux. Please read all. If you run under other than Windows, it not apply for . | |
ID: 52502 | Rating: 0 | rate:
![]() ![]() ![]() | |
I updated the short and long apps (version 923). Should fix the problem. As far as I know Linux was not affected. You are incorrect. The same thing happened last year and in May for the Linux version. The same problem is not limited to Windows. Toni didn't acknowledge it here but did in May. That's what I was pointing out. | |
ID: 52505 | Rating: 0 | rate:
![]() ![]() ![]() | |
Maybe I'm saying something wrong but my antivirus says that you' re site have been infected by IDP.Generic. | |
ID: 52512 | Rating: 0 | rate:
![]() ![]() ![]() | |
Maybe I'm saying something wrong but my antivirus says that you' re site have been infected by IDP.Generic. Exclude the Boinc directories from your a/v as it's a false positive, any real virus will try to infect other files on your computer and be flagged. Boinc doesn't really have viruses, the units come from Servers and that would be bad for business so they are all well protected. What Boinc does have is alot of back and forth communication between the Servers and our pc's making it LOOK like activity a virus might do so the 'heuristics' in all a/v programs now sometimes flag it. | |
ID: 52519 | Rating: 0 | rate:
![]() ![]() ![]() | |
... What Boinc does have is alot of back and forth communication between the Servers and our pc's ... you need to differentiate between the various BOINC projects. While, for example, some sub-projects of LHC (like CMS) does indeed send back and forth a lot of data all the time, this is not true for GPUGRID. Once a task has been downloaded and started crunching, there is no such additional data transmission until the task is finished and being uploaded. So GPUGRID crunching should not induce a false A/V alarm (at least not on basis of an unusual high data traffic). | |
ID: 52520 | Rating: 0 | rate:
![]() ![]() ![]() | |
So I see the Windows folks got another year of license on their acemd2 app. | |
ID: 52525 | Rating: 0 | rate:
![]() ![]() ![]() | |
I 100% agree! | |
ID: 52526 | Rating: 0 | rate:
![]() ![]() ![]() | |
So I see the Windows folks got another year of license on their acemd2 app. Have none of the project team tried rebuilding the current Linux application with no change except a new version number? If one has, what went wrong? | |
ID: 52527 | Rating: 0 | rate:
![]() ![]() ![]() | |
So I see the Windows folks got another year of license on their acemd2 app. Why do they need to rebuild? The latest 2.03(CUDA80) and 2.04(CUDA100) acemd3 apps worked fine on all the test WU's. Just release the bloody work units! | |
ID: 52531 | Rating: 0 | rate:
![]() ![]() ![]() | |
So I see the Windows folks got another year of license on their acemd2 app. WHY aren't they offering Linux workunits? I'm assuming that it is because the latest Linux applications no longer work, but you appear to be assuming that it is some other reason. | |
ID: 52532 | Rating: 0 | rate:
![]() ![]() ![]() | |
Did you not run the acemd3 test WU's back in July when the new acemd3 apps were released as beta? | |
ID: 52533 | Rating: 0 | rate:
![]() ![]() ![]() | |
Did you not run the acemd3 test WU's back in July when the new acemd3 apps were released as beta? I have all the Windows 10 applications enabled, but I have no record of whether I got any acemd3 tasks or not. You seem to be assuming that the server application for creating acemd2 tasks can create acemd3 tasks also with no programmer to update that server application first. We may need to wait for more of the project team to return from their August vacations to find out if that is correct or not. | |
ID: 52535 | Rating: 0 | rate:
![]() ![]() ![]() | |
Acemd3 uses OpenMM does it not? It is probably version 7.2 by now. That is the same one that Folding@Home is testing now for Core_22. And they are having problems that need fixing. | |
ID: 52536 | Rating: 0 | rate:
![]() ![]() ![]() | |
Did you not run the acemd3 test WU's back in July when the new acemd3 apps were released as beta? The license use with acemd2 just needs renewed on Linux. It expired before Windows but has yet to be renewed. Year after year this happens. | |
ID: 52537 | Rating: 0 | rate:
![]() ![]() ![]() | |
But Toni indicated they wanted to get away from this yearly license renewal garbage they have to deal with and just release the apps as BOINC wrapper apps that take care of themselves. | |
ID: 52538 | Rating: 0 | rate:
![]() ![]() ![]() | |
But Toni indicated they wanted to get away from this yearly license renewal garbage they have to deal with and just release the apps as BOINC wrapper apps that take care of themselves. Correct. Delay is just a lack of time due to other projects (lots of quality time must be allotted for addressing scheduler mysteries and follow-up problems). | |
ID: 52539 | Rating: 0 | rate:
![]() ![]() ![]() | |
Keith, do not get impatient. We knew, that the lack of WUs would be for several months... | |
ID: 52540 | Rating: 0 | rate:
![]() ![]() ![]() | |
Well my frustration wouldn't be so high if there were no work units for anybody once all the licenses ran out. But since Windows hosts get work now because they got a new license but we did not, I come to the conclusion that Linux hosts are considered second class citizens not deemed worthy of attention. | |
ID: 52541 | Rating: 0 | rate:
![]() ![]() ![]() | |
Acemd3 uses OpenMM does it not? It is probably version 7.2 by now. That is the same one that Folding@Home is testing now for Core_22. And they are having problems that need fixing. I've forgotten which BOINC recently found problems with the 436 driver. However, you might ask Folder@Home to check whether their problems are also related to the 436 driver. I got a GPUGRID task today, so the Windows portion of the problem might now be workunits not being created as fast as the users become ready for them. | |
ID: 52542 | Rating: 0 | rate:
![]() ![]() ![]() | |
(lots of quality time must be allotted for addressing scheduler mysteries and follow-up problems).This sounds to me like Toni has: miles to go before we're sated, miles to go before we're sated. ____________ ![]() | |
ID: 52545 | Rating: 0 | rate:
![]() ![]() ![]() | |
This sounds to me like Toni has: It just means that it is August in Europe. | |
ID: 52546 | Rating: 0 | rate:
![]() ![]() ![]() | |
Message boards : News : Acemd apps should be fixed